Reading results
How to read a face search result: pages, not a name
A face search result is a list of public pages where a similar face may appear. It is not a name and it is not a verdict. The terms put it in one line: a result means “this public photo looks similar,” not “this is definitely the same person.” What follows is the order I read a list in.

My companion post on the AI image detector result covers a different tool. This one is about the list you get after a face search.
What you see before you pay
Each search shows its real match count for free. You pay to open the exact source pages and see unblurred thumbnails, and only when there are matches to open. Logged out you get 3 searches a day. A free account gets 10 a day, plus 3 free unlock credits once. The pricing page has the current numbers. If a search returns nothing, there is nothing to unlock and nothing to pay.
That count is useful on its own. A count of zero and a count of dozens are different situations before you look at a single page.
Make sure the search used the right face
A search uses the largest face in the photo. There is no picker. If you uploaded a group photo or a screenshot with several faces, the list may be about someone else, so crop and search again before you read anything into it.
Read the pages in this order
- Open the page, not just the thumbnail. A thumbnail is a crop. The page is where the context lives: a caption, a date, an account name, what the site is for.
- Ask what the page says about the person. A page that names someone is the page's claim, not a finding of this tool. Pages get things wrong, and photos get reposted by people other than their subject.
- Look for agreement between unrelated pages. One match is a lead. Several unrelated pages that tell the same story are worth more than one that tells it loudly. Several copies of one image on different sites are one source, not several.
- Use the score only to choose what to open first. A score sits beside each match. I do not treat any score as a ruling on whether two photos show the same person, and I do not publish a cutoff. Face matching is probabilistic, and the terms say matches can be wrong.
- Find something that does not depend on the photo. A video call, a shared contact, a profile with a history. The photo can start a question. It cannot close it.
What the list cannot say
It cannot say who someone is, and the service is not built to. Results link to public pages only, and we do not host the photos. It also cannot say a person is safe, because the list covers only what our own index and the engines we query can see, and the photo goes to Yandex, FaceCheck.ID, and Google Lens for that. Private profiles and login-walled apps do not show up in it. And the list is not a consumer report: the footer says plainly that results cannot be used for decisions about credit, employment, insurance, or tenant screening.
If you reach the end and still do not know, that is a normal outcome. Say so, and do not confront or accuse anyone. The terms say to verify independently first.
What gets stored
The photo you upload is deleted from image hosting within 15 minutes. A small low-resolution thumbnail stays on the result page so you can recognise the search later. We also keep the URLs, site names, and scores the search returned, so you can reopen what you paid for. You can ask us to delete them.
Questions people ask
Does a face search result give me a name?
No. It lists public pages that may show a similar face. A page can carry a name, but the page wrote it, not the search, and the terms say a result means this public photo looks similar, not that it is definitely the same person.
What does the free match count tell me?
Every search shows the real number of matches for free, and you pay only to see the exact pages and thumbnails. If there are no matches, there is nothing to unlock and nothing to pay.
Why did the search pick the wrong face in my photo?
A search uses the largest face in the picture and there is no picker. For a group photo, crop to the one face you mean before you upload.
Written by Nikola Marković. Checked against the live site on October 19, 2026.