Embed location data into your photos

Your JPEG is read by the browser, modified by the browser, and saved by the browser. There is no upload step, no server-side processing and no temporary copy on our end, because the code path that writes the metadata runs in the page you are looking at.

This is not a privacy nicety, it is the whole point. A geotag is a precise coordinate — often a home address for a service business run from home, or a business address you would rather not have sitting in a stranger's storage bucket. A tool that uploads the file has your location whether it means to keep it or not.

The practical consequence: the tool works with no account, no email and no rate limit, and it keeps working if you close the tab mid-way. Nothing was queued anywhere.

EXIF does not store a decimal coordinate. It stores degrees, minutes and seconds, each as a fraction — a pair of whole numbers. So 41.0082 has to be taken apart before it can be written.

The obvious way to do that is to multiply by 60 twice in floating point, and it is subtly wrong. Binary floating point cannot hold most decimal fractions exactly, so the seconds value drifts in the last digits and the coordinate you get back is not the one you typed. Small, but it accumulates in exactly the digits that carry street-level precision.

So the conversion works in whole numbers instead: the coordinate is fixed to seven decimal places as text, then scaled into hundred-thousandths of an arc-second and split with integer division. The seconds field is written as that integer over 100000. What you type is what ends up in the file.

Alongside the coordinates, the file records that they are in WGS-84 — the reference system GPS and Google Maps both use.

Plenty of writers skip this field, and then every program that reads the photo has to assume a datum. The assumption is usually right, which is exactly why the failure is annoying when it is wrong: the photo lands somewhere near your business rather than at it, and nothing in the file explains why. Stating the datum removes the guess.

The GPS version and the hemisphere references are written the same way, so the block is complete rather than minimal. A reader that validates the structure accepts it instead of falling back to partial parsing.

The tool writes your text into two different kinds of field, and they have different character rules. Getting this wrong is why business names come out as mojibake in some photo viewers.

The title, subject, comment, author and keyword fields are the Windows XP metadata tags, which are defined as UTF-16. Turkish characters go in unchanged — ğ, ü, ş, ı, ö, ç all survive, and Windows Explorer, Lightroom and most photo managers read them back correctly.

The classic EXIF fields are a different story: the specification defines them as ASCII. Writing UTF-8 into them is common and produces garbled output in strict readers, so those fields are transliterated instead — ğ becomes g, ü becomes u, and so on. Your name stays legible everywhere rather than being correct in one program and broken in another.

Keywords are joined with semicolons, which is the separator Windows and most cataloguing software expect when reading the keyword tag back as a list.

If your photo already carries EXIF data — and anything from a phone does — the tool overwrites that segment rather than adding a second one.

This sounds like housekeeping and is the difference between working and silently not working. A JPEG can physically contain two metadata segments, but readers stop at the first one they find. Append instead of replace and your new coordinates are in the file, unreachable, while the viewer keeps showing the old ones. There is no error to notice, just a photo that looks untagged.

The tool also checks whether the photo already has GPS coordinates when you open it, and tells you. That is usually worth knowing before you overwrite them: a phone photo taken at your business may already hold the right location.

The tool accepts JPEG and JPG. PNG, HEIC, WebP and RAW files are rejected at the file picker rather than half-processed.

The reason is structural, not effort. EXIF lives in a JPEG as a specific marker segment, and that is what the writer targets. PNG has no equivalent standard location, and HEIC — the default on modern iPhones — is a container format that needs a different library entirely. A tool that accepted them would have to convert first, and converting an iPhone photo to JPEG to add a coordinate throws away image quality to add a field.

If your photos are HEIC, the practical route is to export them as JPEG on the phone or in your photo app, then geotag. Most phones have a setting to shoot JPEG directly.

It does not guarantee a ranking, and it is worth being blunt about why. Google has not confirmed that it reads EXIF GPS from photos as a local ranking signal, and anyone who tells you the correlation is proven is selling something.

What accurate location metadata does do is keep your own files consistent and useful. Your photo library stays organised by real location, the coordinates survive if the image is reused elsewhere, and nothing in your metadata contradicts the address on your profile.

If your goal is map visibility, the levers with actual evidence behind them are elsewhere: your primary category, review volume and reply rate, profile completeness, and posting regularly. Geotagging is cheap and harmless housekeeping — treat it as that rather than as a ranking tactic, and you will not be disappointed by it.