Author: Jason Carter

Which online file tools upload your files? We checked six

All six popular online file tools we checked upload your file to their servers to process it, and say so in their own documentation: Smallpdf, iLovePDF, Adobe Acrobat’s online tools, CloudConvert, Zamzar and remove.bg. They say they delete files after anything from about an hour to seven days. That’s not sinister, it’s how most of these services work, but it’s worth knowing before you drop in a payslip, a passport scan or a client’s contract. Tools that run in your browser, like JFDI, don’t upload anything.

Use tools that don’t upload

What each tool says

We read each company’s own help, security or privacy pages in October 2026. Policies change, so follow the links for the current wording.

Tool	Uploads your file?	What it says about keeping it	Source
Smallpdf	Yes, to servers in Ireland	Free tool usage: “Files deleted after 1 hour”	Smallpdf
iLovePDF	Yes	“Completely deleted within 2 hours of upload”, or straight away with a Delete it now button	iLovePDF
Adobe Acrobat online tools	Yes: “Acrobat online services upload your files to Adobe cloud storage”	Signed in: kept in your account. Not signed in: deleted after an unspecified short period	Adobe
CloudConvert	Yes, to servers in Germany, or the US for US users	“Automatically deleted after 24 hours at the latest”, or straight away with the delete button	CloudConvert
Zamzar	Yes	For people without an account, kept “for no longer than 7 days following receipt”	Zamzar
remove.bg	Yes	Deleted “at the latest about one hour after the upload”	remove.bg

All six are served over encrypted (HTTPS) connections, and several describe further security measures. None of this suggests they misuse files. The point is simply that, for a while, your file exists on a computer you don’t control.

Why it can matter

For a restaurant menu or a holiday photo, it probably doesn’t. It’s worth thinking twice when:

The file is about someone else. Customer lists, patient letters, HR documents and client contracts often come with legal duties. Many employers’ policies forbid putting them into unapproved online services.
The file is about you, in detail. Passports, bank statements, payslips and tax returns are exactly what identity thieves want.
The file is confidential before publication: results, a manuscript, a bid, an unannounced product.

Any company storing files, even briefly, can suffer a data breach, a mistake or a legal demand. A file that never leaves your device can’t be caught up in any of those.

How browser-only tools are different

A browser-only tool sends you the program instead of asking you for the file. The page loads code into your browser, and your browser does the converting, compressing or editing on your own computer. When you click Download, the result is saved from your browser’s memory straight to your device. Our explainer on how browser-only tools work goes into the technology.

That’s how every JFDI tool works: the PDF toolkit, file converter, image editor (including background removal), chart maker, map maker and QR code maker.

Check any tool yourself in two minutes

You don’t have to trust anyone’s privacy page, including ours. Your browser can show you every request a page makes.

Open the tool in Chrome, Edge or Firefox.
Press F12 (or right-click the page and choose Inspect) and open the Network tab.
Tick Preserve log (Chrome and Edge) or Persist Logs (Firefox, in the settings cog), and clear the list with the 🚫 button.
Use the tool as normal: add a file, convert it, download the result.
Look at the list. Sort by Size or look for requests with the method POST or PUT. An upload shows up as a request whose size is roughly the size of your file, sent to the tool’s server.

On a browser-only tool you’ll see the page fetch its own code (and, for some tools, a one-off download such as a font or an AI model), but nothing sent out that’s as big as your file. On an upload-based tool, you’ll see your file go.

JFDI File Converter does not upload your files away from your device. All of the work is performed locally, in private.

Some online file converter tools upload your files to their server - compromising your privacy.

Do browser tools have downsides?

Yes, a few:

Your device does the work, so very large files are faster on a computer than an old phone.
Nothing is saved for you. Close the tab and your work is gone unless you’ve downloaded it.
Some jobs need a server, such as OCR on scanned documents or converting formats browsers can’t read. Those services aren’t wrong to upload; just choose them deliberately.

October 23, 2026

How to redact a PDF properly (a black box isn’t enough)

To redact a PDF properly, the hidden words have to be removed from the file, not just covered. A black rectangle, a black highlight or a white box drawn on top hides text from view, but the text is still there underneath, and anyone can select it, copy it and paste it somewhere else. Use a real redaction tool, then check the result before you send it.

Redact a PDF

The black box mistake keeps happening

This isn’t a beginner’s error. It has caught out lawyers, government departments and newspapers for years:

January 2019: lawyers for Paul Manafort filed a document in his case with black bars over sensitive passages. A Guardian reporter found the text could be copied and pasted into a new document, revealing details about his contacts with Konstantin Kilimnik (Columbia Journalism Review).
2014: redactions in a leaked CIA presentation published by The New York Times could be bypassed by highlighting and copying, exposing an NSA agent’s name.
2011: a federal judge’s opinion in Apple v. Samsung had redacted passages that could be copied into a text document.
A 2011 study estimated that “thousands, and probably tens of thousands” of documents in the US federal courts’ PACER system had redaction failures.

The last three are from the American Bar Association’s round-up of embarrassing redaction failures.

Why covering doesn’t work

A PDF isn’t a picture of a page. It’s a list of instructions: put these words here in this font, draw this shape there, place this image. When you draw a black box, you add one more instruction, draw a black rectangle, on top. The instruction that draws the words is untouched, so:

Copy and paste pulls the text from under the box.
Search finds it.
Screen readers read it aloud.
Text extraction by any program, including search engines, sees it.

The same goes for black highlighting in Word before saving as PDF, and for white-out boxes.

What proper redaction does

Real redaction removes the content under the box from the file. In the JFDI PDF toolkit, the Redact tool works like this:

Open the PDF in the toolkit and go to Edit & sign.
Choose Redact (shortcut X) and drag over every name, number or passage that must go.
Download. Each page with a redaction is rebuilt as a 200 dpi image with the black boxes burned in, so the hidden words no longer exist anywhere in the file.

The trade-off: the rest of a redacted page can no longer be selected or searched, because the page is now a picture. Pages without redactions stay as normal text. Page numbers and watermarks are added afterwards, so they stay sharp.

Nothing is uploaded at any point. The PDF is opened and rebuilt inside your browser, which is the point when the document is confidential.

Don’t forget the hidden information

Visible text isn’t the only thing that leaks. PDFs can carry:

Document details: the author’s name, the software used, creation and editing dates.
Editing history: some programs save changes by adding to the end of the file, so earlier versions of a page can still be inside.
Attachments, scripts and form data.
Text under scans: a scanned page that’s been through OCR has an invisible text layer behind the image.

In the toolkit’s Finish tab, tick Remove hidden information before downloading. It strips the author, software, dates, embedded metadata, scripts, attachments and editing history. And check the file name: “Smith-v-Jones-settlement-£2.4m.pdf” can give away what the redactions hide.

White-out or redact?

	White-out	Redact
Looks like	A white (or coloured) box	A black box
The hidden words	Still in the file. Anyone can copy them.	Removed for good
Rest of the page	Unchanged	Turned into an image
Use it for	Tidying up, hiding a typo, making space for new text	Names, addresses, account numbers, anything confidential

The toolkit’s Change existing text tool also covers the original words rather than removing them, so never use it to remove private information.

Check before you send: five tests

Select all and copy (Ctrl+A, Ctrl+C) on each redacted page, then paste into a text editor. Nothing hidden should appear.
Search the PDF for a redacted word, such as a surname.
Open it in a different PDF reader: your browser’s built-in viewer and one other.
Check the document properties (File › Properties in most readers) for the author and title.
Read the file name.

Common questions

Can a redaction be undone?

Not when the content has been removed, as above. A drawn box can be “undone” by anyone with a PDF editor, by simply deleting the box.

Is redacting a scanned document different?

A scan is already a picture, so a black box burned into the image hides it. But scans often have an invisible OCR text layer; a real redaction tool removes that too, which is why the toolkit rebuilds the page.

Does Adobe Acrobat do this properly?

Yes, when you use its Redact tool and apply the redactions; drawing a black shape with its comment tools doesn’t. The same check above works for any tool.

October 23, 2026

How to convert HEIC to JPG without uploading your photos
To convert HEIC photos to JPG without uploading them, open them in a browser-based converter that does the work on your own device. In the JFDI image editor, drop in your HEIC files, set Save as to JPG, and press Download all. It works in any current browser, on Windows, Mac, Chromebook or phone, with no app to install, no account and no limit on how many photos you convert.

Convert HEIC to JPG

What HEIC is, and why it’s a problem

HEIC is the file type iPhones use for photos by default. Apple switched to the HEIF format, which HEIC files use, with iOS 11 in 2017, on the iPhone 7 and later (Apple). It stores photos at roughly half the size of a JPG with the same quality, which is great for your phone’s storage.

The trouble starts when the photos leave the iPhone. Many websites, upload forms, older Windows computers and photo printers don’t accept HEIC, and some show an error or a blank thumbnail. JPG opens everywhere.

Step by step
1. Open the image editor and choose HEIC to JPG on the start page, then drop your photos in. You can add several hundred at once.
2. Under Save as, choose JPG. (If you leave it on Same as the original, HEIC files become JPG anyway.)
3. Set the quality. 82, the starting point, is hard to tell from the original. Go lower for smaller files to email.
4. Check the preview. Drag the line across the picture to compare before and after; zoom to 100% to see real pixels.
5. Download all (ZIP), or This one for a single photo.
Safari reads HEIC directly. Chrome, Edge and Firefox use a decoder built into the page, which takes a few seconds for a large photo. Either way, nothing is uploaded.

Photos come out the right way up

Phones often save photos sideways, with a note in the file saying which way to turn them. Some converters ignore the note, so portrait photos come out on their side. The image editor turns the picture itself, so it looks right everywhere.

What happens to the date, camera and location

iPhone photos carry hidden details: the date and time, the phone model and settings, and usually the exact GPS location where the photo was taken.

By default, the editor removes all of them (Remove location and camera details is on). That’s the safe choice for photos you’ll post or send to strangers, because the location can reveal where you live. There’s more on this in how to remove location data from photos.

One thing to know: details can only be kept when a JPG is saved as a JPG. Converted HEIC photos always lose them, including the date taken. If you’re converting photos to keep in a photo library, which sorts by that date, use one of the other methods below.

Other ways to get JPGs from an iPhone
- Change the camera setting to save JPGs from now on: Settings › Camera › Formats › Most Compatible. Photos take more space on the phone.
- Transfer settings: with Settings › Apps › Photos › Transfer to Mac or PC set to Automatic, photos copied over USB are converted to JPG as they go.
- Share by email or Messages: iPhones may convert to JPG automatically when the other device doesn’t support HEIC.
- On a Mac, Preview can export a HEIC photo as JPG with File › Export.
- On Windows, the Photos app opens HEIC once Microsoft’s HEIF and HEVC extensions are installed from the Microsoft Store.
Is it really private?

Yes: the conversion happens in your browser tab, and the photos never leave your device. You can check this yourself; our post on which online tools upload your files shows how. Many “free HEIC converter” websites upload every photo to their servers to convert it, which is a lot to hand over for a phone’s worth of family pictures.

Common questions

My HEIC photo won’t open

A few HEIC files, such as some HDR or burst photos, use features the decoder doesn’t support. Open the photo in the iPhone’s Photos app and share or export it as JPG instead.

Should I use PNG or WebP instead?

For photos, JPG is the safe choice that opens everywhere. WebP is smaller at the same quality and works in all current browsers, which makes it good for websites. PNG is lossless but makes photos much bigger.

Can I resize them at the same time?

Yes. Set a size under Resize (for example Longest side 2,000 pixels) and every photo is resized as it’s converted.

Related
October 23, 2026
How to make a UK constituency map from a spreadsheet
To make a map of UK parliamentary constituencies, you need a spreadsheet with one column of constituency names (or ONS codes such as E14001063) and one column of values. Paste it into the JFDI map maker and you get a map of all 650 Westminster seats, drawn on the July 2024 boundaries, without signing up or uploading anything.

Open the map maker

This guide walks through it with real data, then covers the problems that trip most people up: old boundaries, spelling differences and party colours.

What you need

A table with two columns is enough (these values are made up):

Constituency Value
Aberafan Maesteg 41.2
Aberdeen North 38.7
Aberdeen South 44.0

The first row should be a header. The values can be numbers (turnout, majority, population, a survey score) or words (the winning party, Leave or Remain). Words make a map of categories, each with its own colour.

You don’t need to tidy the names first. The map maker recognises constituencies by name however they’re written, and by their nine-character ONS code. If your file has both, that’s fine too.

Step by step
1. Open the map maker and paste your cells into the box (copy them straight from Excel or Google Sheets), or drop a CSV or Excel file onto the page.
2. Check the match count. The Data section says which map it chose and how many rows matched. For constituency data, it picks Constituencies as hexagons (2024) by default.
3. Fix anything unmatched. Press Fix to see rows it couldn’t place. Typos come with a suggestion you can accept in one click.
4. Choose colours and a title. Under Colours, pick a light-to-dark set for numbers, or keep the automatic colours for categories.
5. Download it as PNG, SVG or PDF, or copy the embed code for a web page.
Every constituency gets one hexagon. This one shows each seat’s population, from ONS small-area estimates.

Hexagons or real shapes?

You get hexagons by default because they’re fairer for constituency data. Constituencies have roughly equal numbers of voters, but very unequal areas. Caithness, Sutherland and Easter Ross covers about 11,600 km², more than seven times the area of all 75 London seats put together (about 1,570 km²). On a normal map, the big rural seats dominate and you can barely see the cities.

On the hexagon map, which uses the House of Commons Library’s layout, every seat is the same size. Switch to real boundaries under Data › Map › Constituencies (2024) if you want geography instead. There’s more on this choice in Hex maps vs real maps: why election maps use hexagons.

Election results get party colours

If your values are party names (Labour, Conservative, Liberal Democrat, SNP, Plaid Cymru, Reform UK, Green and so on), the map uses each party’s usual colour, as election maps on TV do. You can switch to any other colour set under Colours.

Common problems

“Some of my constituencies don’t match”

The most common cause is old boundaries. Constituency boundaries changed at the July 2024 general election, and data from before then (including the 2019 results) uses the old seats. The map maker recognises the old ONS codes and tells you they’re from the previous boundaries, rather than quietly putting them in the wrong place.

Many names didn’t change in 2024, so a file can match almost perfectly and still be on old boundaries. Check the date of your data before you publish.

“It picked the wrong map”

If your table mixes levels (some councils, some constituencies), the map maker picks the level with the most matches. Choose the map yourself under Data › Map.

Totals and rates

If you’re mapping a count, such as the number of people claiming a benefit, compare like with like. Constituencies have similar electorates but not identical populations, so a rate (per 1,000 people, or a percentage) is usually fairer than a raw count. The map maker spots columns that look like totals and suggests a rate. Our explainer on what a choropleth map is and when it misleads goes into why.

Where the boundaries come from

The 2024 constituency boundaries are the Office for National Statistics’ generalised boundary files, published under the Open Government Licence, and the hexagon layout is the House of Commons Library’s, under the Open Parliament Licence. The map carries the required credit at the bottom; keep it if you crop the image. You can use the maps commercially.

Is my data private?

Yes. The map is drawn inside your browser, and the spreadsheet never leaves your device. That matters if you’re mapping unpublished polling, membership lists or canvassing returns. If you make a share link, the data travels inside the link itself, so treat the link like the data.

Related
October 23, 2026

Constituency	Value
Aberafan Maesteg	41.2
Aberdeen North	38.7
Aberdeen South	44.0

Why your free QR code stopped working, and how to make one that won’t

If a QR code that used to work now shows an error, an advert or a “subscription expired” page, it’s almost certainly a dynamic code. Dynamic codes don’t contain your link: they contain a short link to the QR company’s server, which forwards people to you, for as long as the company keeps doing it. A static code contains your link itself and keeps working forever. The JFDI QR code maker only makes static codes.

Make a QR code that never expires

Static and dynamic codes

	Static code	Dynamic code
What’s inside the code	Your actual link, Wi-Fi details or text	A short link to the QR company’s server
Works if the company closes or you stop paying	Yes, forever	No
Can be tracked (how many scans, where, when)	No	Yes
Can show adverts or a landing page first	No	Yes
Can change the destination after printing	No	Yes

Dynamic codes are a real product with real uses, such as campaign tracking. The problem is that many sites offering “free QR codes” make dynamic ones by default, without making the catch clear.

What QR companies say themselves

You don’t have to take our word for it. One vendor, Uniqode, writes that dynamic codes “can expire when a subscription lapses, a monthly scan limit is reached”, and that free dynamic codes “expire when the free trial ends or when a monthly scan limit is reached” (Uniqode). Another, QRStuff, says that when a paid subscription expires, “a 50 scans/month limit is automatically applied to each code” (QRStuff).

The printed code looks exactly the same either way. You only find out when a customer tells you the code on your menu, poster or packaging doesn’t work.

How to tell which kind you have

Point your phone’s camera at the code and look at the link it shows before you open it:

Your own website’s address, such as yourcafe.co.uk/menu: it’s static. It will work as long as that page exists.
Someone else’s short address, often a QR company’s domain or an unfamiliar short link: it’s dynamic, and it depends on that company.

How to fix a dead code

You can’t revive a dynamic code without paying the company that controls its redirect. Your options:

Pay to reactivate it, if reprinting would cost more.
Replace it with a static code pointing straight at your page, and reprint.
Cover it with a sticker of a new static code, as a quick fix for menus and posters.

Making a code that won’t expire

Open the QR code maker and choose what the code opens: a link, Wi-Fi, a contact card, an email, a phone call or a text message.
Enter the details. The code updates as you type.
Style it if you like, with colours, rounded dots and a logo. The notes under the preview warn you about anything that could stop it scanning.
Download a PNG, SVG or PDF. Use SVG or PDF for print.
Test it with two or three phones before you print lots.

There’s no sign-up, no watermark and no limit, and the codes are yours to use commercially. Because the code holds only what you typed, nobody, including us, can see when or where it’s scanned.

The one thing to plan for

A static code can’t be changed after it’s printed. If the destination might change, such as a menu, a price list or an event page, point the code at a page on your own website that you can update, such as yourcafe.co.uk/menu, rather than at a PDF file whose address may change. You get the flexibility of a dynamic code without depending on anyone else.

October 23, 2026

How browser-only tools work, in plain English

A browser-only tool sends you the program instead of asking for your file. When you open the page, your browser downloads the tool’s code; when you add a file, the browser reads it from your disk and processes it on your own computer; when you click Download, the result goes straight from your browser’s memory to your device. The website’s server never receives the file, so it can’t keep, leak or look at it.

Try the tools

Two ways a file tool can work

	Upload-based tool	Browser-only tool
Where the work happens	The company’s server	Your computer or phone
Your file travels	To the server and back	Nowhere
Speed depends on	Your upload speed and their queue	Your device
Size limits	Often capped on free plans	Your device’s memory
Works offline after loading	No	Often, for much of the work
Can save your work for you	Yes, if you have an account	No: you download it

Most well-known online file tools are upload-based; we looked at what six of them say about it in Which online file tools upload your files?

What makes it possible

Ten years ago, a web page couldn’t realistically edit a PDF or convert a video. Several things changed that:

WebAssembly. A way to run fast, compiled programs inside a web page, at close to the speed of a desktop app. It became an official W3C web standard on 5 December 2019, described as “a virtual instruction set architecture that enables high-performance applications on the Web” (W3C). It lets existing software, written in languages like C, C++ and Rust, run in the browser.
Access to your device’s media hardware. The WebCodecs interface lets a page use the same chips your device uses to record and play video, so converting a video in a browser can be as fast as in a desktop app.
Access to your graphics card. WebGPU lets a page run heavy maths, such as an AI model, on the graphics processor.
Background threads. Web Workers let a page do heavy work without freezing, so you can keep using it while a batch of 300 photos processes.
Reading and saving files locally. Browsers can read a file you choose and save a result, all without a server in between. Chrome and Edge can even write results straight into a folder you pick.

How each JFDI tool uses them

Tool	What runs in your browser
PDF toolkit	Mozilla’s PDF.js draws the pages; pdf-lib rebuilds the file with your changes
File converter	SheetJS for spreadsheets; Mediabunny with WebCodecs for video and audio
Image editor	A HEIC decoder compiled to WebAssembly; the IS-Net background-removal model on WebGPU, or WebAssembly when there’s no usable graphics card
Chart maker and map maker	The d3 drawing library, plus boundary files served from this site
QR code maker	A QR encoder written in plain JavaScript

All of these are open-source libraries, credited in our licences file.

What still gets downloaded

“Nothing is uploaded” doesn’t mean “nothing is downloaded”. Your browser fetches:

The tool’s code, when you open the page.
Extra pieces when they’re needed: the PDF engines in the background after the PDF toolkit opens, the 44 MB background-removal model the first time you use it, map boundaries for the US or UK when you make those maps. Your browser keeps them, so later visits are quick.
The font for the pages, from Google Fonts. Like any website, Google can see your IP address when your browser fetches it. Our privacy page explains this.

None of these requests contain your files.

How to check for yourself

Open the browser’s developer tools (F12), choose the Network tab, and use the tool. You’ll see every request the page makes. An upload shows up as an outgoing request about the size of your file; on a browser-only tool, there isn’t one. The steps are in Which online file tools upload your files?

The trade-offs

Your device does the work, so a huge video converts faster on a laptop than an old phone.
Nothing is saved for you. That’s the privacy, but it also means closing the tab loses unsaved work. Download as you go.
Some jobs still need a server, such as recognising text in scanned documents (OCR) or reading formats browsers can’t, like old Word .doc files.

October 23, 2026

Map of the month: equal electorates, very unequal populations

Westminster constituencies were drawn to have nearly equal numbers of voters, but the number of people living in them varies almost twofold. Leaving aside five protected island seats, the smallest constituency by population, Congleton, has about 84,000 residents; the largest, Cities of London and Westminster, has about 163,000. Thirty-one of the fifty most populous seats are in London.

Make your own constituency map

A hexagon map of UK constituencies shaded by population, with the darkest hexagons concentrated in London and major cities — Population of each Westminster constituency (2024 boundaries). One hexagon per seat. Source: ONS small-area population estimates (mid-2019), via mySociety.

The rule: equal electorates

For the 2023 boundary review, which set the seats used at the July 2024 general election, each constituency had to have between 69,724 and 77,062 registered electors, within 5% of a quota of 73,393, based on the electoral register of 2 March 2020. Five island seats were protected from the rule: the two Isle of Wight seats, Na h-Eileanan an Iar, Orkney and Shetland, and Ynys Môn (2023 review).

So every MP outside the islands represents roughly the same number of voters. They don’t represent the same number of people.

The numbers

Across the 645 seats subject to the rule:

	Population
Smallest (Congleton)	83,907
10th percentile	92,181
Median	100,479
90th percentile	117,075
Largest (Cities of London and Westminster)	162,641

The ten most populous seats are:

Constituency	Population
Cities of London and Westminster	162,641
Birmingham Ladywood	157,574
Kensington and Bayswater	147,231
Hampstead and Highgate	144,676
Poplar and Limehouse	141,599
Erith and Thamesmead	139,649
Tottenham	138,358
Holborn and St Pancras	137,909
Bethnal Green and Stepney	135,919
Coventry South	135,355

And by region, the average London seat has about 119,000 residents, against about 96,000 in Scotland and 97,000 in the South West.

Why the gap?

Equal electorates and unequal populations aren’t a contradiction: lots of residents aren’t on the electoral register. The seats with the most people per voter tend to share features that are likely to explain much of it:

Children. Under-18s can’t vote in general elections, so seats with more young families have more residents per elector.
People who can’t vote in general elections. Most foreign nationals, including most EU citizens, can’t, while Irish and qualifying Commonwealth citizens can. Seats with many residents from elsewhere have fewer electors per head.
People who aren’t registered. Registration rates are lower among renters, people who move often and students, all common in inner cities.
Timing. The populations here are mid-2019 estimates and the electorates come from March 2020; places growing fast have moved on since.

This data can’t separate these effects, so treat them as likely contributors rather than a measured breakdown.

What it means

Equal electorates are the legal standard, and there are good arguments for it: MPs are elected by voters. But MPs also serve residents who can’t vote, taking up their casework and representing them in Parliament. An MP for Birmingham Ladywood represents nearly twice as many people as one for Congleton.

How this map was made

The population figures are the ONS’s mid-2019 small-area estimates, added up for each 2024 constituency by mySociety, and they’re built into the JFDI map maker: it’s the UK constituencies example on the start page. The map uses the House of Commons Library’s hexagon layout, so every seat is the same size. On a normal map, London’s 75 seats would cover less than 1% of the picture; see why election maps use hexagons.

Open the example, change the colours, switch to real boundaries or add your own data, and download it. Every map you make is yours to use, with the boundary credit kept at the bottom.

Map of the month takes a public dataset each month and maps it. Suggestions are welcome.

October 23, 2026

How to open a CSV in Excel without losing leading zeros

Excel drops leading zeros because a CSV file has no column types, so Excel guesses, and anything that looks like a number becomes one. 00123 turns into 123, phone numbers lose their first 0, and long ID numbers get rounded to 15 digits. To stop it, either turn off the conversion in Excel’s options, import the file with the column set to Text, or convert the CSV to an Excel file that already stores those columns as text.

Convert CSV to Excel, zeros kept

What goes wrong

In the CSV	After opening in Excel
`00123` (product code)	`123`
`07700 900123` (UK mobile)	stays text only because of the space; `07700900123` becomes `7700900123`
`02134` (US ZIP code)	`2134`
`4111111111111111` (16-digit number)	`4.11111E+15`, and the last digit becomes 0

The last row is the nastiest. Excel stores numbers to 15 significant digits, so the sixteenth digit is replaced with a zero for good. Save the file and the original number is gone.

Fix 1: change Excel’s settings (Microsoft 365 and Excel 2024)

Newer versions of Excel let you switch off the conversions. On Windows go to File › Options › Data; on a Mac go to Excel › Preferences › Edit. Under Automatic Data Conversion, untick:

Remove leading zeros and convert to number
Keep the first 15 digits of long numbers and display in scientific notation if required

Microsoft documents both in Set automatic data conversions. The settings apply to every workbook you open from then on, so they’re the best fix if you handle codes and IDs all the time. Older versions of Excel don’t have them.

Fix 2: import instead of opening

Instead of double-clicking the CSV, open a blank workbook and use Data › From Text/CSV. In the preview, choose Transform Data, select the columns that hold codes, and set their type to Text. Then Close & Load.

This works in any recent Excel, but you have to do it every time, for every column.

Fix 3: convert the CSV to an Excel file first

The JFDI file converter reads each value for what it is. Codes such as 00123, phone numbers and postcodes stay exactly as written, and numbers longer than 15 digits are kept as text with every digit. Real numbers, dates and percentages still become numbers and dates, so you can sort and calculate with them.

Open the file converter and drop in your CSV (or a whole folder of them).
Choose Excel under Convert to.
Press Convert, then Download.

The workbook also gets a bold header row that stays put when you scroll, filter buttons and sized columns. Nothing is uploaded: the conversion runs in your browser, which matters when the file is a customer list.

Fix 4: Google Sheets

In Google Sheets, use File › Import › Upload, and untick Convert text to numbers, dates, and formulas before importing. Leading zeros and long numbers stay as written.

Already lost the zeros?

If the file has been saved after Excel removed them, the zeros are gone from the file. For codes with a fixed length you can add them back with a formula, for example =TEXT(A2,"00000") for five-digit ZIP codes. Long numbers rounded to 15 digits can’t be recovered; go back to the original export.

Why it matters

Lost leading zeros cause quiet failures: postcodes that don’t match, FIPS county codes that leave Alabama off a map, product codes that don’t find their stock. The same habit of guessing types is behind Excel turning text into dates, which famously corrupted gene names in published research. That’s the subject of why dates change when you open a CSV.

October 23, 2026

Hex maps vs real maps: why election maps use hexagons
Election maps use hexagons because constituencies have roughly equal numbers of voters but wildly unequal areas. On a normal map, a handful of huge rural seats fill most of the picture and the cities shrink to specks. A hex map gives every seat one identical hexagon, so the picture shows seats, which is what decides an election.

Make a hex map of your data

The numbers behind the distortion

I added up the areas of all 650 Westminster constituencies on the July 2024 boundaries (ONS boundary files):

Area Seats Share of seats Share of UK land
London 75 11.5% 0.6%
Scotland 57 8.8% 32%
Rest of the UK 518 79.7% 67%

London elects more MPs than Scotland, yet on a normal map Scotland takes up about fifty times as much space. Fifty-nine seats are smaller than 20 km², while the largest, Caithness, Sutherland and Easter Ross, covers about 11,600 km². Colour each seat by the winning party and the map mostly shows who won the Highlands.

The same data both ways. On the left, real boundaries; on the right, one hexagon per seat.

What a hex map is

A hex map is a kind of cartogram: a map where the size of each area stands for something other than land. Here, each constituency is one hexagon, placed roughly where it really is so neighbours stay neighbours. London becomes a recognisable cluster, and you can see every seat in Manchester or Glasgow.

The layout used in the JFDI map maker is the House of Commons Library’s, the same one many newsrooms use. It’s published under the Open Parliament Licence.

When to use hexagons

Use a hex map when each area counts equally in what you’re showing:
- election results and vote shares
- polling or MRP estimates by seat
- anything per constituency that an MP or campaign will act on, such as casework or survey responses
Use real boundaries when geography matters:
- something physical, such as flood risk, broadband coverage or distance to a hospital
- when readers need to find a specific place by shape or coastline
- local data within one region, where size differences are smaller
The trade-offs

Hex maps aren’t free of distortion; they move it somewhere else.
- Shapes and positions are approximate. A seat may sit a hexagon away from where you expect. Labels or a hover tooltip help people find their own.
- Population isn’t exactly equal. Seats were drawn to have electorates within 5% of 73,393 (with five protected island seats excluded), but total population, including children and people not registered, varies much more.
- Readers need a moment. Add a subtitle such as “One hexagon per constituency” so nobody mistakes it for geography.
Councils: a different kind of cartogram

Local authorities vary in population as well as area, from the Isles of Scilly to Birmingham, so one hexagon each would be unfair the other way. For councils, the map maker offers Local authorities sized by population, also from the House of Commons Library, where each council is built from small hexagons of about 8,000 people. Big cities get room in proportion to their residents.

How to switch

In the map maker, load your constituency data, then open Data › Map and choose between Constituencies (2024) and Constituencies as hexagons (2024). Your colours, title and data carry over, so you can download both and compare them. The step-by-step guide covers getting your data in.

Sources
- Constituency areas: ONS Westminster Parliamentary Constituencies (July 2024) boundaries, Open Government Licence.
- Electoral quota: 2023 review of Westminster constituencies.
- Hexagon layout: House of Commons Library, Open Parliament Licence.
October 23, 2026

Area	Seats	Share of seats	Share of UK land
London	75	11.5%	0.6%
Scotland	57	8.8%	32%
Rest of the UK	518	79.7%	67%

Which chart should I use? A guide to 42 common charts

Choose a chart by the question it needs to answer. Comparing items? Use bars. Change over time? Use a line. Parts of a whole? Use a stacked bar or, for two or three parts, a pie. Relationship between two numbers? Use a scatter plot. Flows between stages? Use a Sankey. The rest of this guide covers the cases in between.

Try every chart with your own data

Every chart below is in the JFDI chart maker. Paste your data once and the gallery redraws all 130 chart types with it, putting the ones that fit your data’s shape first.

1. Comparing items

Chart	Use it when	Avoid when
Bar chart	Comparing a handful to a few dozen categories, especially with long names	Showing change over time
Column chart	Comparing a few categories, or a few time periods	Category names are long
Grouped bar or column	Comparing two or three series per category	More than about four series
Dot plot	Many categories with values close together	You need to show zero as a baseline
Lollipop chart	Like a bar chart, but lighter when bars are many	Values are negative and positive
Bullet chart	Actual against a target, with good/bad bands	The audience hasn’t seen one before
Slope chart	Who rose and who fell between two points in time	More than two points in time
Diverging bar chart	Values above and below zero, or a midpoint	Everything is on one side

Sort bars by value unless the categories have a natural order (age groups, months).

2. Change over time

Chart	Use it when	Avoid when
Line chart	Continuous change, many time points	You have only two or three points
Column chart	A few periods, or totals per period	Dozens of periods
Step chart	Values that change in jumps, such as prices or interest rates	Change is gradual
Area chart	Emphasising volume over time for one series	Comparing several series that overlap
Stacked area chart	A total and its components over time	Readers need to compare the middle layers
Line with final dot	Highlighting the latest value	—
Year-over-year line	Comparing the same months across years	Data isn’t seasonal
Fan chart	A forecast with widening uncertainty	You don’t have a range
Bump chart	Rankings rising and falling	The actual values matter more than ranks
Calendar heatmap	Daily values across a year, showing weekly patterns	Data is monthly

3. Parts of a whole

Chart	Use it when	Avoid when
Stacked bar or column	Parts of a total, compared across categories	More than about five parts
100% stacked bar	Shares, not totals, across categories	Totals matter
Pie or donut	Two or three parts of one whole	Parts are similar in size, or there are many
Waffle chart	Showing a percentage people can count, such as 37 in 100	Values need decimals
Treemap	Many parts, or parts within parts	Precise comparison matters
Sunburst	Hierarchies with two or three levels	More than three levels
Waterfall chart	How a start value becomes an end value	Values aren’t sequential changes

Pie or bar? Bars win for anything but two or three slices, because people compare lengths far more accurately than angles.

4. Distributions

Chart	Use it when	Avoid when
Histogram	The shape of one set of values	Comparing many groups
Box plot	Comparing medians and spread across groups	Audience isn’t used to them
Violin plot	Comparing the shape of distributions	Groups are small
Beeswarm plot	Showing every individual value, up to a few hundred	Thousands of points
Population pyramid	Two groups (such as men and women) by age band	Groups aren’t comparable

5. Relationships

Chart	Use it when	Avoid when
Scatter plot	Two numbers per item: is there a pattern?	Only a handful of items
Scatter with trend line	Showing the direction of a relationship	The relationship isn’t straight
Bubble chart	A third number shown by size	Bubbles overlap heavily
Connected scatter plot	How two numbers moved together over time	More than one or two series
Heatmap	Values across two categories, such as day × hour	Precise values matter
Quadrant chart	Sorting items into four groups, such as impact × effort	Axes have no natural midpoint

6. Flows and processes

Chart	Use it when	Avoid when
Sankey diagram	Quantities flowing between stages	More than about 20 nodes
Alluvial chart	How groups move between categories over stages	Categories change between stages
Funnel Sankey	A conversion funnel with drop-offs	Steps aren’t sequential
Income statement chart	Revenue flowing to costs and profit	—

7. Ranking races and small multiples

Chart	Use it when	Avoid when
Bar chart race	An animated ranking over time, for social or presentations	Printed or static reports
Grid of line charts (small multiples)	Many series that would tangle on one chart	Series have very different scales

Three rules that apply to every chart

Title it with the finding. “Sales doubled after the price cut” tells readers what to look for; “Sales by month” doesn’t.
Start bars at zero. The length of a bar is the value, so a cut-off axis exaggerates differences. Lines can start elsewhere, if you say so.
Use colour for meaning. Highlight the one series that matters and grey the rest, rather than giving each its own bright colour.

October 23, 2026

Author: Jason Carter

What each tool says

Why it can matter

How browser-only tools are different

Check any tool yourself in two minutes

Do browser tools have downsides?

Related

The black box mistake keeps happening

Why covering doesn’t work

What proper redaction does

Don’t forget the hidden information

White-out or redact?

Check before you send: five tests

Common questions

Can a redaction be undone?

Is redacting a scanned document different?

Does Adobe Acrobat do this properly?

Related

What HEIC is, and why it’s a problem

Step by step

Photos come out the right way up

What happens to the date, camera and location

Other ways to get JPGs from an iPhone

Is it really private?

Common questions

My HEIC photo won’t open

Should I use PNG or WebP instead?

Can I resize them at the same time?

Related

What you need

Step by step

Hexagons or real shapes?

Election results get party colours

Common problems

“Some of my constituencies don’t match”

“It picked the wrong map”

Totals and rates

Where the boundaries come from

Is my data private?

Related

Static and dynamic codes

What QR companies say themselves

How to tell which kind you have

How to fix a dead code

Making a code that won’t expire

The one thing to plan for

Related

Two ways a file tool can work

What makes it possible

How each JFDI tool uses them

What still gets downloaded

How to check for yourself

The trade-offs

Related

The rule: equal electorates

The numbers

Why the gap?

What it means

How this map was made

Related

What goes wrong

Fix 1: change Excel’s settings (Microsoft 365 and Excel 2024)

Fix 2: import instead of opening

Fix 3: convert the CSV to an Excel file first

Fix 4: Google Sheets

Already lost the zeros?

Why it matters

Related

The numbers behind the distortion

What a hex map is

When to use hexagons

The trade-offs

Councils: a different kind of cartogram

How to switch

Sources

1. Comparing items

2. Change over time

3. Parts of a whole

4. Distributions

5. Relationships