# siteb.biz — full text > The complete text of every guide, in the current language, so that it can be read in one request rather than page by page. ## Calculating Exact Age: Free Online Tools and the Leap-Year Problem https://siteb.biz/guides/how-to-calculate-exact-age Reviewed 2026-08-09 · Dates and time - Calendar subtraction and the 365.25 shortcut can produce different ages for the same person. - The borrowing method is the standard used by most institutions, but leap-day birthdays vary by jurisdiction. - Schools, sports leagues and pension schemes usually use a fixed cut-off date rather than a precise age. - Free online tools are fine for quick checks if they run entirely in your browser and respect your privacy. - For legal, financial or medical decisions, confirm any calculation with a qualified professional. Most people think age is simple subtraction: today's date minus your birth date. But subtract 15 March 2019 from 14 March 2024 and you get four years, eleven months and twenty-eight days, not five years. The mismatch matters the moment you are one day short of a pension threshold, a visa age limit or a school cut-off. Free online tools can handle this arithmetic instantly, provided they run entirely in your browser and never send dates to a server. Some tools do exactly that, quietly. This guide explains the borrowing method institutions use, the 29 February trap, and when a spreadsheet or a professional is the safer route. ### Why subtracting dates and dividing by 365.25 rarely agree A common shortcut divides days lived by 365.25 to account for leap years. That gives a rough average, but it is not what pension offices, courts or schools use. Calendar age is computed by rolling the calendar forward from the birth date until it either reaches or passes today's date. The result can differ from the 365.25 figure by days or even weeks because months are not uniform and leap years do not occur every four years without exception. - Dividing by 365.25 assumes every year is exactly 365.25 days long. - It ignores the fact that months have 28, 29, 30 or 31 days. - It treats 1900 and 2100 as leap years, which they are not. - It can make someone appear eligible a day early or a day late. Free online tools that show only a decimal age are usually using the shortcut, not the calendar method. ### The borrowing method: a worked example The final answer is 32 years, 8 months and 22 days. The day count is not simply 24 minus 14 because February only had 28 days in 2025. If the period had included a leap year, the borrowing would have used 29 February instead of 28 February, adding one extra day to the result. This is why two people born on the same calendar date can show slightly different day counts depending on how many leap days fall between their birthdays. | Start | 2025-03-08 minus 1992-06-14 | needs borrowing | | Borrow 1 month as days | March → February, 8 days from 28 | 2024-02-28 with 8 days | | Borrow 1 year as months | 2024 → 2023, 12 months + February | 2023-14-28 | | Subtract | 2023-14-28 minus 1992-06-14 | 32 years, 8 months, 22 days | ### The 29 February problem People born on 29 February create the most commonly argued-about edge case. In non-leap years there is no 29 February, so the anniversary does not exist. Different institutions solve this differently: some treat 28 February as the birthday, some treat 1 March, and some wait for the next leap year. The choice changes whether a person is considered one year older on 28 February, 1 March or 29 February of the following year. - Born 29 February 2020; in 2025 some systems flag 28 February, others 1 March. - UK law traditionally treats 28 February as the anniversary in common years. - New Zealand law treats 1 March as the birthday in non-leap years. - Many database systems crash or return null when 29 February is missing. ### How different legal systems treat a 29 February birthday The split matters for contracts, driving licences, criminal-age thresholds and pension claims. If you rely on a free online tool, check whether it follows the rule that applies to your document. A tool hosted in California may default to 28 February even if your solicitor in London or Auckland needs 1 March. The safest approach is to know the jurisdiction's rule first, then find a tool that lets you select it or compute it manually in a spreadsheet. | England and Wales | 28 February is the legal birthday | | Scotland | 28 February is the legal birthday | | New Zealand | 1 March is the legal birthday | | Hong Kong | 1 March is the legal birthday | | United States | Varies by state; often 28 February federally | | Civil-code countries | Code may specify 28 February or 1 March | ### Why schools, sports leagues and visas use a date, not an age Institutions rarely ask 'are you five?' Instead they ask 'were you born on or before 1 September 2019?' A fixed date removes ambiguity about time zones, leap years and whether age is counted in whole years or rounded years. It also makes batch processing possible: every child in a school year is treated the same, regardless of whether they were born on the first or last day of the window. - UK school reception year: children must be four by 1 September. - FIFA youth competitions: cut-off is the calendar year of birth, not the exact birthday. - US Medicare: eligibility begins at the start of the month you turn 65, not your birthday. - UK state pension age: based on date of birth and gender history, not a simple birthday. ### When a free online tool is enough, and when it is not For a quick check of how many days until a birthday, free online tools that run entirely in the browser are fine. Look for a clear privacy statement that says no data leaves your device; otherwise your birth date could be logged on a server you do not control. The risk is low for a birthday, but higher if the date is combined with a name, address or national insurance number. - Use a tool for: party planning, quick curiosity, draft spreadsheets. - Use a spreadsheet for: recurring payroll, membership renewals, audit trails. - Ask a solicitor for: wills, contracts, criminal-age or pension disputes. - Ask a tax adviser for: pension crystallisation or lifetime allowance calculations. No online calculation in this guide is financial, legal or medical advice. ### A five-step sanity check before you trust any answer - Confirm whether the tool uses calendar subtraction or the 365.25 shortcut. - Check the leap-year rule for 29 February birthdays, if relevant. - Verify the output against a second tool or a spreadsheet formula. - Read the privacy policy to see whether dates are sent to a server. - For any legal, financial or medical decision, confirm with a qualified professional. Q: Can I trust a free online age calculator for legal documents? A: Free online tools are useful for rough checks, but legal documents require the rule used by the relevant jurisdiction. A tool hosted in one country may default to a different convention for leap-day birthdays, or may round ages to the nearest whole year rather than counting complete years and months. For leases, contracts, pension claims, visa applications or criminal-age questions, confirm the calculation with a solicitor or the issuing authority. Treat any online result as a draft, not evidence, and keep a record of which rule you applied. Q: Why does my spreadsheet give a different age from the website? A: Spreadsheets store dates as serial numbers and may use a 365.25-day year, whole-year truncation or calendar subtraction depending on the formula. Functions like DATEDIF use calendar subtraction in Excel and Google Sheets, but custom formulas often do not. Check whether the formula accounts for leap years, month lengths and the convention used for 29 February. The safest check is to work through the borrowing method manually for a few test dates and compare the results. If the numbers disagree, the formula is probably using a different method from the website. Q: How much does it cost to get an age calculation checked professionally? A: A solicitor or tax adviser will typically charge €100–€300 per hour depending on location and complexity, with a simple date confirmation taking well under an hour. Many local authorities, pension services and school admissions teams will confirm cut-off dates for free. The cost rises if the question involves cross-border law, missing historical records or disputed identity documents. For most everyday questions, a reputable free online tool plus a phone call to the relevant office is enough; reserve professional fees for decisions with money or legal risk attached. ## Working Days, Business Days and Free Online Tools: A Deadline Guide https://siteb.biz/guides/working-days-explained Reviewed 2026-08-09 · Dates and time - ‘Working day’ has no single global meaning; the weekend and public holidays differ by country and by contract. - Many countries treat Friday and Saturday, or Friday alone, as the weekend, so a Monday-to-Friday assumption can be wrong. - Public holidays, substitute days, and regional observances make automatic counting risky without a verified local calendar. - The safest deadline wording combines a working-day count with a defined jurisdiction, start-day rule, time zone, and a hard calendar date. - Use free online tools for quick checks, but turn to spreadsheets, native apps, or a lawyer when the stakes are high. ‘Five working days’ sounds precise until the day you need it. In some offices it means Monday to Friday; in others it excludes public holidays, half-days, or days when the relevant court is closed. If you are setting a delivery date, a cancellation window or a response deadline, the wrong assumption can turn a clear contract into a dispute. This guide compares how working weeks are defined around the world, explains why public holidays cannot be left to free online tools alone, and shows how to phrase a deadline so it survives being forwarded, questioned, or translated. We also point to the better options—spreadsheets, native apps, or a lawyer—when a browser calculator is not enough. ### There is no universal definition of a working day A working day is any day on which business is normally conducted, but that definition is circular. In London it usually means Monday to Friday excluding bank holidays. In Riyadh it has historically meant Sunday to Thursday, while in Tel Aviv it means Sunday to Thursday with Friday and Saturday off. Even within one country, a ‘business day’ for a bank can differ from a ‘working day’ for a court, a port, or a school. Contracts often assume the reader knows which calendar applies, which is exactly how a five-day deadline becomes a seven-day argument. Before you calculate anything, you have to know whose calendar you are counting. - The country’s weekend pattern - Public and religious holidays - Half-days, bridge days, and company closures - Whether the start day is counted as day one - The specific institution’s opening hours, such as courts, banks, or ports ### Working weeks around the world The weekend is not always Saturday and Sunday. In many countries Friday is the religious day of rest, so the weekend runs from Friday to Saturday or Friday to Sunday, and the working week shifts to Sunday through Thursday. Some governments have moved the weekend to align with global markets, so the same country can change its pattern within a decade. Always check the current rule for the specific jurisdiction, because a calculator that assumes a Monday start will be wrong by a day or more. | United States, UK, Canada, Australia | Saturday–Sunday | Monday–Friday | | Israel | Friday–Saturday | Sunday–Thursday | | Egypt, Saudi Arabia, Kuwait, Qatar | Friday–Saturday | Sunday–Thursday | | United Arab Emirates | Saturday–Sunday (from 2022) | Monday–Friday | | Iran | Friday (Thursday often half-day) | Saturday–Wednesday or Thursday half-day | | Brunei and some Malaysian states | Friday–Saturday or Friday–Sunday | Varies by state | Weekend rules change; check the current local regulation rather than relying on memory. ### Why public holidays cannot be automated safely A date calculator can add five days to a start date, but it cannot know which days your contract treats as non-working. Public holidays are the biggest source of error. They can be fixed, lunar, regional, or observed only by certain institutions. Substitute-day rules differ: in the United Kingdom, a Christmas Day that falls on a Sunday moves to the following Monday, while in other countries it does not. A free online tool that ships with a generic holiday list will miss these details unless you feed it the exact calendar you intend to use. | Fixed-date public holiday | Substitute-day rules vary by country and contract | Christmas Day moving to Monday in the UK | | Lunar or religious holiday | Date changes every Gregorian year | Eid al-Fitr, Chinese New Year, Diwali | | Regional or state holiday | Not observed everywhere in the country | Bavaria’s public holidays, US state holidays | | Half-day or bridge day | Often treated as non-working though not official | Christmas Eve, afternoon before Eid | | Substitute holiday rules | A weekend holiday may or may not create a Monday off | UK yes; some European countries no | Always verify the current holiday calendar for the relevant jurisdiction before relying on any automatic count. ### How to phrase a deadline so it cannot be read two ways The safest wording converts a working-day count into a specific calendar moment and defines every term. Do not rely on the other party’s assumptions. State the time zone, the cut-off time, and the holiday list. If the contract says only ‘five working days’, one side may count the start day and the other may not. The examples below are drafting habits, not legal advice; for high-value or regulated contracts, have a lawyer review the clause. - Name the jurisdiction whose holidays apply, for example ‘England and Wales’ or ‘Dubai, UAE’. - Define ‘working day’ explicitly: ‘any day other than Saturday, Sunday and the public holidays observed in [jurisdiction]’. - State whether the start day counts as day one or day zero. - Say what happens if the final day lands on a holiday: does it roll to the next working day? - Add a calendar date and time fallback in brackets, such as ‘no later than 17:00 CET on 15 May 2026’. ### Tools that can do the counting For a quick, single-jurisdiction check, free online tools are convenient. At siteb.biz we list independent mini projects on their own subdomains and publish our own browser-based date calculators; none of them uploads your dates to a server, and none requires a login. That matters because many generic date calculators quietly send your input to analytics or cloud services. For recurring calculations, a spreadsheet with a custom holiday table is usually more reliable. For cross-border or high-stakes deadlines, a native legal-calendar app or a professional review is the safer route. | Browser date calculator | Quick, single-jurisdiction checks | May use generic holidays or upload data | | Spreadsheet with custom holiday list | Recurring deadlines in one organisation | You must maintain and audit the holiday table | | Native calendar app | Personal planning and reminders | Rarely supports multiple jurisdictions | | Legal or professional calendar service | High-stakes or cross-border contracts | Costs €150–€400 per hour or €30–€100 per month | | Lawyer or contract professional | Drafting and dispute prevention | Fee reflects jurisdiction and complexity | ### When a free online tool is the wrong choice Browser calculators are fine for informal planning, but they are the wrong place to stop when the consequence of a missed day is a penalty, a lost right, or a regulatory breach. Free online tools do not read your contract, do not know your local court rules, and cannot interpret substitute-holiday conventions. If the amount at stake is more than a few hundred euros, the cost of a professional check is usually small compared with the cost of being wrong. A lawyer will typically charge €150–€400 per hour, with the lower end for a straightforward contract review and the higher end for cross-border or regulated work. - Cross-border contracts where two countries have different weekends and holiday calendars. - Employment-law deadlines, such as notice periods or tribunal filing limits. - Financial settlement dates that depend on banking holidays. - Medical, regulatory, or tax response windows. - Any situation where a missed day triggers a penalty or loss of rights. ### A practical checklist before you sign or send Before you commit to a working-day deadline, run through the same checks a sceptical colleague would use. Confirm the jurisdiction, list the holidays, and write the calendar date in the message or contract. If you are using a free online tool, treat its answer as a first draft, not the final word. The strongest deadlines are the ones that still make sense when someone forwards the email six months later. - Confirm the jurisdiction whose holidays apply. - Convert the working-day count into a calendar date and time zone. - Check whether the start date counts as day one. - Verify substitute-holiday rules for that year. - Add the calendar date in brackets after the working-day phrase. Q: Do weekends count as working days? A: Usually not, but it depends on the contract and the country. In the United States, the United Kingdom, and similar jurisdictions, a working day is normally Monday to Friday, excluding public holidays. In Israel and many Middle Eastern countries, the weekend falls on Friday and Saturday, so Sunday to Thursday are treated as working days. Some countries, such as the United Arab Emirates, have shifted to a Saturday–Sunday weekend. Your contract can also define the term differently, so always read the clause before relying on a calculator. Free online tools are useful for a quick check, but they cannot override the wording you have agreed. Q: Can I trust a free online date calculator for a contract deadline? A: For informal planning, yes; for a binding deadline, no. A browser calculator can add days, but it does not know which holidays your jurisdiction observes, whether your contract counts the start day as day one, or what happens when the final day lands on a public holiday. Many tools also send your input to remote servers, which is a concern if the dates are sensitive. At siteb.biz we list tools that run entirely in the browser and require no login, but even those should be treated as a first draft. When money or a legal right is at stake, verify the date manually or ask a professional. Q: What is the safest way to write a deadline in a contract? A: Give both a working-day count and a hard calendar date, and define every term. For example: ‘Within five working days, meaning any day other than Saturday, Sunday and public holidays in England and Wales, starting the day after receipt and ending no later than 17:00 GMT on 15 May 2026.’ This removes ambiguity about the start day, the holiday list, the time zone, and substitute holidays. It also gives the reader a calendar date to diary. This wording is a drafting habit, not legal advice; for high-value or cross-border contracts, have a lawyer adapt it to your jurisdiction. ## Days Between Two Dates: Why Free Online Tools Can Get It Wrong https://siteb.biz/guides/days-between-two-dates Reviewed 2026-08-09 · Dates and time - The difference between inclusive and exclusive counting can shift a deadline by exactly one day, even when the total is the same. - Leap years follow a divisible-by-4 rule that is cancelled by centuries and restored by every fourth century. - Finance uses day-count conventions such as actual/365 and 30/360 that treat months and years differently from ordinary calendars. - Browser-only tools protect privacy, yet a spreadsheet or a professional may be the safer choice for binding contracts or complex accruals. - Always document the convention you used, because two reputable calculators can disagree and both be correct under their own rules. Counting the days between two dates is the kind of calculation most people assume is too simple to get wrong. Yet a landlord’s notice period, a software trial expiry, a statutory cooling-off window or an interest accrual can all shift by a day depending on whether the start and end dates are included in the total. That single day can mean the difference between a valid cancellation and a late fee. Free online tools can handle the arithmetic instantly, but only if you understand what they are actually counting. Some count every calendar day; others count whole 24-hour intervals; finance uses conventions that treat months as 30-day blocks. This guide explains inclusive and exclusive counting, leap-year traps and the day-count conventions that spreadsheets, browser calculators and contracts use differently. ### Inclusive vs Exclusive: The One-Day Gap Imagine you give 30 days’ notice on 1 June. If the notice period is counted exclusively, the clock starts on 2 June and the 30th day is 1 July. If it is counted inclusively, 1 June itself is day one and the 30th day is 30 June. Both totals are exactly 30, yet the last valid day differs by 24 hours. That is how a contract, a tenancy or a refund window can be honoured on one reading and missed on another. Always check whether the document says “within 30 days”, “30 days from” or “not less than 30 days before”, because each phrase points to a different convention. | Exclusive: start not counted | 1 July | 30 | | Inclusive: start counted | 30 June | 30 | ### Leap Years and the Century Exception A leap year adds 29 February, but the rule is more specific than “every four years”. The extra day is skipped in years divisible by 100, unless they are also divisible by 400. That means 1900 was not a leap year, while 2000 was. For date spans, the only thing that matters is whether the interval crosses 29 February. A one-year contract from 1 March 2023 to 29 February 2024 contains 366 days; the same contract from 1 March 2024 to 28 February 2025 contains 365 days, even though both are roughly a year. Most modern free online tools handle this correctly, but older spreadsheets and hand-coded calculators sometimes store 1900 as a leap year to remain compatible with Lotus 1-2-3, which can silently shift 1900-era dates by a day. - A year is a leap year if it is divisible by 4. - It is not a leap year if it is divisible by 100. - It becomes a leap year again if it is divisible by 400. - 1900 had 28 days in February; 2000 had 29. - A leap day only affects spans that actually cross 29 February. - Century years are the most common source of calculator disagreement. ### Day-Count Conventions: 30/360 and Actual/365 Finance does not count days the way a wall calendar does. In the actual/365 convention, every genuine calendar day in the period is counted, but the year is treated as 365 days even in a leap year. In the 30/360 convention, every month is treated as 30 days and every year as 360 days, so a two-month loan from 31 January to 31 March is counted as exactly 60 days, while actual/365 would count 60 days in a non-leap year and 61 days in a leap year. The result is that two contracts with the same nominal rate can produce different interest amounts. A €10,000 loan at 5% for 90 days costs about €12 more under actual/365 than under 30/360 in an ordinary year. You will usually see actual/365 in sterling and euro corporate loans, while 30/360 appears in US corporate bonds and many mortgage-backed securities. | Actual/actual | Real calendar days; real year length | Government bonds, some derivatives | | Actual/365 | Real calendar days; year fixed at 365 days | Sterling and euro corporate loans, some credit cards | | 30/360 | Each month = 30 days; each year = 360 days | US corporate bonds, many mortgage-backed securities | | Actual/360 | Real calendar days; year fixed at 360 days | Some short-term bank loans | None of these examples is financial, legal or tax advice. ### Can You Trust Free Online Tools? Most browser-based date calculators are accurate for simple spans, but accuracy is not the only risk. Some sites send your dates to a server to generate a shareable link or to run analytics. For anything involving a contract, a patient, a court filing or a tax deadline, you should prefer a tool that runs entirely in the browser and never uploads your input. The same caution applies to spreadsheets stored in cloud accounts with sync enabled. The tools this directory lists include browser-only options, but a directory review is not a guarantee of privacy or accuracy. - Runs entirely in the browser: no network request carries your dates. - Open source or auditable: you can inspect the calculation logic. - No login or cloud save: sensitive deadlines stay off remote servers. - Clear inclusive/exclusive toggle: prevents the one-day shift. - Handles 1900 and century years correctly: avoids the Lotus 1-2-3 bug. - Shows its working: lists counted days or states the convention used. ### When a Browser Calculator Is the Wrong Tool For one-off checks, a browser calculator is fast and private. For repeated business calculations, a spreadsheet gives you an audit trail and lets you adjust conventions. For legally binding deadlines, a solicitor or tax adviser is the safer route. The cost of professional advice varies widely: a solicitor reviewing a notice clause will typically charge €150–€400, depending on jurisdiction and complexity, while a chartered accountant verifying bond accruals may charge €200–€500. A native desktop date calculator or a Python/R script may suit developers who need batch processing. Cloud spreadsheets are convenient, but remember that co-workers or algorithms may see dates you would rather keep local. | Browser calculator | One-off, private checks | Ambiguous inclusive/exclusive settings | | Spreadsheet | Repeated, auditable calculations | 1900 leap-year bug; cloud sync | | Native app or script | Offline, sensitive or batch work | Update lag; maintenance burden | | Professional adviser | Binding contracts or disputes | Cost €150–€500 depending on jurisdiction | ### Five Quick Checks Before You Rely on a Number Before you paste a day count into an email, a contract or a spreadsheet formula, run through these sanity checks. They take less than a minute, yet they catch the one-day slips that turn a valid notice into a missed deadline and the convention mismatches that make two interest figures diverge. Treat the first result from any calculator as a hypothesis, not a fact. If two calculators disagree, the cause is almost always a setting, not a bug. - State the convention: inclusive or exclusive, and which date is day zero. - Count the leap day: does the span cross 29 February in a leap year? - Match the domain: legal notices, finance accruals and project schedules use different rules. - Reconcile tools: run the same dates through two calculators and investigate any difference. - Document the source: save a screenshot or formula so you can explain the number later. If two reputable tools disagree, the difference is usually a convention, not a bug. Q: Should I include the start date when I count days between two dates? A: It depends on the convention you are expected to use. In many legal contexts, a period is counted exclusively: the day the notice is served is day zero, and the period ends at midnight on the 30th day after. In project management and some contract schedules, counting is inclusive, so the start date is day one. The safest approach is to look for wording such as “within 30 days of” or “not less than 30 days before”, which usually signals exclusive counting. If you use a free online tool, choose one that lets you toggle inclusive or exclusive mode and shows the counted days. When the answer matters, write down which convention you used. Q: Why does 1900 matter if I am calculating dates in 2024? A: For dates after 1 March 1900, the 1900 leap-year bug will not change your result, but it is a warning sign. Microsoft Excel and Google Sheets originally copied a Lotus 1-2-3 quirk that treated 1900 as a leap year, so any date serial number before 1 March 1900 is off by one day in those spreadsheets. The same code path is why the century rule matters: 2100 will not be a leap year, and a calculator that simply divides by four will get long-dated bonds and leases wrong. If your calculation crosses a century year, confirm whether the tool applies the divisible-by-400 exception. Q: Why does my loan interest not match the calendar days? A: Lenders do not always use calendar days. The actual/365 convention counts every real day but divides by 365, while the 30/360 convention treats each month as 30 days and each year as 360 days. A 90-day loan in February, March and April will produce a different interest amount under each method. Your loan agreement should name the convention; if it does not, ask before you try to check the figure yourself. A browser calculator can verify the arithmetic once you know the convention, but it cannot tell you which convention applies. This is not financial advice. ## Digital Business Cards: vCard, QR Codes and Free Online Tools Compared https://siteb.biz/guides/vcard-and-contact-sharing Reviewed 2026-08-08 · QR codes and sharing - A vCard QR puts the contact file straight into the phone; a link QR merely opens a web page the scanner may ignore. - Keep the vCard minimal—name, one phone number and one email—to maximise import success across iOS and Android. - Photos, long notes and social links often drop or break when imported on different devices. - Static vCard QRs protect privacy; dynamic link QRs can log who scanned you and where. - Free online tools are fine for simple cards, but professional design, a CRM or a native app is better for teams, frequent updates or regulated fields. Business cards persist because handing someone a physical object still works. The digital equivalent is a vCard: a small text file that tells a phone where to store your name, number, email, address and more. Plenty of free online tools will turn that file into a QR code, but not all of them produce something every phone can read, and some quietly add tracking links or non-standard fields that break the import. This guide compares vCard QR codes, link-to-page QR codes and app-based digital cards. It explains which vCard fields import cleanly into iOS, Android and desktop address books, which ones silently drop, and why a QR that opens a landing page is usually less useful than one that adds a contact directly. We also cover what a scanned card reveals about you, and when a free online tool is the wrong place to start. ### What a vCard really is A vCard is just a plain-text file with a .vcf extension and a rigid syntax: each line starts with a property name, a colon, and a value. The current standard, vCard 4.0, can hold names, phone numbers, email addresses, URLs, postal addresses, photos, job titles, notes and even social-media links. Older phones still speak vCard 3.0, and a few budget devices only understand the much simpler 2.1 format. If the generator you use writes vCard 4.0-only syntax—such as certain international phone formatting or embedded base64 photos—the file may look fine on your iPhone but fail to import on an older Android handset. That is why the simplest vCard is usually the most reliable one. ### Which vCard fields phones actually import Not every field you add to a vCard will survive the trip into someone else's address book. The safest approach is to treat the vCard as a business card, not a CV: name, phone, email and perhaps one URL. Add photos, multiple addresses, long notes or social links and you enter a lottery. iOS, Android, Samsung Contacts, Google Contacts and Outlook each parse the file slightly differently, so a field that imports on one device may vanish on another. The table below shows the fields we have seen import cleanly in tests, and the ones that frequently disappear or arrive broken. | Full name (FN/N) | Yes | None significant | | Phone (TEL) | Yes | Formatting can split work/mobile labels | | Email (EMAIL) | Yes | Label may become 'Other' | | URL | Mostly | May be treated as a note on old phones | | Postal address (ADR) | Sometimes | Line order and country code often lost | | Photo (PHOTO) | Rarely | Large images break; iOS and Android differ | | Organisation/title (ORG/TITLE) | Sometimes | Title often dropped; org name may be appended to contact name | | Note (NOTE) | No | Truncated or ignored on many devices | | Social profiles (X-SOCIALPROFILE) | No | Proprietary; often unread | If a field is not in the table above, assume it will not survive a cross-platform import. ### vCard QR, link QR or app-based card: the trade-offs There are three common ways to share a digital business card. A vCard QR encodes the contact file itself; scan it and the phone offers to save the person. A link QR sends the scanner to a web page, where they may read your details, download a vCard or do nothing. An app-based card lives inside a proprietary service and usually requires the other person to install the same app or view an ad-supported profile. Each has different costs, reliability and privacy implications. | vCard QR | A .vcf file the phone can import directly | Free to generate with free online tools; dense codes can be hard to scan; printed cards cost €0.05–€0.40 each depending on paper and quantity | | Link-to-page QR | A mobile web page with your details | Hosting €0–€12/month plus domain €10–€50/year; page can go offline, track visitors, or demand a login | | App-based card | A profile inside a specific app or platform | Freemium €0–€15/month, team plans €8–€40/user/month; platform can change terms, show ads, or shut down | | Printed NFC card | A tap-to-save vCard or link | €3–€25 per card depending on material and volume; not every phone reads NFC; still needs hosting | A link QR is only as good as the hosting behind it; if the bill is not paid, your card becomes a dead link. ### Why a QR-to-landing-page often ends up in the bin A link QR sounds flexible because you can change the destination later, but flexibility is only useful if the scanner reaches the destination. In practice the user must unlock their phone, open the camera, scan the code, wait for the browser, load the page, read the page, then decide to save the contact. Each step loses people. The page may also require JavaScript, ask for location permission, or serve a cookie banner before showing your phone number. By the time the user sees your details, the moment has passed. A vCard QR shortens that chain to one action: save the contact. - The scanner must have a data signal at that exact moment. - The page has to load faster than the person's patience. - The visitor must complete an extra tap to download or save. - The landing page can break, move or expire without warning. - The owner often gains analytics the visitor did not agree to. ### What you leak when someone scans your card A vCard is not anonymous. At minimum it contains the fields you chose to include, but the QR code itself can reveal more. Static vCard QRs are usually safe because the data lives only in the image; no server sees when or where it is scanned. Dynamic or link QRs route the scan through a shortener or analytics platform, which can log the device type, approximate location, time and referrer. Some free online tools generate dynamic codes by default because they want usage data. If the QR points to a profile page, that page may set cookies, request permissions or ask the visitor to create an account before showing your details. - Your name, number, email and any address you included. - The fact that the scan happened, plus device and location if the code is dynamic. - The hosting platform's tracking cookies and third-party scripts. - Whether the scanner saved the contact or bounced, if the link is tracked. If you would not print it on a paper card, do not encode it into a vCard or a link QR. ### When a spreadsheet, native app or professional is the better choice Free online tools are fine for one-off personal cards or small teams, but they are not always the right tool. If you need hundreds of cards with consistent branding, a print shop or a designer will produce a better result and can pre-validate the vCard across devices; expect to pay €50–€250 for design, plus €20–€80 for a print run. If you update contact details frequently, a native contacts app or a CRM with built-in sharing may be easier than regenerating QR codes. If the card carries legal, medical or financial credentials, a generic generator is the wrong choice: you need a professionally controlled identity system, not a browser tool. This guide does not replace legal, financial or identity advice; it only explains how the technology works. ### Making a vCard QR code that phones can read Start by creating the smallest vCard that still does the job. Use version 3.0 for maximum compatibility, include only one phone number and one email unless you genuinely need more, and avoid embedding a photo unless you have tested it on both iOS and Android. Generate the QR code at a medium error-correction level; very high levels make the pattern denser and harder to scan from a distance. Print it at least 2.5 cm square, and always test the final printed version with a three-year-old phone before you order five hundred. - Write the vCard in a text editor or export it from your contacts app. - Strip out every field except the essentials. - Validate the file by importing it into iOS, Android and one desktop mail client. - Generate a static QR code from the file text, not from a hosted URL. - Print a test card and scan it under poor light with an older phone. - Keep the source file so you can regenerate the code if your details change. ### Test, update and retire your card Digital business cards age quickly. A promotion, a new phone number or a rebrand can make last year's QR code misleading. Because a static vCard QR contains the data directly, you cannot change it after printing; you must reprint. Link QRs can be redirected, but only if you control the domain and the shortener. Set a reminder to review your card every six months, and retire old cards by removing the landing page or marking the number as outdated. A working card that shows last year's job title is worse than no card at all. Q: Can I put my company logo and photo in a vCard QR code? A: You can, but you probably should not. A photo must be base64-encoded inside the vCard or served from a URL, and both approaches break on different phones. iOS tends to accept small embedded images, while many Android contacts apps ignore them or import them at the wrong size. Logos are safer on the printed card itself, not inside the vCard. If you must include an image, keep it under 15 KB, test it on an iPhone and an Android device, and have a fallback version without the photo. Remember that the more data you stuff into the QR pattern, the denser it becomes and the harder it is to scan quickly. Q: Is a link QR better because I can update my details later? A: Only if you are willing to keep paying for hosting and domain renewals for as long as the card exists. A link QR can be redirected, but that also means it can break, expire, or start showing ads if the service changes hands. Most people who receive a link QR will not save the contact immediately, so a later update helps only the minority who bookmarked the page. A static vCard QR cannot be edited after printing, which is a feature: what you hand out today will read exactly the same next year. If you need updatability, use a link QR on a domain you own and control, and accept the ongoing cost of around €10–€50 per year. Q: Do digital business cards expose me to spam or tracking? A: A static vCard QR exposes only the data printed in the code, so the main risk is over-sharing information such as your personal mobile number or home address. A link QR or an app-based card can expose far more: the service may log the scanner's device, approximate location, time of scan and referrer. Some platforms then use that data for analytics or marketing. To minimise exposure, generate a static vCard QR with a tool that runs in your browser and does not upload the file, include only the fields you would print on paper, and avoid services that require the visitor to create an account before seeing your details. ## How to Share a Wi-Fi Password Using Free Online Tools https://siteb.biz/guides/share-wifi-password-safely Reviewed 2026-08-08 · QR codes and sharing - A Wi-Fi QR code stores your network name and password in plain text, so the image is as sensitive as the password itself. - Phone-to-phone sharing works well for family but is awkward or impossible for strangers and mixed ecosystems. - A guest network is the only method that limits what visitors can reach on your local network. - Never post a Wi-Fi QR code or password where the public, search engines, or future guests can keep a copy. - Free online tools are handy for a quick QR, but a café or holiday let should use router-level guest networking and get professional advice if payment data is involved. Giving someone your Wi-Fi password usually means spelling it out or watching them mistype it. Free online tools can skip that ritual: a QR code, a phone-to-phone prompt, or a printed card lets a visitor connect without anyone saying a word. The catch is that a QR code is just the password in picture form; anyone who scans it has the same access as someone who typed it in. For a home with friends, that trade-off is usually fine. For a café, a holiday let, or anyone renting rooms, the right answer is usually a guest network, not a prettier way to hand out the main password. This guide compares the QR method, phone-to-phone sharing, and guest networks for three real situations, and flags where a free online tool is the wrong tool for the job. ### Why spelling the password aloud is the weakest link Reading a password out loud solves the immediate problem, but it creates several lasting ones. Anyone within earshot — a customer at the next table, someone on a video call, or a child repeating what they heard — gets the same credentials. Typed passwords are also saved to the guest’s keychain or phone, so they remain valid long after the visit ends. If you later change the password to revoke access, you must track down every device that stored it. For a home with trusted friends this matters little; for a rental or café it is a slow leak of access. If your password is short enough to read aloud quickly, it is probably short enough for a neighbour to guess. ### What a Wi-Fi QR code really contains A Wi-Fi QR code is not a magical key. It is a standardised string that tells the phone which network to join and what password to use. The image contains the SSID, the security type (usually WPA2 or WPA3), the password in plain text, and sometimes a hidden-network flag. Because the password is inside the image, anyone who can scan or save the QR can extract it. That makes a printed poster in a holiday let the functional equivalent of writing the password on the wall. The safest free online tools are the ones that build the QR image entirely inside your browser, because the SSID and password never leave your machine. | Network name (SSID) | Tells the device which network to join; not a secret on its own. | | Security type (WPA2/WPA3) | Lets the phone negotiate the connection correctly. | | Password | Anyone with the image can extract and reuse it. | | Hidden-network flag (optional) | Reveals that a hidden network exists. | A laminated QR poster in a holiday let is effectively a permanent leak to every future guest. ### Phone-to-phone sharing: Apple, Android, and their limits Modern phones can share Wi-Fi credentials over Bluetooth or a local peer-to-peer link, which avoids the awkwardness of reading a password aloud. On an iPhone running iOS 11 or later, tapping the network and choosing Share Password sends the details to a nearby contact who is in your address book. On Android, Nearby Share can pass Wi-Fi details to another Android device within range. Both methods are convenient, but both still store the password on the recipient’s phone, and neither works well across iOS and Android. They are built for family and friends, not for customers or short-stay guests. - iOS Share Password: both devices need iOS 11+, Bluetooth and Wi-Fi on, and the recipient in your Contacts. - Android Nearby Share: the receiving device must accept the prompt while within a metre or two. - Cross-platform: sending from iPhone to Android usually means falling back to reading or typing the password. - Both methods save the password to the guest’s saved networks, so access persists after they leave. - Neither method gives you a way to revoke access without changing the Wi-Fi password. Phone-to-phone sharing is only as private as the relationship between the two people. ### Guest networks: the real answer for cafés and holiday lets A guest network is a separate wireless network broadcast from the same router. It gives visitors internet access while keeping them off the network where your printer, file server, payment terminal, or smart home devices live. Most modern routers include a guest-network setting at no extra cost; you simply give it a different name and password. Business routers add useful controls such as bandwidth limits, scheduled hours, captive portals, and voucher codes that expire at checkout. If your current router lacks the feature, a consumer mesh or Wi-Fi 6 unit costs €60–€180 depending on speed and coverage, while a business-grade access point runs €120–€400 depending on PoE, central management, and Wi-Fi 7 support. Professional installation for a small café or holiday let is typically €150–€600, rising with cabling length and the number of access points. | Device isolation | Stops guests from seeing your printer, NAS, or payment hardware. | | Separate password | Lets you rotate or disable guest access without affecting your own devices. | | Bandwidth limits | Prevents one visitor from saturating the connection. | | Captive portal | Shows terms of use before access; useful for cafés and co-working spaces. | | Vouchers or schedules | Holiday lets can issue codes that expire at checkout or disable the network off-season. | A guest network separates visitors from your devices, but it does not encrypt the internet traffic end-to-end. ### Where not to put your Wi-Fi QR code or password The most common mistake is treating a Wi-Fi password like a public amenity. Once it leaves your mouth or your private message, you lose control over where it travels. Search engines cache listing pages, guests photograph sticky notes, and email inboxes are shared or hacked. A QR code posted on social media is a password that the whole world can save. The safer rule is to share credentials only through channels you can delete or replace, and only with people who are physically present. - On public social media, listing sites, or review replies — search engines and strangers keep copies. - On a sticky note stuck to the router in a rental property. - In a pre-arrival email that sits in the guest’s inbox indefinitely. - On a poster visible from the street or in photographs shared online. - In a group chat with people who are not the current guest. If you cannot delete it or change it remotely, it is the wrong place for a password. ### When a free online tool is the wrong choice Free online tools are handy for one-off tasks: generating a QR code for a family dinner, counting characters in a sign, or creating a strong password. They are not network architects. If you run a café, a holiday let, or any premises that handles payment or booking data, a browser QR generator will not give you device isolation, logging, or compliance. In those cases the right tools are a business router, a managed Wi-Fi service, or a network professional. If you are managing dozens of rooms, a spreadsheet and a consumer router will not scale; you need a property-management system or managed Wi-Fi. The same applies if you are subject to data-protection rules: a QR code is not a substitute for a properly segmented network. No browser tool can create a guest network for you; that has to be configured in your router. ### A practical workflow for three common situations Choosing a method depends on how much you trust the visitor and how often the crowd changes. A trusted friend at home needs little ceremony. A café needs a rotating crowd without giving away the keys to the back office. A holiday let needs easy access that expires when the booking does. The table below matches each situation to a sensible method and the main risk to watch. | Home with friends | Phone-to-phone sharing or a browser-only QR | The password stays on their phone after the visit. | | Café or co-working space | Guest network with captive portal | A QR poster alone hands the same password to every visitor forever. | | Holiday let or short-stay rental | Guest network plus a printed QR only for that network | Future guests may keep the code or share it online. | ### The bottom line: convenience versus control Free online tools are a sensible starting point when you need a quick QR code for a trusted guest and you use a browser-only generator that never uploads your SSID or password. They stop being the right choice when the visitor is a paying customer, the device is a business asset, or the network touches payment or personal data. In those cases, spend money on the right layer of the problem: a guest network, a business access point, or professional advice. No sharing shortcut removes the need to think about who has access and for how long. This guide is not technical, legal, or financial advice for your specific premises. Q: Can someone hack my Wi-Fi from a QR code? A: A QR code does not hack your network in the cinematic sense; it simply contains the password in a machine-readable format. Anyone who scans the image gets the same access as someone you told the password to. If that image is posted publicly, saved in a guest’s photos, or shared online, you have effectively given the password to the world. The encryption between your router and the phone still works, but encryption does not help once the password is known. The fix is to change the password and, for regular visitors, use a guest network instead of sharing the main credentials. Q: Is it safe to use a free online QR generator for my Wi-Fi password? A: It is safe only if the generator runs entirely in your browser and never sends the SSID or password to a server. You can check this by opening the site, disconnecting from the internet, and trying to create the QR: a genuine browser-only tool will still work. Server-side generators may log your credentials, and some monetise by storing or analysing inputs. For a one-off home use, a reputable browser-only tool is fine. For a business, a holiday let, or any environment with turnover or payment data, skip the QR generator and configure a guest network on your router. Q: Should guests get my main Wi-Fi password or a separate guest network? A: For a home with friends or family you trust, sharing the main password is usually acceptable, though you should still change it if the relationship changes or a device is lost. For a café, a holiday let, or any property with paying guests, a separate guest network is the minimum sensible setup. It keeps visitor devices away from your printer, payment hardware, and personal files, and it lets you rotate or disable the guest password without touching your own devices. If your router does not support guest networks, a new consumer unit costs €60–€180 depending on Wi-Fi speed and coverage, while professional installation for a small business runs €150–€600. ## QR Codes: Why Yours Will Not Scan and Which Free Online Tools to Trust https://siteb.biz/guides/qr-code-guide Reviewed 2026-08-08 · QR codes and sharing - A QR code is only as reliable as its quiet zone, contrast, and module size. - Higher error-correction levels protect against damage but reduce capacity or increase density. - Dynamic short links let you change destinations later, but your printed code dies with the service. - Logos and colours can work, yet each one eats the safety margin that error correction provides. - Test with an old phone, real lighting, and the final substrate before approving any print run. A QR code is a two-dimensional barcode that a phone camera reads as a string of text, usually a URL. Error-correction algorithms then reconstruct the message even if part of the pattern is dirty, creased or obscured by a logo. That resilience is real, but it is not infinite: poor contrast, a missing quiet zone, or too much data can still make a code unscannable. Free online tools make it tempting to generate a code in ten seconds and paste it onto a flyer. That is fine for a one-off test, yet business use introduces real risk. Print resolution, colour choice, error-correction level, and the difference between a static code and a tracked short link all affect whether a customer reaches your site or gives up. This guide covers the mechanics that matter and the failure modes that actually show up in the field. ### What a QR code actually stores A QR code stores text as a grid of black and white modules. The most common payload is a URL, but the specification also supports plain text, contact cards, phone numbers, Wi-Fi credentials and SMS messages. The data is converted into a binary stream, padded, and spread across the grid using Reed-Solomon error correction. Version 1 is a 21×21 grid; higher versions hold more data at the cost of density. A longer URL means more modules, smaller individual squares at a fixed print size, and a harder scan for older cameras. - Numeric mode: digits only, highest capacity. - Alphanumeric mode: uppercase letters, digits and a limited set of symbols. - Byte mode: any UTF-8 text, including lowercase URLs and special characters. - Kanji mode: Shift JIS characters, rarely useful outside Japan. A vCard or long URL can force a higher version and smaller modules, so keep the payload short. ### Error correction: the safety margin you can choose QR codes offer four error-correction levels. The level determines how much of the pattern can be damaged before the code becomes unreadable. It is not about making the code stronger in every sense; higher levels consume space that could otherwise hold data, so they reduce capacity or increase density. Level L is enough for a clean digital screen. Level H is the choice for outdoor stickers or codes that will carry a logo, but it cuts usable data capacity by roughly a third. Most generators default to M, a reasonable middle ground. | L (Low) | ~7% damaged modules | Clean screens, controlled indoor use | | M (Medium) | ~15% damaged modules | Flyers, handouts, general print | | Q (Quartile) | ~25% damaged modules | Outdoor labels, slight abrasion risk | | H (High) | ~30% damaged modules | Heavy wear, logos, manufacturing marks | Higher correction does not make a blurry or low-contrast code readable; it only repairs missing modules. ### Quiet zone, contrast, and colour choices The quiet zone is the empty margin around the three position markers. Without it, the scanner cannot reliably locate the code. A safe quiet zone is at least four modules wide on every side; many designers crop it to make the code fit a layout, which is the single most common cause of intermittent scans. Contrast matters more than colour. Black on white works because the luminance difference is large. Pale blue on white, dark grey on black, or gradients can drop below the threshold. Colour is fine if the dark modules remain clearly darker than the light ones. - Keep a quiet zone of at least four modules on all sides; more is safer. - Use high luminance contrast, not just different hues. - Avoid gradients, photographs, or tints behind the code. - If the brand demands colour, make the modules the darkest available colour and the background the lightest. - Never reverse the code out of a dark background unless every module is clearly lighter than the surround. A glossy laminate can create specular highlights that wash out modules; matte finishes scan more reliably. ### Print size and scanning distance: a rule of thumb The smallest printable unit is the module, not the overall code. A useful rule of thumb is that the module width should be at least one three-hundredth of the expected scanning distance. If you expect customers to scan from half a metre, each module should be about 1.7 mm wide. A version-1 code is 21 modules across, so the whole code would need to be roughly 36 mm. A longer URL may push the code to version 5 or higher, requiring a proportionally larger print to keep modules the same size. Printing smaller than the calculation invites failure. | 0.3 m (handheld) | 1.0 mm | 21 mm | | 0.5 m | 1.7 mm | 36 mm | | 1.0 m | 3.3 mm | 70 mm | | 2.0 m | 6.7 mm | 140 mm | | 5.0 m (posters/billboards) | 17 mm | 350 mm | These are minimums for ideal conditions; add 30–50% for low light, curved surfaces, or older phones. ### Putting a logo in the middle: a calculated gamble A logo sits in the centre because that area is least critical for decoding. Error correction rebuilds the missing modules, but only up to the limit of the chosen level. A central image that covers 5–10% of the code is usually safe at Level H; covering 15% or more begins to trade away the safety margin. The logo must also be treated as damage: it should sit on a flat, contrasting background, not bleed into surrounding modules. Many branded codes fail because the logo overlaps position markers or the timing patterns. If brand guidelines demand a large logo, increase the print size so modules stay large enough to compensate. - Use Level H if you must include a logo. - Keep the image under 10% of the total area. - Centre it away from the three finder patterns in the corners. - Place the logo on a solid block, not a transparent overlay. - Test with the oldest phone and worst lighting you can find. A logo is not the only source of 'damage': creases, stains, and low resolution also consume the same error-correction budget. ### Static codes versus dynamic short-link services A static QR code embeds the final URL directly. Print it, and it will work as long as the URL exists. A dynamic service encodes a short link that redirects to a destination you can change later. That is genuinely useful for campaigns, A/B tests, and correcting typos after a poster has gone to print. The trade-off is dependency. If the service changes its redirect rules, raises prices from €0 to €10–€25 per month as usage grows, or shuts down, your printed code becomes a dead link or worse, redirects somewhere else. For long-life assets such as vehicle decals or product labels, a static code usually wins. - Gain: change the destination URL without reprinting. - Gain: scan counts, locations, and device statistics. - Risk: the service controls the redirect; outages become your outage. - Risk: subscription costs can rise from €0 to €10–€25 per month as usage grows. - Risk: if the service closes, your printed materials become landfill. Download a vector file and a tested screenshot of the decoded URL before you commit to a large print run. ### When free online tools are enough, and when they are not For a single code, a short URL, and a one-off flyer, free online tools are perfectly adequate. Look for a generator that runs in the browser, exports SVG or high-resolution PNG, and lets you set the error-correction level. Avoid tools that force you to log in, watermark the output, or redirect through the provider's own short-link domain. For batch work, event badges, asset labels, or brand-controlled packaging, use a spreadsheet mail-merge, a design programme with a QR plug-in, or a professional printer who provides verified proofs. The cost of fixing ten thousand printed items far exceeds the cost of doing it properly once. If a generator asks for an email address to 'send your code', your data is leaving the browser. ### Test before you commit to print Testing is not optional. Scan the code with at least two different phones, one of them more than three years old. Try it under fluorescent light, daylight, and at the intended scanning distance. If the code will be on a curved surface such as a bottle or a vehicle, scan it while it is wrapped, not flat. Print a single proof on the actual substrate and laminate if one will be used. Only after it passes those checks should you approve a run of €50–€500 worth of materials, depending on quantity and finish. - Scan with iOS and Android devices from different years. - Test in low, bright, and mixed lighting. - Try the actual scanning distance and angle. - Test on the final material, including any lamination or coating. - Confirm the decoded URL matches exactly, including https://. A code that scans perfectly on a glossy monitor may fail on matte paper under a shadow. Q: Why does my QR code scan on my phone but not on my customer's phone? A: Newer phones have better sensors, faster autofocus, and more aggressive decoding software. A code that scans on a flagship device may fail on a three-year-old budget phone, especially if the modules are small, the contrast is marginal, or the lighting is poor. Camera apps also vary: some tolerate a damaged quiet zone or a glossy reflection better than others. The only reliable test is to scan with the oldest and cheapest device your audience is likely to own, under real-world conditions. If it fails there, it will fail for some customers. Q: Can I change the URL in a QR code after printing? A: Only if the code encodes a dynamic short link. A static code contains the actual URL, so what is printed is fixed. Dynamic services let you change the destination through a dashboard, which is useful for correcting mistakes or rotating campaigns. The catch is that the redirect is controlled by the service, not by you. If the service changes its terms, suffers an outage, or disappears, your printed code points to a broken or hijacked link. For materials that will last years, a static code is usually the safer choice. Q: Are free online QR code generators safe to use? A: Many are safe for a quick test, but not all handle your data the same way. A generator that runs entirely in your browser never sends the URL to a server, which is ideal. Others upload the payload, create the code on their server, and may store or log it. Some wrap your URL in their own short link so they can track scans, or they place the redirect under their control. For sensitive destinations, business use, or anything you will print in volume, prefer a browser-only tool that exports SVG or high-resolution PNG and never asks for a login. ## Are Free Online Tools Safe? A Checklist Before You Paste https://siteb.biz/guides/is-this-free-online-tool-safe Reviewed 2026-08-07 · Passwords and privacy - Open the browser's Network tab, paste a distinctive dummy string, and trigger the action to see whether your input leaves your device. - HTTPS only encrypts data in transit; it does not stop a site from storing or sharing what you submit. - Never paste passwords, private keys, legal documents, medical records or financial details into an untrusted browser tool. - Mandatory accounts, missing privacy pages, and results that arrive after a suspicious delay are strong warning signs. - For money, health, legal or reputation-sensitive tasks, use a native app, a spreadsheet, or a qualified professional instead of a free online tool. Free online tools are convenient until the moment you paste something you should not. A browser-based converter, password generator or date calculator can solve a small problem in seconds, but many of them send your input to a server to do the work. That is fine for a public weather forecast and a mistake for a contract clause, a medical record number or the only copy of a private key. This guide shows how to check, in about a minute, whether a tool processes your data inside your browser or ships it elsewhere. It also explains the legal gaps that terms-of-service pages usually leave open, and when you should close the tab and use a spreadsheet, a native app or a paid professional instead. Nothing here is legal, financial or medical advice; it is a practical filter for a common risk. ### What 'runs in your browser' really means When a tool says it runs locally, the calculation, conversion or generation happens on your own computer using JavaScript loaded from the site. The page may still fetch the code itself, fonts or analytics scripts from the internet, but your input should not appear in a network request. True local tools can often work offline after the page loads, and they do not need your email address, a login or a credit card for a one-off task. A privacy policy that says the site processes data on its servers is at least honest; a missing page is a stronger negative signal. ### How to check the Network tab in under a minute Every mainstream desktop browser has a built-in Network panel that logs every request a page makes while you interact with it. The test takes about sixty seconds and needs no extensions or technical background. You simply paste dummy data that is distinctive enough to spot in a request, trigger the action, and watch for outbound traffic while the result appears. If the only new lines are analytics, fonts or style sheets, the tool is likely local. If your dummy string shows up in a request body or URL, the site is sending your input somewhere else. - Open the tool in a new tab and press F12 (or Ctrl+Shift+I / Cmd+Option+I) to open Developer Tools. - Click the Network tab, then clear any existing log by clicking the clear icon or pressing Ctrl+Shift+E. - Type or paste a distinctive but fake string, such as TEST1234-siteb-priv-check, into the tool. - Trigger the action: convert, generate, count or calculate. - Watch the Network log. If your string appears in a request URL or body, the tool uploaded it. - Repeat once with a file drag-and-drop if the tool accepts documents, since some tools upload files while claiming the preview is local. A few tools send only a hashed or encrypted version of your input; that is better than plain text, but you still have to trust the operator. ### Red flags that should make you close the tab Some warning signs are easier to spot than a network request. They do not prove the operator is malicious, but they do prove the site was not built with your privacy as the default. A red flag should shift the burden of proof: the site now has to show you why it is safe, rather than you having to prove it is dangerous. These six patterns come up again and again on free online tools that later turn out to be careless with data. - A mandatory account or email for a trivial calculation, such as a word count or unit conversion. - No privacy page, terms page or contact details, or a privacy policy dated years ago and copied from a template. - A noticeable delay between input and result for a task your phone could do instantly, suggesting a remote server. - Requests for permissions that have nothing to do with the job: camera, microphone, contacts or location. - An upload button that opens immediately rather than processing locally after the file is selected. - Aggressive monetisation: multiple pop-ups, auto-downloads, or prompts to install browser extensions. One red flag alone is not proof of harm, but it should move the burden of proof onto the site. ### What the terms of service usually allow Most free tool sites rely on standard legal clauses that favour the operator. You rarely ‘sell’ them your data in a literal sense, but you usually grant a broad licence to use, display, adapt and distribute whatever you submit. That licence is often described as irrevocable, worldwide and royalty-free, and it can survive even after you close the tab. The terms also tend to disclaim accuracy and liability, which means a wrong conversion or leaked file is your problem, not theirs. | Broad licence to user content | The site can keep, modify and share what you pasted, including in product demonstrations or datasets. | | No warranty / as-is | If the calculation is wrong and you lose money, the operator is not liable. | | Right to share with service providers | Your data may be passed to hosting, analytics or advertising partners. | | Public data may be indexed | If the tool produces a shareable link, search engines or archive sites may store it. | | Arbitration and jurisdiction | You may have to sue in another country and waive class-action rights. | Terms pages are written to protect the operator; read them as a description of risks, not a guarantee of safety. ### Data you should never paste into a third-party tool The safest rule is simple: if losing control of the data would cause harm, embarrassment, financial loss or legal exposure, do not paste it into a browser tab you do not trust. Browser tools are fine for public or low-value information, but they are the wrong place for secrets, regulated data or anything protected by a confidentiality agreement. The following categories are almost always better handled on your own device or by a professional bound by duty. | Passwords and private keys | A server can store or leak them; keyloggers and malicious scripts are also a risk. | A password manager or offline generator. | | Legal contracts and NDAs | Confidentiality obligations may be breached; the site may gain a licence to the text. | A local word processor or solicitor-reviewed workflow. | | Medical record numbers and test results | GDPR and HIPAA-style rules may apply; accidental disclosure is irreversible. | Your health portal or a clinician. | | Bank account numbers, card details and tax identifiers | Fraud and identity theft; many terms disclaim liability for errors. | Your bank's app, accountant software or a spreadsheet on an encrypted drive. | | Unpublished source code and trade secrets | Broad content licences and indexing can expose intellectual property. | A local IDE or company-managed code host. | | Scanned identity documents | Facial images and document numbers are high-value targets for fraud. | Offline scanning or the requesting institution's official portal. | This list is illustrative, not exhaustive. When in doubt, keep the data on a device you control. ### When free online tools are the wrong choice Free browser tools suit quick, low-stakes tasks where a small error does not matter much: converting a recipe from cups to grams, counting characters in a social post, or generating a one-time QR code for a public Wi-Fi password. Once the output affects money, health, legal rights, employment or reputation, the convenience stops being worth the risk. At that point you need an audited app, a regulated professional, or at minimum a spreadsheet you control on an encrypted device. | Currency or tax calculations for a business | Exchange rates may be stale; terms disclaim accuracy. | Accounting software or a qualified accountant (€50–€250 per hour depending on complexity). | | Medical dosage or date calculations for treatment | A wrong result can affect care; no tool knows your full history. | A clinician or pharmacy professional. | | Drafting legal notices or contracts | Template errors and confidentiality breaches are expensive. | A solicitor (€150–€400 per hour in most EU markets). | | Generating passwords for high-value accounts | You cannot verify the generator's entropy or honesty. | A reputable password manager with audited code. | | Batch processing sensitive files | Uploading dozens of documents multiplies exposure. | A local script, desktop app or private server. | Prices are rough EU ranges and vary by country and specialism; always agree a fee before engaging a professional. ### A practical workflow for everyday tasks You do not need to audit every site. A short routine keeps most risks low. Start by asking whether the data is public, then run the network-tab test on any new tool, keep a short list of trusted tools for repeated jobs, and default to native apps when the stakes rise. Siteb.biz is an independent directory of free online tools and runs its own small utilities entirely in the browser; use this checklist on any third-party subdomain the same way you would on any unfamiliar site. - Treat every new tool as untrusted until the Network tab confirms it is local for your task. - Use a fake but identifiable string during the test, never real data. - Bookmark two or three tools that pass the test for each category you use often. - Avoid tools that ask for accounts, permissions or file uploads for jobs that do not need them. - Keep sensitive work on your device: spreadsheets, password managers, desktop apps or professional services. If a tool needs a server for a legitimate reason, such as generating a QR code image on demand, read its privacy policy and decide whether the trade-off is acceptable. Q: If a tool says it is secure, should I trust it? A: ‘Secure’ is a marketing word, not a technical guarantee. A site can use HTTPS — which only protects data in transit — and still store everything you submit on its servers. Look for specific claims: the code runs in your browser, no data leaves your device, no account is required. Then verify those claims with the Network tab test described above. A privacy policy helps, but it is not a promise of local processing. If the tool handles something sensitive and the operator does not clearly explain where the computation happens, treat it as server-side and choose a local app or a professional instead. Q: Can I safely use a QR code generator or password generator online? A: It depends on what the generator creates and how it works. A QR code for a public café menu is low risk; a QR code embedding a private Wi-Fi password or payment request is not. Password generators are only safe if the code runs locally and you can inspect or trust the entropy source. Even then, generating a password inside your browser exposes it to any malicious script loaded by the page or a compromised extension. For high-value accounts, use an audited password manager instead of a browser tab. For one-time, low-stakes codes, a local browser generator you have tested is usually acceptable. Q: What is the difference between a local browser tool and a server-side tool? A: A local browser tool downloads JavaScript to your computer and performs the calculation or conversion there. Your input never has to leave the device for the core task, though the page may still load code, fonts or analytics from the internet. A server-side tool sends your input to a remote computer, which processes it and returns a result. Server-side is necessary for heavy tasks, large files or features that need a database, but it creates a permanent record on someone else's hardware. The Network tab test is the fastest way to find out which model a site is using for your specific action. ## Password Managers: What They Solve, When Free Online Tools Fail https://siteb.biz/guides/password-manager-basics Reviewed 2026-08-07 · Passwords and privacy - A password manager mainly protects you from password reuse and weak credentials, not from every possible attack. - An encrypted vault breach is far less dangerous than a plaintext leak, but only if your master password is strong. - Your master password cannot be reset by the provider, so forgetting it usually means losing the vault. - Write down a recovery plan on paper before you need it, and keep the paper somewhere safe. - A free online tool is fine for generating passphrases locally, but it is the wrong place to store your actual vault. A password manager is the one piece of security software that almost every expert agrees is worth using. It will not make you immune to breaches, but it removes the two habits that cause most harm: reusing passwords across sites and storing them in easy-to-read places. For many people, a reputable manager is safer than doing nothing; for others, the wrong one is simply another cloud account they do not control. This guide covers what a manager protects you from, how a vault breach differs from a plaintext leak, and why the master password is both the strongest and weakest link. We also explain when a free online tool is the wrong place for credentials, and when a spreadsheet, native app, or professional is the safer bet. Siteb.biz is an independent directory of browser tools, not a vendor; this is guidance, not a recommendation. ### What a password manager actually stops The main protection is not against the vault provider being hacked; it is against the everyday attacks that work because people reuse passwords or choose weak ones. When every account has a unique, random password, a breach at one service cannot be used to log in to another. A manager also fills credentials only on the site they belong to, which makes some phishing attacks harder, and it removes the temptation to keep passwords in email drafts, spreadsheets, or notes on a phone. - Credential stuffing: attackers take a leaked password from one site and try it on dozens of others. - Weak or predictable passwords: the manager generates strings you would never memorise. - Phishing by lookalike domains: most managers refuse to fill credentials on the wrong site. - Passwords stored in plain sight: email drafts, spreadsheets, and sticky notes. - Device loss: your vault can be restored on a new phone or laptop, provided you know the master password. ### Encrypted vaults vs plaintext leaks Not all breaches are equal. If a service stores your password as plaintext, anyone who steals the database can read it immediately. If it stores only a hashed password, the attacker must crack it, and a strong unique password usually survives. A password manager adds another layer: your vault is encrypted on your device before it leaves, so the provider typically holds only ciphertext. Even if that ciphertext is stolen, an attacker still needs your master password to unlock it. That is a meaningful difference, but it is not invulnerability. | Plaintext leak | Your actual password, readable immediately | Nothing; change the password at once | | Hashed password leak | A scrambled hash that must be cracked | A long random password and slow hashing | | Encrypted vault theft | Ciphertext without the decryption key | Your master password and the key derivation settings | | Master password guessed | Full access to the decrypted vault | Length, randomness, and not reusing it elsewhere | Some managers have been criticised for weak default iteration counts; check the security settings if you are concerned. ### The master password problem The strongest encryption in the world becomes useless if your master password is short, personal, or reused. Because the vault provider does not know this password, they cannot reset it for you. That is the point: you alone hold the key. It also means that if you forget it, the data is gone. The best master passwords are long passphrases—five or six random words, or a long string of characters you can type reliably. For this job, the right free online tool is a passphrase generator that runs entirely in your browser; anything that sends candidate passwords to a server is learning your future master password. - Use at least 16 characters; 20 or more is better if you can type it consistently. - Do not reuse your master password anywhere else, not even as a security answer. - Avoid names, birthdays, pet names, and favourite quotes that appear online. - Generate it with a browser-based passphrase tool that does not upload the result. - Practise typing it for several days before you rely on it; muscle memory matters. ### The recovery plan you should write down before you need it Most people only think about recovery after they are locked out. Write the plan down on paper before that happens. Store the paper somewhere safe: a locked drawer, a safe, or with a trusted person, depending on your threat model. Do not keep the only copy next to the computer. The sheet should not include your master password unless you have decided, consciously, that the risk of forgetting it outweighs the risk of someone else finding it. For most people, a strong hint is enough. | Vault provider name and login email | You cannot recover what you cannot find | Paper and a secondary email account | | Two-factor authentication method | Losing the authenticator can lock you out | Backup codes in a sealed envelope | | Emergency contact or inheritance contact | If you are incapacitated, someone must access accounts | With your will or a trusted relative | | Master password hint | Jogs your memory without revealing the phrase | On paper, separate from the device | | Recovery key or export file | Lets you move to another manager or restore access | Encrypted USB drive or safe deposit box | A paper record is not old-fashioned; it is offline, durable, and harder to steal remotely than a cloud note. ### When a password manager is the wrong choice There are situations where a manager adds friction without adding safety. If you cannot remember a strong master password and have no secure place to store a hint, you will eventually lock yourself out. If you need family members to access accounts when you cannot, a shared spreadsheet in a secure location may be more practical—though less secure in other ways. Regulated businesses sometimes need audit trails and centralised control that consumer managers do not provide. In those cases, a free online tool is the wrong place for your vault; the right answer may be an enterprise product, a native OS keychain, or a professional. - You have no recovery plan and no trusted contact to hold a backup. - You share credentials with family members who will not use the same app. - Your employer or regulator requires logging, escrow, or centralised administration. - You need access on devices that cannot run the manager or its sync service. - You are unable to memorise or reliably type a long master password. ### Choosing among the four common types Consumer password tools fall into four categories, and the best one depends on how much control you want and how much you are willing to spend. Browser-built-in managers are convenient but tied to one ecosystem. OS keychains are better integrated and encrypted, but still limited. Cloud managers sync across devices and add sharing, at the cost of trusting a third party. Self-hosted options give you full control, yet they also give you full responsibility for backups and updates. | Browser built-in | €0 | Convenient, but tied to one browser account and harder to export | | OS keychain | €0 (included with the device) | Tightly integrated, yet limited sharing and recovery options | | Cloud password manager | €0–€60/year (free tier vs paid features) | Cross-device sync and sharing, but you trust the provider's security | | Self-hosted vault | €0–€200/year (hardware, hosting, and your time) | Full control, but you are solely responsible for backups and updates | This is general guidance, not legal, financial, or regulated security advice; consult a professional if you handle sensitive organisational data. ### A short setup checklist - Choose the type of vault that matches your recovery needs and budget. - Generate a long master passphrase using a browser-based tool that runs locally. - Enable two-factor authentication on the vault account and save backup codes. - Add a few low-risk accounts first, then gradually migrate the rest. - Write down the recovery sheet and store it offline before deleting old password lists. Do not treat setup as a one-off event; review your recovery kit at least once a year. Q: Can the password manager company see my passwords? A: In a well-designed service, the answer is no. The provider stores only encrypted data and never receives your master password in a usable form; decryption happens on your own device. This is often called zero-knowledge architecture. It is not magic, though. You are still trusting that the code, the update mechanism, and the cryptography are implemented correctly. A dishonest or compromised provider could, in theory, serve a malicious update that captures your master password when you type it. That is why many people prefer open-source managers or those that have been independently audited, and why you should keep the app updated. Q: What should I do if my password manager reports a breach? A: First, find out whether the stolen data was encrypted vault files or plaintext account information. If only ciphertext was taken, change your master password as a precaution and review your two-factor authentication settings; the old vault files will remain protected by the previous master password unless it was weak. If plaintext data such as email addresses or URLs was exposed, watch for targeted phishing and consider rotating the most sensitive account passwords. If you had reused your master password anywhere else, change those copies immediately, because that reuse is now the real danger. Q: Is a free online password generator safe to use? A: It depends on what it does with the data. A generator that runs entirely in your browser and never uploads the result can be perfectly safe for creating one-off passwords. The risk appears when a tool sends your generated password to a server, stores it in analytics, or runs on a site you do not trust. For everyday passwords, a local browser-based generator is reasonable. For your master password, use a dedicated passphrase generator that also stays local, and never paste the result into an online form just to test it. When in doubt, generate passwords offline. ## How Strong Does a Password Need to Be? A Free Online Tools Guide https://siteb.biz/guides/how-strong-should-a-password-be Reviewed 2026-08-07 · Passwords and privacy - Password strength is measured in bits of entropy, and each extra bit doubles the attacker’s expected work. - Length matters far more than symbol substitution, because attackers already model tricks like P@ssw0rd1. - A leaked database lets an attacker test billions of guesses per second offline, so short passwords fall quickly. - NIST no longer recommends forced 90-day rotation or mandatory complexity rules; use long, unique passwords and screen for breaches. - Browser-based free online tools can generate and check passwords safely, but a password manager is the right place to store high-value credentials. Most password advice is folklore dressed as engineering. People still believe that swapping letters for symbols makes a password strong, or that changing it every three months keeps attackers out. The truth is simpler and more numerical: a password’s real strength is measured in bits of entropy, which is just a count of how many guesses an attacker must make before they expect to find yours. Free online tools can calculate that number in a browser, but only if you understand what the number means and what it ignores. At siteb.biz we list free online tools that run entirely in your browser, including password generators and entropy checkers that never upload your secret. This guide explains the arithmetic behind the rating, shows why length dominates complexity, and tells you when even the best browser-based checker is the wrong place to put a high-value credential. ### Entropy in plain English Entropy is not a magic score; it is the logarithm of the number of possible passwords of a given length and character set. If your alphabet has N symbols and your password is L symbols long, there are N^L possibilities, and the entropy is L × log₂(N) bits. Each extra bit doubles the attacker’s expected work. A totally random eight-character password from the 95 printable ASCII characters has about 52.6 bits of entropy. That sounds high until you learn that a single modern GPU can test tens of billions of hashes per second. | Lowercase a–z | 26 | ~4.7 | | Lowercase + uppercase | 52 | ~5.7 | | Alphanumeric | 62 | ~5.95 | | All printable ASCII | 95 | ~6.57 | | Diceware word list | 7,776 | ~12.9 | These figures assume the password is chosen randomly; human-chosen passwords usually have far less entropy. ### Why length beats symbol substitution Security trainers used to praise passwords like P@ssw0rd1 because they contained uppercase, lowercase, digits and symbols. Modern rule-based crackers model those exact substitutions before they ever try brute force. Hashcat and John the Ripper come with ‘leetspeak’ rules that turn ‘password’ into ‘P@ssw0rd’, ‘Passw0rd!’ and hundreds of siblings in milliseconds. Adding a symbol to a short, predictable base is decoration, not structure. It may satisfy an outdated policy, but it does not meaningfully change the attack path. - Replace ‘a’ with ‘@’ and ‘o’ with ‘0’ - Append the digit 1 or the current year - Capitalise the first letter and add an exclamation mark at the end - Run keyboard walks such as ‘qwerty’ and ‘1q2w3e’ - Combine a dictionary word with a single digit or symbol If your password is on a leaked list, no amount of symbol swapping will help; the attacker already has it. ### The offline crack in numbers When a service is breached and its password database leaks, the attacker works offline. They can rent hardware, try guesses for days, and never trigger a lockout. The table below shows how long a single mid-range GPU (roughly 40 billion SHA-256 guesses per second) would need to exhaust the key space of random passwords. Real attacks stop early because users cluster around common patterns, but the table shows the theoretical ceiling for fully random credentials. | 8 characters | Lowercase a–z only | < 10 seconds | | 8 characters | All 95 printable ASCII | ~2 days | | 10 characters | All 95 printable ASCII | ~50 years | | 12 characters | All 95 printable ASCII | ~400,000 years | | 16 characters | All 95 printable ASCII | ~10^13 years | | 6 random words | 7,776-word Diceware list | ~150,000 years | These figures ignore early hits from common patterns; they are worst-case for the attacker, best-case for you. ### What a leaked-database attack really looks like Attackers rarely ‘guess’ a password through a login form. They download a stolen database, identify the hash algorithm, and run it at hardware speed. The first pass uses previously leaked passwords, then dictionary words with leet rules, then masks for known patterns, and finally brute force for the shortest credentials. A single leaked password from Site A is then tried on Site B, Site C and Site C’s banking partner because people reuse. A cloud GPU instance costs €0.50–€4.00 per hour depending on whether you rent a consumer card or a cluster, and the whole workflow is automated. - Obtain the hashed password database from a breach or purchase it on a forum. - Identify the hash algorithm and whether a unique salt was used. - Run a wordlist plus rules; this cracks 60–80% of typical user passwords within hours. - Apply brute force or masks to the remainder, starting with short lengths. - Use recovered credentials for credential-stuffing attacks against other services. Salting and slow hashes such as Argon2 or bcrypt make this far slower, but many older sites still use fast algorithms like SHA-1 or MD5. ### NIST’s revised advice on rotation and complexity The US National Institute of Standards and Technology withdrew the old ‘change every 90 days’ rule in SP 800-63B. It now says password policies should require a minimum length, allow long passphrases, and reject passwords known to be compromised. Complexity rules that demand uppercase, lowercase, digit and symbol are no longer recommended, because they encourage predictable mutations such as P@ssw0rd1 and do not improve entropy. Rotation is only required when there is evidence of compromise. - Minimum length is more important than character-class requirements. - Do not force periodic changes unless a breach is suspected. - Screen new passwords against lists of compromised credentials. - Allow paste and password managers, which enable longer, unique passwords. - Use multi-factor authentication rather than relying on passwords alone. These are US federal guidelines, but they have been widely adopted by security standards bodies worldwide. ### When a browser-based tool is enough, and when it is not A password generator that runs entirely in your browser can produce excellent randomness and never sees your secret. That makes it useful for one-off accounts, temporary keys, or checking the entropy of a candidate password. It is not the right place to create or store the master password for your password manager, your cryptocurrency seed, or the admin key to a business system. For those, use a vetted native password manager or offline process. A reputable password manager costs €20–€60 per year depending on features and family plans. Free online tools are conveniences, not vaults. - Use a browser-based generator for short-lived or low-risk passwords. - Use a dedicated password manager for banking, email, and primary accounts. - Do not paste a real master password into any online checker, even one that claims no upload. - Store recovery codes on paper or in encrypted storage, not in a notes app. - For shared business credentials, use an enterprise password manager with audit logs. siteb.biz lists tools on their own subdomains and writes these guides; we do not provide the service ourselves, so judge each tool by its privacy policy and whether it runs client-side. ### A practical minimum for 2025 For accounts you care about, use a password manager to create and store a unique password of at least 16 characters drawn from the full printable set, or a passphrase of at least five random words. Turn on multi-factor authentication, ideally using an authenticator app or hardware key rather than SMS. Check whether your email appears in known breaches through Have I Been Pwned, and change any password that is reused or compromised. Treat password strength as one layer in a larger risk decision, not a guarantee. - Use 16+ random characters or 5+ random words for high-value accounts. - Enable multi-factor authentication on email, banking, and password-manager accounts. - Never reuse a password across services. - Check breach-notification services and rotate only compromised credentials. - Prefer a hardware security key or authenticator app over SMS codes. A longer, memorable passphrase is usually stronger than a short complex password you have to write down. Q: Is a 12-character password with symbols strong enough? A: It depends on how it was created and what you are protecting. Twelve fully random characters from the full printable set give roughly 77 bits of entropy, which would take a single GPU hundreds of thousands of years to exhaust offline. However, if the base is a dictionary word with symbols added for decoration, modern rule sets will crack it in minutes. For high-value accounts such as email, banking, or your password manager, 16 random characters or five random words is a safer minimum, and multi-factor authentication is essential. Q: Are free online password generators safe to use? A: They can be, if they run entirely in your browser and do not upload the password or log keystrokes. The best tools use the browser’s built-in cryptographically secure random number generator and work without a login. They are fine for generating one-off passwords or estimating entropy. They are not the right choice for creating a master password, a cryptocurrency seed, or any secret you will store long term. For those, use an audited native password manager or an offline process, and never paste a real master password into an online form. Q: Why did NIST stop recommending password rotation? A: Research showed that forced rotation every 60 or 90 days produced weaker security, not stronger. Users responded by making tiny, predictable changes such as Password1, Password2, Password3, or by writing passwords down. NIST SP 800-63B now advises against mandatory periodic changes unless there is evidence that the password has been compromised. Instead, organisations should enforce minimum length, allow long passphrases, screen new passwords against breach lists, and require multi-factor authentication. ## Free Online Tools for Title Case, Sentence Case and Cleanup https://siteb.biz/guides/text-case-and-cleanup-guide Reviewed 2026-08-06 · Text tools - Free online tools can clean pasted text quickly, but they are not always the right choice for confidential or complex work. - Automatic title case mishandles brand names, acronyms and non-English words because it follows rules, not meaning. - Invisible characters such as non-breaking spaces, smart quotes and zero-width joiners can break CMS imports, CSV files and URL slugs. - Clean lists in a repeatable workflow: normalise line endings, remove duplicates, fix case, then strip whitespace and special characters. - For recurring, sensitive or large-scale jobs, use a spreadsheet formula, script or native app instead of a browser tool. Most pasted text carries baggage—non-breaking spaces, smart quotes, inconsistent line breaks, zero-width joiners and duplicate blank lines. Free online tools can strip these in seconds, but only if you know what to look for and what the tool is actually doing with your text. Some cleaners upload your copy to a server; the ones we discuss run entirely in your browser. This guide explains when a browser-based cleaner is the right choice, why automatic title case mishandles short prepositions, brand names and non-English words, and how to build a quick cleanup workflow for lists bound for a CMS, spreadsheet or publication. We will also say when a spreadsheet formula is the safer bet. ### Automatic Title Case Is Not As Smart As It Looks Rule-based title case tools apply a simple pattern: capitalise the first and last word, plus every word that is not a short preposition, conjunction or article. That works for straightforward headlines, but it falls apart on brand names, technical acronyms, non-English phrases and words that contain internal capitals. The tool does not know whether iPhone is a proper noun or whether COVID-19 is an acronym; it only sees a string of letters. The result is text that looks edited but is quietly wrong. | the lord of the rings | The Lord Of The Rings | The Lord of the Rings | | macOS Ventura and iPhone | Macos Ventura And Iphone | macOS Ventura and iPhone | | la vie en rose | La Vie En Rose | La Vie en Rose | | CEO of IKEA in the UK | Ceo Of Ikea In The Uk | CEO of IKEA in the UK | | COVID-19 and eBay | Covid-19 And Ebay | COVID-19 and eBay | ### The Invisible Characters That Break a CMS Copying from a PDF, web page or word processor often imports characters that do not show up until something breaks. A non-breaking space can defeat a search-and-replace, a smart quote can turn into a rectangle on a cheap display, and a zero-width joiner can make two visually identical strings fail to match. These characters are valid Unicode, so your browser will render the page correctly while your database, CSV importer or URL slug generator chokes. | Non-breaking space (U+00A0) | Looks like a space but is not one | CSV imports, find/replace, code | | Smart quote (U+2018/2019) | Curly apostrophe or quotation mark | Code blocks, plain-text exports | | Zero-width joiner (U+200D) | Joins emoji or script components | URL slugs, string comparisons | | Zero-width space (U+200B) | Invisible word break | Tokenisers, database lookups | | Carriage return (U+000D) | Legacy line ending | Unix tools, single-line fields | ### A Five-Step Workflow for Cleaning a List If you are pasting a list of names, URLs or product titles into a spreadsheet or CMS, clean it first. Free online tools that run in the browser are ideal for this because the work is mechanical and the file never needs to leave your machine. The goal is not to make the text beautiful; it is to make it predictable. - Paste the raw text into a plain-text editor or a browser-based cleaner and inspect it for odd spacing. - Normalise line endings so every item sits on its own line. - Remove duplicate lines and empty lines, then sort if the order does not matter. - Apply case rules and manually fix brand names, acronyms and non-English words. - Strip leading and trailing whitespace, then convert any remaining special characters to their plain equivalents. ### When Free Online Tools Are the Wrong Choice Browser-based cleaners are excellent for one-off jobs, but they are not the right tool for every text problem. If you process the same list every week, you will save time with a repeatable method. If the data is sensitive, you should avoid any tool whose privacy policy allows server-side processing. And if the transformation is complex, a spreadsheet or a short script will be more reliable than a web form. - Recurring tasks: a spreadsheet formula or a script can apply the same cleanup every time the source changes. - Sensitive data: patient records, legal documents and financial figures should stay in tools you control. - Large files: browsers can slow down or freeze on files above a few hundred thousand lines. - Complex logic: conditional changes, lookups and validation are better handled in a database or scripting language. ### Sentence Case vs Title Case: Which to Use When The choice between sentence case and title case is a matter of house style, not grammar. Sentence case capitalises only the first word and proper nouns; title case capitalises most words. News sites and blogs often prefer sentence case for headlines because it is calmer and easier to read. Marketing copy, book titles and formal reports tend to prefer title case. The important thing is to pick one and apply it consistently. - Use sentence case for blog headlines, email subject lines and interface labels where speed of reading matters. - Use title case for book titles, report headings and formal documents where a traditional look is expected. - Never apply title case to URLs, file names or code variables; these should almost always be lowercase. - Always preserve the official capitalisation of brand names, even when it breaks your chosen style. ### What 'Runs in the Browser' Actually Means Many free online tools process your text on a remote server. That is fine for a shopping list, but it is a risk for client data, draft contracts or unpublished articles. Tools that run entirely in the browser perform the transformation with JavaScript on your own machine; nothing is uploaded. You can verify this by checking the network tab in your browser's developer tools: if no request is sent after you paste, the text is staying local. Even browser-based tools can be compromised by malicious updates, so treat highly sensitive text with the same caution you would apply to any third-party software. ### Quick Checks Before You Publish Before you commit a cleaned text to a CMS or spreadsheet, run through a short sanity checklist. It takes less than a minute and catches the errors that automated tools miss. Look for double spaces, inconsistent punctuation and words that the case converter has mangled. If you are importing into a spreadsheet, open the CSV in a text editor first to confirm that commas and quotes have not created new problems. - Search for double spaces and replace them with single spaces. - Check that apostrophes are straight or consistently curly, depending on your style. - Verify brand names and acronyms against their official spelling. - Confirm that line breaks appear where you expect them in a plain-text view. Q: Are free online text tools safe for confidential copy? A: It depends on where the processing happens. Tools that run entirely in your browser keep the text on your machine, which is safer than server-based tools that receive a copy of everything you paste. Even then, browser-based tools can be updated remotely, and a compromised or malicious site could exfiltrate data. For confidential client work, medical records, legal drafts or anything covered by GDPR, use software you own and control rather than a free online tool. Q: Why does automatic title case capitalise small words like ‘a’ and ‘the’? A: Most title case converters use a list of short words to leave lowercase, typically prepositions, conjunctions and articles under four or five letters. The trouble is that the same word can be different parts of speech: The Wind in the Willows keeps in lowercase because it is a preposition, but Log In to the System should capitalise In because it is part of a phrasal verb. A rule-based tool cannot tell the difference, so it either over-capitalises or under-capitalises. Q: How do I know if hidden characters are causing problems? A: Start by looking for symptoms: a search that should match does not, a CSV import splits a field in the wrong place, or a URL slug contains unexpected characters. Paste the suspect text into a tool that reveals Unicode code points, or use a plain-text editor that can show whitespace and control characters. Once you identify the culprit—often a non-breaking space, smart quote or zero-width joiner—you can search for that exact code point and replace it. ## URL Slugs: Write Web Addresses That Last with Free Online Tools https://siteb.biz/guides/url-slug-basics Reviewed 2026-08-06 · Text tools - A slug is a permanent identifier, not a cleaned-up headline, so write it for the long term. - Use lowercase letters and hyphens only, and avoid underscores, spaces, capital letters and dates. - Transliterate or drop accents based on your audience: German often keeps 'ue', while French and Spanish accents usually disappear. - Changing an indexed slug requires a 301 redirect, internal-link updates and weeks of monitoring, so get it right before you publish. - Free online tools are fine for drafting, but site migrations, rebrands and multilingual structures usually need a developer or SEO professional. A URL slug is the human-readable tail of a web address: the part after the domain and any folder path. In https://example.com/guides/url-slug-basics, the slug is url-slug-basics. It tells visitors and search engines what the page is about, and—unlike a query string—it is meant to stay fixed for the life of the page. That permanence is why slugs deserve more thought than most people give them. Plenty of free online tools will turn a headline into a hyphenated string in a second, and many of them do the job well. The trouble starts when you treat the output as a final decision. A slug is not a cleaned-up title; it is a contract with every future link, bookmark, social post and search result. Once a page is indexed, changing that string is possible, but it is rarely free. ### What a slug actually is (and what it is not) A slug is the final, path-level identifier for a single page. In a full URL such as https://siteb.biz/text-tools/url-slug-basics, the domain is siteb.biz, the folder path is /text-tools/, and the slug is url-slug-basics. It is not the page title, not the meta description, and not the headline that appears above the article. Some content management systems generate a slug automatically from the title, but the two fields serve different purposes: the title can change with a redesign or a seasonal campaign; the slug should not, because every external link points to it. | Protocol + domain | https://siteb.biz | Identifies the site | | Folder path | /text-tools/ | Groups related pages | | Slug | /url-slug-basics | Identifies one specific page | | Query string | ?id=123 | Passes temporary data; usually bad for permanence | ### Why a slug is not the title with hyphens A headline is written to catch attention in a feed; a slug is written to survive for years in a link. That difference explains why you should strip marketing language, year numbers and product version strings from slugs. A title like '10 Best Free Online Tools for 2024' becomes a slug such as 10-best-free-online-tools-for-2024, which looks fine today and broken in January 2025. The same applies to promotional phrases: 'Limited-Time Offer' ages badly, while 'offer' stays neutral. Keep the noun, lose the adjective. - A title can change every season; a slug should change almost never. - Marketing words such as 'ultimate', 'best' and 'free' date quickly and add no meaning. - Numbers are fine when they are part of the topic, not the calendar. - Hyphens are the only safe separator; underscores and spaces break in email and some social platforms. - Lowercase is safer because most servers treat URLs as case-sensitive. ### The transliteration question: accents, diacritics and non-Latin scripts Accented characters are valid in URLs, but they create two problems. Browsers usually display ü as ü while the underlying address is percent-encoded as %C3%BC, which is ugly in plain text and fragile when copied between systems. For German, the convention is to transliterate ü to ue, ä to ae and ö to oe, but the shorter 'u' form is acceptable in many international contexts. French and Spanish accents are more often dropped: résumé becomes resume, not reesume. For non-Latin scripts—Cyrillic, Arabic, Chinese—percent-encoding is unavoidable unless you use a Latin transliteration layer, which introduces its own inconsistencies. | Über | uber | Short and recognisable; German formal 'ueber' is also valid | | Café | cafe | Drop accents in French and Spanish loanwords | | München | muenchen | German 'ue' convention for umlauts | | 日本語 | nihongo | Latin transliteration avoids percent-encoded URLs | | naïve | naive | Diaeresis is usually dropped in English contexts | If your audience is almost entirely German and your site is German-language, 'ueber' is often the safer choice; for an international audience, 'uber' is easier to type. ### Stop words, length and dates in URLs Stop words—the, a, an, of, and—can usually be removed from slugs without losing meaning. 'Guide to URL Slugs' becomes url-slugs-guide or guide-url-slugs; both are readable. Length matters because long slugs truncate in search results, social previews and email clients. Aim for roughly 40 to 60 characters in the slug itself, including hyphens. Dates are the most common regret: /blog/2024/url-slug-basics locks the content to one year and forces a redirect when the guide is updated in 2026. If the date is part of the archive structure, keep it in the folder path, never in the slug. - Remove stop words unless the slug becomes ambiguous. - Keep slugs under 60 characters; very short slugs can be too vague. - Put dates in folder paths, not slugs, if you need them at all. - Avoid promotional seasons, version numbers and temporary qualifiers. - Read the slug aloud: if it sounds like a file name from 2003, rewrite it. ### The real cost of changing a slug Changing a slug after a page is indexed is not a typo fix; it is a migration. You must return a 301 redirect from the old URL to the new one, update every internal link, and wait for search engines to re-process the signal. During that window, rankings can slip, referral traffic can drop, and any bookmarks or backlinks that miss the redirect become 404 errors. The work is straightforward on a small site and expensive on a large one: a redirect audit from an SEO consultant typically runs €400–€1,200 depending on site size, while a developer implementing the redirects may charge €150–€600 depending on the CMS and the number of rules. | 301 permanent | Passes most link equity; preferred for slug changes | The page has moved permanently | | 302 temporary | Does not pass full equity; confuses long-term indexing | A temporary test or maintenance page | | Meta refresh / JS redirect | Unreliable; search engines may ignore | Avoid for structural changes | | No redirect | Old URL becomes 404; equity is lost | Only if the page never had traffic or links | ### A five-year checklist for a durable slug Before you publish, imagine someone reading the URL in five years. Will the topic still be accurate? Will the year in the slug feel stale? Will the file still exist if the author leaves? A durable slug is descriptive, language-appropriate, and free of anything that ties it to a campaign or a quarter. It should be easy to type from memory and safe to paste into an email without breaking. The best slugs look obvious in retrospect and survive redesigns because they name the thing, not the moment. - Use the core noun or phrase the page is actually about. - Strip years, seasons, version numbers and campaign slogans. - Transliterate or drop accents based on your primary audience. - Keep it under 60 characters and remove stop words where meaning stays clear. - Check that the slug is unique across your site and does not clash with a folder name. - Confirm you can live with it for five years before you click publish. ### When a free online tool is the right choice (and when it is not) Browser-based slug generators are useful for the first pass: paste a title, get a hyphenated string, copy it into your CMS. The five tools we publish on siteb.biz run entirely in your browser, so nothing leaves your machine, and the independent mini projects we list each sit on their own subdomain with their own privacy policies. Still, a slug generator cannot know your content strategy, your existing URL patterns or your redirect history. For a one-off blog post, a free online tool is enough. For a site migration, a rebrand or a multilingual setup, hire a developer or an SEO professional; the cost of getting it wrong is usually higher than the cost of the advice. Always read the privacy policy of any slug generator that asks you to sign in or that processes text on a server; if it is not clear, do not use it. ### What to do when you must change a slug If a slug is genuinely wrong—misleading, offensive, or blocking a restructure—change it, but change it once. Map the old URL to the new one with a 301 redirect at the server level, not a plugin-only redirect that can be disabled by accident. Update every internal link, including navigation, footer links and image srcset references if they contain the URL. Submit the changed URL through your search console and monitor 404 logs for at least 30 days. Keep the redirect live indefinitely; removing it after a year is the most common way to recreate the original problem. - Audit existing internal links and backlinks before you change the slug. - Set a server-level 301 redirect from the old URL to the new one. - Update internal links so they point directly to the new URL instead of relying on the redirect. - Submit the new URL to search engines and request re-indexing. - Monitor 404 logs and backlink reports for at least 30 days. - Keep the redirect in place permanently; do not treat it as temporary cleanup. Q: Should I include keywords in my URL slug for SEO? A: Including a relevant word or phrase can help a little, because search engines still use the URL as a signal and because users scan addresses before they click. The benefit is modest, however, and it is easily outweighed by keyword stuffing or by a slug that becomes awkward when the topic shifts. A slug such as /url-slug-basics is clearer and more durable than /seo-url-slug-best-practices-guide-2024. Write the slug for humans first, make sure it describes the page accurately, and let the title, headings and body copy carry the heavier SEO load. A readable slug also earns more trustworthy clicks over time. Q: Can I use capital letters or underscores in a slug? A: Technically yes, but practically no. Most web servers treat URLs as case-sensitive, so /URL-Slug-Basics and /url-slug-basics can be two different pages, which creates duplicate-content risk and broken links when someone types the address by hand. Underscores are even worse: some social platforms and email clients do not underline or wrap them cleanly, and search engines have long treated hyphens as word separators while treating underscores as part of a word. Use lowercase letters and hyphens only. The result is easier to read, harder to break and more consistent across phones, desktops and printed materials. Q: How do I fix a slug that is already indexed by Google? A: First, decide whether the problem is serious enough to justify a migration. A slightly awkward slug that ranks well is usually better left alone. If the slug is misleading, offensive or structurally wrong, change it once and set a 301 redirect from the old URL to the new one. Update every internal link, request re-indexing through your search console, and monitor 404 reports for at least 30 days. Expect a temporary dip in rankings while search engines reassign signals. Keep the redirect active forever; removing it is the easiest way to resurrect the original error. ## Why Free Online Tools Give Different Word Counts for the Same Text https://siteb.biz/guides/word-count-guide Reviewed 2026-08-06 · Text tools - Word counting has no universal standard, so the same text will produce different numbers in different tools. - The count that matters is the one used by the system paying you or publishing your text. - Character limits for SEO, social media, SMS, and databases each measure different things. - CJK text and emoji expose the difference between visible characters, code points, and bytes. - For confidential or high-stakes work, use a browser-only or native counter and document which rule you agreed on. Every writer has a story about the invoice that came up short. You paste a finished article into the client’s free online tools and the total is three per cent lower than your own count. A 1,200-word piece becomes 1,164 words, and at typical web-copy rates of €0.08–€0.40 per word, that gap is real money. The same thing happens in reverse when a form rejects your bio because its counter sees five characters you cannot find. It is not a bug, and it is rarely dishonesty. Word and character counting is a set of arbitrary conventions, and every application—Word, Google Docs, a browser-based counter, a CMS field—applies a slightly different rulebook. This guide explains why the numbers diverge, where the limits that matter actually live, and how to choose the count you should be paid by, or the one that will let your post through. ### The same paragraph, three different numbers To see the problem in miniature, paste the same short paragraph into three free online tools or word processors. The text below is only one sentence long, yet the word count changes depending on whether hyphens, contractions, and punctuation are treated as word boundaries. Microsoft Word and most word processors keep hyphenated compounds and contractions intact; a regex-based browser counter may split them. The character count is more stable, but even it shifts if trailing spaces or line breaks are included. | Microsoft Word | 14 | 100 | | Browser counter A (splits on hyphens) | 19 | 100 | | Browser counter B (splits on hyphens and apostrophes) | 20 | 100 | The exact numbers depend on the tool version and settings; treat them as illustrations, not specifications. ### Why the numbers diverge There is no ISO standard for what counts as a word. A word processor usually treats a word as a string of letters, digits, or connectors such as apostrophes and hyphens, bounded by spaces or punctuation. A web form may simply split on whitespace. A programming script may use a regular expression that splits on anything that is not a letter or number. Because the source text is identical, the disagreement looks like a mistake; in fact, each tool is answering a different question. - Hyphens: some tools treat 'start-up' as one word; others count 'start' and 'up' separately. - Contractions and possessives: 'it's' may be one word or two, and 'client's' may keep or lose the apostrophe-s. - Numbers and currency: '€50' can count as one token or split into '€' and '50'. - Punctuation attached to words: trailing commas, full stops, and quotation marks may be stripped or counted as separators. - Line breaks and multiple spaces: a double space can create an empty word in some counters. - Non-Latin scripts: CJK text has no spaces, so whitespace-based counting fails entirely. ### The character limits people actually hit Character limits are usually hard stops: a form simply refuses the extra character, or a search engine truncates your snippet. The limit that matters is the one enforced by the system that will publish the text, not the one shown by the tool you drafted in. If you are writing for a CMS, test in the CMS; if you are writing a meta description, check the pixel width as well as the character count, because Google truncates by pixels, not characters. | Google title tag | ~60 characters | Characters and spaces; pixel width is the real limit | | Google meta description | ~155–160 characters | Characters and spaces; mobile often truncates earlier | | X post | 280 characters | Spaces, punctuation, and URLs counted as shortened t.co links | | Bluesky post | 300 characters | All characters including spaces | | SMS (GSM-7 alphabet) | 160 characters per message | Letters, digits, and a few symbols; longer texts split into parts | | SMS (UCS-2 encoding) | 70 characters per message | Non-Latin scripts and some emoji; longer texts split into parts | | Database VARCHAR(255) | 255 characters | Characters in the declared encoding; emojis may use multiple bytes | A limit is a contract; draft in whatever tool you like, but validate in the final system. ### How CJK and emoji break naive counting Naive counters usually return the length of the string in bytes, Unicode code points, or grapheme clusters. Those three numbers can be very different. A single Chinese character is one user-perceived character and one code point, but it may be three bytes in UTF-8. An emoji with a skin-tone modifier is one visible glyph, two code points, and four bytes. A family emoji can be seven code points and one glyph. If your counter is counting bytes, your Chinese novel will look enormous. | Latin 'A' | 1 | 1 | 1 | | Chinese '中' | 1 | 1 | 3 | | Emoji thumbs up | 1 | 1 | 4 | | Emoji thumbs up with skin tone | 1 | 2 | 8 | | Flag emoji | 1 | 2 | 8 | | Family emoji | 1 | 7 | 25 | For user-facing limits, count grapheme clusters; for storage, count bytes. ### Which number should you argue from? If you are invoicing by the word, the only number that matters is the one in your contract or the client's brief. If the brief says 'Microsoft Word count', use Word. If it says 'per source word', count the source text in the agreed tool. For ad-hoc work, agree on a sample before you start. Freelance rates for web copy typically run €0.08–€0.40 per word, with the lower end for bulk blog posts and the upper end for technical or rushed work; the counting method should be specified alongside the rate, because a 5 per cent difference is not noise at those prices. - One-off web check: a browser-based counter is fine. - Bulk invoices or client reports: a spreadsheet with a consistent formula, or a native app like Word's built-in counter, is better. - Legal, medical, or financial documents: hire a professional or use the tool mandated by the regulator; never rely on a free online tool for compliance. - Programming and database limits: use the language's string-length function in the target encoding. - Social posts: draft anywhere, but paste into the platform for the final count. This is guidance, not legal or financial advice; put the counting method in your contract. ### Privacy: the tools that quietly upload your text Many free online tools are not calculators on your device; they are front ends for a cloud service. Paste your text and it travels to a server for analysis, storage, or AI training. That is fine for a public blog post, but it is a poor choice for an unpublished manuscript, a client brief under NDA, or a draft containing personal data. Browser-only tools—such as the five counters siteb.biz publishes—run entirely in your browser and never upload the text, though you should still check the network tab if you are cautious. - The site asks you to sign in to count words. - The counter offers grammar checking, summarising, or translation—features that require understanding the text. - Pasting triggers network requests to analytics or AI endpoints. - The privacy policy says content may be used to improve services. - There is no privacy policy at all. If in doubt, use a native app or a spreadsheet. ### How to choose a counter you can trust A good counter tells you what it is counting. Look for a clear label: 'words separated by whitespace', 'Unicode code points', or 'grapheme clusters'. Test it with a known string containing a hyphen, a contraction, an emoji, and a Chinese character. If the result matches your expectations, the tool is probably suitable for that job. Remember that no single tool is right for every job; the right tool is the one that matches the limit you are trying to satisfy. - Identify the limit: word count for an invoice, characters for a form, bytes for a database. - Find the system that enforces the limit and use its count as the reference. - Choose a tool that explains its counting rule, or verify it with a test string. - For sensitive text, prefer a browser-only or native app counter. - Keep a screenshot or saved copy of the count if money or compliance is at stake. Q: Why does Microsoft Word give a different word count from a free online tool? A: Microsoft Word uses an internal word-breaking algorithm that keeps hyphenated compounds, contractions, and numbers together when they are not separated by spaces. A browser-based counter often uses a simple regular expression that treats hyphens and apostrophes as boundaries, so 'start-up' becomes two words and 'it's' becomes two. Some tools also split punctuation differently. Neither result is wrong; they are simply following different definitions of a word. The practical fix is to agree on the tool and the counting rule before any money changes hands, and to record that agreement in your contract or brief. Q: Are browser-based word counters safe for confidential documents? A: It depends on whether the tool runs entirely in your browser or uploads text to a server. Browser-only counters process the text locally and never transmit it, which makes them a reasonable choice for drafts you would not want to share publicly. However, many free online tools that also offer grammar checking, summarising, or AI features send your text to the cloud. For NDAs, patient records, financial drafts, or any material covered by data-protection rules, use a native app, an offline spreadsheet, or the tool mandated by your organisation rather than an unknown website. Q: How should I count text that includes Chinese, Japanese, Korean, or emoji? A: Use a counter that returns grapheme clusters, not bytes or raw code points. A single Chinese character is one grapheme, but three bytes in UTF-8. An emoji with a skin-tone modifier is one visible character, yet two code points. A flag or family emoji can be one visible glyph made of several code points joined by zero-width joiners. For user-facing limits, grapheme clusters are the fair measure because they reflect what a reader sees. For database storage and programming, byte length or code-point length is what matters, because that is what the system must allocate. ## Converting Files Online: What Free Online Tools Lose https://siteb.biz/guides/file-format-conversion-guide Reviewed 2026-08-05 · Converters - Free online tools are convenient, but every conversion trades something away. - Lossy formats like JPEG and MP3 permanently discard data; changing the wrapper does not bring it back. - PDF to Word conversions often break fonts, spacing, forms, and annotations. - Uploaded files may be stored, logged, or used under terms you did not read. - For sensitive or precise work, use a local application and keep your originals untouched. Free online tools make file conversion look effortless: drag, drop, done. Behind that simplicity is a chain of decisions about compression, encoding, and format limits that can quietly degrade a photograph, scramble a layout, or leave a copy of your document on a server you have never heard of. This guide explains what actually happens when you convert common file types in the browser. It is written for anyone who has hesitated before uploading a contract, a medical scan, or a master image to a free converter and wondered whether the result would be private, accurate, or worth the zero-euro price tag. Nothing here is legal, financial, or medical advice; it is simply a practical map of what each conversion loses. ### Lossy versus lossless: the distinction that decides quality The first thing to understand about any file conversion—whether you use a desktop program or one of the many free online tools—is whether the format is lossy or lossless. Lossless formats keep every original bit, so converting between them only changes the container. Lossy formats discard information to save space, and once that detail is gone it cannot be reconstructed by saving the file under a different extension. This guide is written from the perspective of an independent directory that has tested dozens of these tools, not from a vendor trying to sell you one. - Lossless: PNG, TIFF, FLAC, ALAC, ZIP – every pixel, sample, or byte is preserved. - Lossy: JPEG, MP3, AAC, HEIC, H.264 – data is thrown away according to perceptual models. - Re-wrapping: RAW to DNG, MKV to MP4 – may change metadata but usually keep the essence. - Flattening: layered PSD to JPEG – layers, masks, and adjustment layers are merged. - Metadata stripping: EXIF, geotags, edit history, and accessibility tags often disappear. - Re-compression: opening and re-saving a lossy file adds new artifacts on top of old ones. ### Common conversions and what each one loses Most browser converters handle the same dozen conversions, from image resizes to document exports. The table below shows what usually survives and what does not, assuming a typical consumer-grade tool rather than a specialised professional pipeline. Use it as a quick reference, but remember that individual sites vary: some use better OCR engines, some preserve more metadata, and some quietly downgrade colour depth or strip security settings to keep processing fast. | JPEG → PNG | The blocky, compressed image is copied exactly; no new artifacts are added. | The quality lost during earlier JPEG saves is gone forever; the file becomes larger. | | PNG → JPEG | The overall picture and colours remain recognisable. | Sharp edges, transparency, exact colours, and clean text become blurry or artefact-ridden. | | PDF → Word | Text flow and embedded images usually come through. | Precise fonts, spacing, forms, annotations, and page geometry are often distorted. | | Word → PDF | The final visual layout is preserved. | Editable layers, comments, revision history, and live hyperlinks may be lost unless explicitly retained. | | MP3 → WAV | The audio plays back without further compression. | No quality is recovered; the file grows because silence and discarded data are encoded as zeroes. | | HEIC → JPEG | A viewable image is produced. | 10-bit colour, depth maps, burst frames, and the smaller-file-size advantage are lost. | ### Why JPEG to PNG cannot restore quality Converting a JPEG to PNG is one of the most misunderstood operations. PNG is a lossless format, so the new file stores every bit of the image it receives. The problem is that the image it receives has already been mauled by JPEG compression. JPEG works in 8×8 pixel blocks and throws away fine detail according to a quality setting. When you save that damaged image as PNG, you lock in the blockiness and colour banding permanently. The PNG file is larger not because it contains more quality, but because it treats compression artifacts as real detail. - JPEG compression splits the image into 8×8 blocks and discards high-frequency detail. - A PNG can only preserve what it is given; it cannot reconstruct the discarded detail. - The resulting PNG is often two to five times larger than the JPEG with no visible improvement. - Repeated JPEG saves compound the damage; each generation adds new artifacts. - If you need a clean master, return to the original RAW, TIFF, or PSD source. ### PDF to Word: where the formatting goes PDF and Word are fundamentally different creatures. PDF is a fixed-layout page description: every glyph, line, and image is positioned for print or display. Word is a reflowable document built around paragraphs, styles, and editable objects. Converting PDF to Word therefore asks the tool to reverse-engineer structure from appearance. The result is usually usable for text-heavy pages and a mess for complex layouts, forms, or scanned archives. | Text with embedded fonts | Font substitution and small spacing shifts are common. | | Vector graphics and diagrams | Often turned into images or grouped shapes; text may become uneditable. | | Interactive forms | Flattened to static text; fields and JavaScript are lost. | | Annotations and comments | Usually stripped unless the converter specifically preserves them. | | Tables drawn as lines | Reconstructed as Word tables, frequently with misaligned rows or columns. | | Scanned pages | Require OCR; accuracy is typically 95–99% on clean printed text, lower on handwriting or faded pages. | ### The privacy question: what converter sites do with your file Free converter sites live on advertising, premium upgrades, or data harvesting. Their privacy policies often grant the site a broad licence to process your file, and many keep copies for hours or days for 'quality assurance' or machine-learning training. Even if the operator is reputable, the file usually passes through a cloud server, a content-delivery network, or a subcontractor in another legal jurisdiction. That means a document you upload for a ten-second conversion can outlive your visit on several systems you did not choose. - A licence to process, and sometimes train models on, the content you upload. - Retention periods from immediate deletion up to 24 hours, 30 days, or indefinite 'as necessary'. - Logging of IP address, browser fingerprint, file size, and conversion settings. - Third-party processing by cloud providers under different data-protection rules. - Public or shareable download links that remain active until you manually delete them. Before uploading anything you would not print on a bus stop poster, read the retention and third-party clauses in the privacy policy. ### When a local application is the better choice Browser-based free online tools are fine for quick, non-sensitive jobs, but they are the wrong tool for anything that matters. If the file contains personal data, trade secrets, legal pleadings, or medical images, process it on hardware you control. Local applications range from free suites such as LibreOffice to professional tools costing €50–€300 for a perpetual licence or €10–€30 per month for a subscription, with the spread reflecting features like OCR, colour management, batch processing, and PDF form editing. - Confidential documents: use LibreOffice, Microsoft Office, Adobe Acrobat, or a local OCR engine. - Batch or automated workflows: command-line tools like ImageMagick, ffmpeg, or Pandoc run on your machine. - Precise colour and print work: Affinity Photo, Photoshop, or Capture One keep profiles intact (€50–€300). - Complex PDFs with forms, layers, or digital signatures: Adobe Acrobat or PDF-XChange Editor (€0–€240 per year). - Archival or regulatory needs: maintain originals, audit trails, and backups on storage you control. ### How to pick a converter you can trust If you still need a browser converter, treat it like a public photocopier. Choose a tool that states clearly that files are processed locally and never uploaded, or one with a short, readable privacy policy. Test with a harmless file first, inspect the output for missing pages, shifted margins, or colour shifts, and never overwrite your original. Remember that 'free' often means you are paying with attention, data, or both. - Prefer tools that run entirely in the browser and explicitly say files are not uploaded. - Read the privacy policy for retention periods, third-party processors, and jurisdiction. - Check that the output format supports the features you need: transparency, layers, fonts, forms. - Test with a non-sensitive file and compare the result side by side with the original. - Keep the original untouched; save converted copies under a new name. Q: Is converting a JPEG to PNG lossless? A: PNG compression is lossless in itself, but converting a JPEG to PNG is not. A JPEG has already discarded detail using 8×8 block compression and a quality setting, and that lost information is gone for good. When you save that JPEG as a PNG, the new file stores every remaining block and artifact exactly, so the image does not improve and the file size often grows. The PNG simply preserves the damage. The only way to obtain a higher-quality version is to go back to an uncompressed original such as a RAW, TIFF, or PSD file and export from there. Q: Can I safely upload a contract to a free online converter? A: Usually no. Most free converter sites ask you to grant them a licence to process the file, and their terms often permit retention for hours or days, use for quality assurance, or inclusion in machine-learning datasets. The file may also pass through third-party cloud servers in jurisdictions with weaker data-protection rules. Even if the operator is trustworthy, a security breach or legal request could expose your document. For contracts, medical records, or financial statements, use local software on a machine you control and keep the original untouched. This is not legal advice. Q: Why does my PDF look different after converting to Word? A: PDF is a fixed-layout format: every character, line, and image is positioned precisely on the page. Word is a reflowable format built around paragraphs, styles, and editable objects. A converter has to guess how the PDF's visual structure maps onto Word's logical structure, so fonts get substituted, spacing shifts, vector graphics become images, and interactive forms flatten into static text. Scanned pages add OCR errors on top. If you only need small edits, a dedicated PDF editor is usually safer; if you need heavy editing, recreating the document in Word may be faster than cleaning up a conversion. ## Metric vs Imperial: A Practical Guide for Users of Free Online Tools https://siteb.biz/guides/metric-vs-imperial-explained Reviewed 2026-08-05 · Converters - US and imperial units often share names but differ in size by up to twenty per cent. - The Mars Climate Orbiter was destroyed because pound-seconds were read as newton-seconds. - Free online tools convert accurately only when you select the correct source system. - If a source does not state its unit system, convert both possibilities and check against a physical limit. - For legal, financial, engineering or medical work, use documented tools and professional advice, not a browser converter. A ‘gallon’ in Boston is not the same as a ‘gallon’ in Birmingham. The names match, but the volumes differ by twenty per cent, and the same trap waits inside pints, fluid ounces and tons. Free online tools convert these in seconds, yet they cannot tell you which system the original figure used. That is the user's job, and it is where costly errors start. This guide compares metric and imperial units that share a name but not a size. It includes a quick-reference table, explains the Mars Climate Orbiter as the textbook cost of mixing systems, and gives a rule for what to do when a source does not say whether it means US customary, imperial or metric units. ### The Name Game: Why Shared Labels Hide Different Sizes The metric system is not the danger. The danger is the set of imperial and US customary units that borrowed the same words and then quietly changed the quantities. A US fluid ounce is roughly four per cent larger than an imperial fluid ounce, yet a US pint is twenty per cent smaller than an imperial pint because the US uses sixteen fluid ounces per pint while the UK uses twenty. Free online tools will convert any of these accurately once you pick the right system, but they assume you already know which system you are looking at. Many people do not. ### Same Name, Different Size: A Quick-Reference Table | Gallon | 3.785 litres | 4.546 litres (imperial) | | Quart | 0.946 litres | 1.136 litres (imperial) | | Pint | 0.473 litres | 0.568 litres (imperial) | | Fluid ounce | 29.57 ml | 28.41 ml (imperial) | | Ton | 907.18 kg (short ton) | 1,016 kg (long ton) / 1,000 kg (tonne) | | Hundredweight | 45.36 kg (short cwt) | 50.80 kg (long cwt) | | Cup | 240 ml (legal US cup) | 250 ml (metric cup, Australia) | The differences are not rounding errors; they are separate definitions that happen to share a name. ### Liquid Measures: The US Gallon Is Not the Imperial Gallon Most recipe disasters and fuel-economy confusions come from liquid units. The US gallon is defined as 231 cubic inches, while the imperial gallon is defined as the volume of ten pounds of water at a specified temperature. That difference makes the imperial gallon about 1.2 times the US gallon. A British pint of beer is therefore larger than an American pint, but a British fluid ounce is smaller than an American fluid ounce. The arithmetic is consistent once you know the starting point; the problem is that labels rarely say which starting point they used. - A US pint is 473 ml; an imperial pint is 568 ml. - A US fluid ounce is 29.6 ml; an imperial fluid ounce is 28.4 ml. - US recipes usually measure by volume, not weight, so a ‘cup’ of flour can vary by density. - Fuel economy in miles per gallon cannot be compared across the Atlantic unless you know which gallon. - Paint, resin and adhesive data sheets may list coverage per gallon without stating the system. ### Weight: Short Ton, Long Ton and Tonne The word ‘ton’ is even more slippery. The US short ton is 2,000 pounds, the British long ton is 2,240 pounds, and the metric tonne is 1,000 kilograms. A shipping quote that ignores the difference can be ten per cent out on freight costs or customs duty. The hundredweight follows the same split: a US hundredweight is 100 pounds, while an imperial hundredweight is 112 pounds, which is why an imperial ton contains twenty hundredweight and a US ton contains twenty. Free online tools handle these conversions, but only if you select the correct ton before pressing convert. | Ton | 907.18 kg (2,000 lb) | 1,016 kg (2,240 lb) / 1,000 kg (tonne) | | Hundredweight | 45.36 kg (100 lb) | 50.80 kg (112 lb) | | Pound | 0.4536 kg | 0.4536 kg (same) | | Ounce (mass) | 28.35 g | 28.35 g (same) | The avoirdupois pound and ounce are identical in both systems; only the ton and hundredweight diverge. ### The Mars Climate Orbiter: What Mixing Systems Cost The most famous price tag for a unit mix-up is NASA's Mars Climate Orbiter. In 1999, one team sent thrust data in pound-seconds, while another team read it as newton-seconds. The ratio between those units is roughly 4.45, so the spacecraft's trajectory was off by that factor. The probe entered the Martian atmosphere too low and was destroyed. The mission cost around $125 million, or roughly €115–€135 million at contemporary exchange rates. It is the example every engineering course uses because the failure was not a calculation error; it was a metadata error. The numbers were right, but the label on the numbers was wrong. The same failure mode still appears today in spreadsheets, CAD files and shipping manifests. ### A Rule for Sources That Do Not Say Which System They Used When a source gives a number but not a system, the safest assumption is not the one closest to your own country. The safest assumption is that you do not know. Start with the context: a 1990s British cookbook will use imperial pints, an American brewery will use US pints, and a European technical drawing will use metric. If the source is ambiguous, convert both possibilities and compare them against a physical limit. A twenty-litre result and a twenty-gallon result cannot both fit in the same tank. - Look for a country, date or standard reference on the source. - If none appears, convert both US customary and imperial versions. - Check the result against a physical limit, cost or known weight. - Record which system you assumed, so the next person can check it. - For contracts, engineering or medicine, insist on written confirmation. ### When Free Online Tools Help, and When They Do Not Browser-based converters are excellent for one-off checks when the source system is known. They are fast, need no installation and leave no data on a server if they run entirely client-side. The five tools on siteb.biz, for example, process everything in the browser. However, a free online tool is the wrong choice for repeated business calculations, regulated work or any figure where a mistake could cost money or safety. In those cases, a spreadsheet with locked cells, a native unit-conversion app or a qualified professional is the better investment. The tool converts; the user still owns the judgement. - Use a browser converter for quick recipe scaling, travel fuel checks or luggage weight estimates. - Use a spreadsheet with documented formulas for repeated quotes, invoices or inventory. - Use a native app when you are offline or need audit trails. - Consult a professional for customs declarations, engineering specs or medical dosing. - Never treat a converted figure as legal, financial or medical advice on its own. ### Final Check: Treat the Unit as Part of the Number The only reliable way to avoid these traps is to treat the unit as part of the number, not as decoration. Write ‘5 US gal’ or ‘5 imp gal’ rather than ‘5 gal’. Store the unit beside the value in every spreadsheet cell. Ask the supplier, the author or the lab which system they used. Free online tools will then give you a clean conversion; without that label, they can only give you a guess. Q: Is a US pint the same as a UK pint? A: No. A US pint is 473 millilitres, while an imperial pint is 568 millilitres, so the UK pint is roughly twenty per cent larger. The confusion exists because both systems use sixteen fluid ounces in some contexts, but the US fluid ounce is 29.6 millilitres and the imperial fluid ounce is 28.4 millilitres, while the US pint contains sixteen US fluid ounces and the imperial pint contains twenty imperial fluid ounces. When you see ‘pint’ on a label, recipe or data sheet, you need to know the country of origin before converting. Free online tools can do the arithmetic once you select the correct system, but they cannot guess which pint the original author meant. Q: Why did the Mars Climate Orbiter fail? A: The orbiter failed because two engineering teams used different force units without realising it. Lockheed Martin provided thruster performance in pound-seconds, while NASA's navigation team expected newton-seconds. A pound-second is about 4.45 newton-seconds, so the calculated trajectory was wrong by that factor. The spacecraft approached Mars at the wrong angle and burned up in the atmosphere. The direct mission cost was about $125 million, or roughly €115–€135 million at the exchange rates of the late 1990s. The lesson is that the numbers themselves were calculated correctly; the disaster came from the missing label that should have identified the unit system. Q: When should I avoid free online unit converters? A: Avoid browser-based converters for any calculation where a mistake could have legal, financial, safety or medical consequences. They are fine for quick personal checks, but they offer no audit trail, no guarantee of accuracy and no professional accountability. For repeated business use, a spreadsheet with locked formulas or a certified native app is more reliable. For customs, engineering, aviation or medical dosing, use the system mandated by the regulator and confirm the unit system in writing. Free online tools are a convenience, not a substitute for professional judgement, documented processes or advice from a qualified expert. ## Unit Conversion: A Practical Guide to Free Online Tools https://siteb.biz/guides/unit-conversion-guide Reviewed 2026-08-05 · Converters - A conversion factor is a ratio that equals one; it re-labels a quantity without changing it. - Modern definitions such as 1 inch = 25.4 mm and 1 lb = 0.45359237 kg are exact, so errors usually come from rounding or using the wrong standard. - Rounding early or using a truncated factor compounds quickly, especially on large quantities. - Temperature, fuel economy and currency are not simple scaling problems; they need offsets, inverse ratios or live rates. - For engineering tolerances, medication doses, legal filings and financial contracts, use an official source or professional advice, not a generic converter. Most people looking for a free online tool for unit conversion want the same thing: paste a number, choose the labels, and copy the result. That works for an enormous range of everyday jobs—cooking, travel, parcel weights, screen sizes—but it only works because someone, somewhere, has already fixed the exact relationship between the two units. A conversion is not a guess; it is a multiplication by a defined factor. The danger is that the interface rarely tells you whether the factor is exact, rounded, or appropriate for your purpose. The best free online tools make the arithmetic transparent and keep the calculation inside your browser. This guide explains what a conversion factor is, where rounding introduces real error, and why temperature, fuel economy and currency are not like other conversions. It also flags the jobs—engineering tolerances, medication doses and legal filings—where an online converter is the wrong tool entirely. ### What a conversion factor actually is A conversion factor is a ratio that equals one, expressed in different units. Because multiplying by one changes the label but not the quantity, you can convert 12 inches to 304.8 millimetres by multiplying 12 by 25.4 millimetres per inch. The result is exact because the inch is defined as 25.4 millimetres exactly. Good converters store these definitions; weaker ones store rounded approximations and pass the error on to you. The factor is the entire logic of the conversion, so understanding it matters more than trusting the design of the page. ### Exact definitions you can rely on Modern metrology fixes many common units to exact SI values. That means the conversion is not approximate at the source; any error is introduced later by rounding or by using the wrong national standard. | 1 inch | 25.4 millimetres | Defined exactly since 1959; all inch-derived units rest on it | | 1 pound (avoirdupois) | 0.45359237 kilograms | Exact since 1959; use for mass, not weight in physics | | 1 mile | 1609.344 metres | Defined from the exact yard of 0.9144 metres | | 1 yard | 0.9144 metres | Exact since 1959; three feet equal one yard exactly | | 1 litre | 1 cubic decimetre | Exact since 1964; tied to the metre even though it is not an SI unit | If a converter shows 1 inch as 25.44 mm, it is wrong, not approximate. ### Where rounding destroys a result Rounding is harmless once, but it compounds. Converting 10.00 kg to pounds using a factor rounded to 2.20 gives 22.0 lb, close enough for luggage. Using the same rounded factor to calculate a shipping load of 10 000 kg gives 22 000 lb, an error of about 4.6 kg—possibly the difference between a courier band and a freight charge. For repeated conversions, keep the full precision until the final step and round only the answer. | Exact (0.45359237) | 22.046226 | 22 046.226 | 0 | | 2.20462 (5 digits) | 22.0462 | 22 046.2 | ≈ 0.01 | | 2.20 (3 digits) | 22.0 | 22 000 | ≈ 4.6 | | 2.2 (2 digits) | 22 | 22 000 | ≈ 46 | Round the final answer to the precision of your least precise measurement, not the tool’s default. ### Temperature: the conversion you cannot multiply Celsius and Fahrenheit are not different-sized versions of the same zero. Water freezes at 0 °C and 32 °F, and boils at 100 °C and 212 °F. The scales have different zero points and different step sizes, so the formula is F = (C × 9/5) + 32. A temperature interval of 10 °C equals 18 °F, but a temperature of 10 °C equals 50 °F. Many casual converters get this right; the risk is using them for scientific temperature differences or thermal calculations where you need kelvin and absolute intervals. For temperature differences, use kelvin or rankine; 10 °C of change equals 10 K, not 18 °F as a temperature reading. ### Fuel economy and currency: conversions of a different kind Fuel economy is not a simple length-over-volume ratio because the comparison is inverted between regions. Miles per gallon and litres per 100 kilometres move in opposite directions: higher mpg means lower consumption, while a lower L/100 km figure means lower consumption. Currency is even more volatile; the rate changes every second and carries fees. For both, a browser converter can give a snapshot, but a spreadsheet with a live feed is better for repeated work. - UK mpg to L/100 km: divide 282.481 by the mpg figure. - US mpg to L/100 km: divide 235.215 by the mpg figure. - Higher mpg means lower consumption; lower L/100 km means lower consumption. - Currency conversions need a timestamp and the name of the rate provider; without both the number is already stale. - For invoices or tax, use the rate your tax authority publishes for that date. ### When a free online tool is the wrong choice Browser-based converters are excellent for quick, low-stakes checks. They are not appropriate when the result affects safety, money that is being disputed, or a part that must fit. In those cases you need traceability: a documented method, a known uncertainty, and someone who can explain it. - Engineering tolerances: a shaft specified at 25.00 mm ±0.01 mm is not safely checked by pasting 0.98425 inches into a generic converter. - Medication doses: mg, mL, IU and percentage concentrations each carry clinical risk; a converter has no idea about the formulation. - Legal filings and contracts: courts and tax offices care about the source of the conversion factor and the rounding convention, not the number on your screen. - Recipe scaling across countries: cups and tablespoons differ by country; a volume converter cannot tell you which standard the author used. - Precious metals and gemstones: troy ounces, carats and pennyweights have historical definitions that many generic tools mishandle. ### How to pick a converter that respects your data The safest free online tools run entirely in your browser. That means no upload, no account, and no server-side processing. You can check this by opening the tool, disconnecting from the internet, and trying the conversion. If it still works, your data never left your device. - Look for a statement that calculations happen locally; avoid tools that ask you to upload a file or create an account. - Prefer converters that show the factor they used, or at least let you copy a result with plenty of decimal places. - Test the tool with a known exact value: 1 inch should equal exactly 25.4 mm, not 25.44 mm. - Be wary of pages covered in ads for unrelated products; they are often scraped content with outdated factors. - For currency, use the website of a central bank or a reputable financial data provider, not a generic converter. ### Better alternatives for serious work For one-off everyday conversions, a browser tool is fine. For repeated business calculations, a spreadsheet lets you document the factor, lock the formula, and audit the result. For engineering, use calibrated measurement software or a dimensional metrology app. For medication, ask a pharmacist or use the specific calculator supplied by your health service. For legal or tax matters, use the official conversion tables published by the relevant authority. The right tool is the one that leaves a paper trail. Q: Are online unit converters accurate? A: Many are accurate for everyday tasks because modern definitions such as 1 inch = 25.4 mm and 1 pound = 0.45359237 kg are exact. The problem is usually context, not arithmetic: a generic tool does not know your required precision, the standard you are working to, or whether you are converting a temperature or a temperature interval. It also cannot judge whether the result will be used for a recipe, an engineering drawing or a medicine dose. Treat the result as a starting point, keep full precision through intermediate steps, and round only the final answer to match the precision of your original measurement. Q: Why does the same conversion give different results on different websites? A: Differences usually come from rounding, outdated definitions or confusion between similar units. Some sites round the conversion factor itself, which magnifies errors when the input is large. Others mix US gallons with imperial gallons, troy ounces with avoirdupois ounces, or short tons with metric tonnes. A few pages scrape data from older sources and never update it. If two sites disagree, check which standard each claims to use and compare it with an official source such as NIST in the United States or the BIPM in France. Q: Can I use a free online converter for my business invoices? A: For rough internal estimates, yes. For invoices, tax returns, shipping declarations, contracts or any document that may be disputed, you should use the official conversion rate and rounding rules published by the relevant authority or agreed between the parties. A browser tool may produce the right number, but it cannot provide the audit trail, the date of the rate or the legal certainty that a tax office or court expects. Keep a screenshot or a record of the source, and when in doubt ask an accountant, lawyer or metrologist.