How to Find Photo Location Without EXIF Data
· 4 min read
Short answer
To find photo location without EXIF you work from the image instead of the file. Reverse image search looks for copies of the same scene online, AI geolocation estimates a region from visual features, and manual clue-reading narrows it further using signage, road markings and vegetation.
This is the normal case, not the awkward one. Almost every image that has passed through a social feed has had its metadata removed on upload, so "no EXIF" describes most photos you will ever be handed.
Confirm it first, though. Checking for coordinates takes about 10 seconds and occasionally saves an hour of guesswork — and when they are there, they are accurate to roughly 5 m rather than to a city.
Why is the metadata usually missing?
Because publishing platforms re-encode images on upload and discard the metadata as a side effect. It is also absent if location services were off, if the file was converted, or if the image is a screenshot.
Which routes destroy metadata and which preserve it is worth knowing precisely — the route-by-route breakdown covers it, and Instagram specifically strips before the post is ever visible. The short version: publishing strips, transferring does not.
Does reverse image search work on a stripped photo?
Yes, because it never used the metadata. Reverse search compares pixels, so removing EXIF changes nothing about its chances. What decides success is whether the same scene has been published somewhere an engine has indexed.
This is the highest-value first move: when it works it gives an exact answer, and it costs a minute. Run more than one engine — indexes differ enough that a photo drawing a blank on one matches immediately on another. The comparison covers which to try for which subject.
Crop first. A full street scene gives the matcher a dozen competing signals; a single shopfront or monument gives it one.
What can AI geolocation do that search cannot?
It works when the scene has never been published. Instead of matching against an index it reads the image as evidence — architecture, road markings, vegetation, signage, light — and infers a probable region from what it sees.
How that inference actually works is worth understanding before trusting a result. The trade is precision for coverage. Reverse search is exact or nothing; AI always returns something, and that something is a city or region rather than an address. On a personal photo of an ordinary street, "always returns something" is the difference between an answer and none.
How much can you work out by eye?
Often more than an automated tool. Legible text, the side of the road traffic drives on, kerb markings, utility pole design, plant species and shadow direction each narrow the search, and together they can reach a single street.
| Clue | Typically narrows to |
|---|---|
| Legible shop or street name | A specific street, via a map search |
| Script and language | A country or small group of countries |
| Driving side and road markings | A country or region |
| Utility poles, bollards, postboxes | A country |
| Licence plate shape and colour | A country |
| Vegetation and soil colour | A climate band |
| Shadow direction and length | A latitude range and time of day |
No single row is decisive, and the ordering matters more than the list. The method is intersection: each clue removes possibilities, and the answer is whatever survives all of them. The full four-method guide works through the order to apply them in.
When is a photo genuinely unplaceable?
When it contains no geographic evidence. Indoor scenes, close-ups, plain sky or water, studio shots and heavily cropped images give every method nothing to work with, and the honest result is no answer rather than a confident guess.
Modern construction is the other hard case. Apartment blocks, motorway junctions and retail parks are built to similar patterns across whole continents, so a photograph of one is close to unplaceable regardless of what analyses it.
Frequently asked questions
- Can you recover EXIF data that was stripped?
- No. The tags are gone from that copy and cannot be reconstructed. If someone still holds the original file, that copy has them — but the stripped version is permanently stripped.
- Does a screenshot have any location data?
- Not from the original photo. A screenshot is a new image created by your device, and it carries its own metadata, which normally has no GPS coordinates attached.
- Which method should you try first?
- Reverse image search, because it is fast and, when it hits, exact. Fall back to AI geolocation when nothing matches, and use manual clue-reading to confirm or reject whatever either suggests.
- How accurate is the result without metadata?
- Reverse image search is exact when it finds the scene. AI geolocation typically reaches the correct city. Manual analysis ranges from a continent to a single street depending on how much is visible.
Sources
- Exif 3.0 (CIPA DC-008-2023) — Camera & Imaging Products AssociationDefines the GPS tags that are absent in a stripped file.
- GPS.gov — GPS Accuracy — U.S. GovernmentWhat the missing coordinates would have offered.
Keep reading
How to Find Out Where a Photo Was Taken
Four ways to work out where a photo was taken: GPS metadata, reverse image search, AI geolocation and reading the image itself. What each can and cannot do.
· 6 min read
The Best Reverse Image Search Tools, Compared
Google Lens, Yandex, Bing and TinEye do different things. A comparison of what each engine is good at, and when to reach for AI geolocation instead.
· 6 min read