OSINTsearch vs fingerprint.to vs Revealer: Three Handles Tested

Three handles, three engines, nine exports. On 2026-09-05 we put davidbombal, srhoe and deadcatx3 through OSINTsearch, fingerprint.to and Revealer.US, exported every result as JSON, and counted what was in the files. On davidbombal the totals are 59 from OSINTsearch, 69 from fingerprint.to and 57 from Revealer. On srhoe, 37, 46 and 35. On deadcatx3, 21, 22 and 11 real rows (Revealer's export for that handle is from Sep 4, and 17 of its 28 rows are artifacts; more on that below). Across the three tools, davidbombal resolves to an 80-platform union, 46 of which all three found.

This is the OSINTsearch vs fingerprint and OSINTsearch vs Revealer test that our earlier username search benchmark did not run. That post put fingerprint.to first and Revealer second, and OSINTsearch was not in it. Disclosure: we build Revealer.US. Assume bias and check the files; Revealer loses more than it wins here.

How we tested OSINTsearch, fingerprint.to and Revealer

Three handles, picked to be small. davidbombal, srhoe and deadcatx3 resolve to a few dozen profiles at most, so every row can be read and every platform name compared. davidbombal is a public YouTube educator and can be named. srhoe and deadcatx3 are handles only; nothing from their bios, names or locations is printed here.

Two days, not one. Every OSINTsearch and fingerprint.to export carries a 2026-09-05 timestamp, as do Revealer's davidbombal and srhoe exports. Revealer's deadcatx3 file is an "advanced" export from 2026-09-04, with an older schema (no join date, user ID or avatar field), so the deadcatx3 comparison is not same-day, and the file is labelled as such everywhere it appears.

Two OSINTsearch runs on srhoe. We ran srhoe on OSINTsearch at 09:24 and again at 21:20. The morning file (export schema 2) has 37 rows and no endpoint total. The evening file (schema 3) has 31 rows and records 770 endpoints scanned, 698 not found and 39 that could not be checked. The six platforms that dropped between runs are Buy Me a Coffee, Ko-fi, Rocket League Tracker, Snapchat, TikTok and a second Instagram row. The headline uses the 37; the scoreboard shows both.

What we counted. For every export: rows, distinct platforms after normalising names to lowercase with spaces, underscores and dots removed, rows with a non-empty bio, a non-empty location, a join date that is not null and not the zero time 0001-01-01, an avatar URL, a user ID, a former-username or name-history field, and a ban or account-state flag. Then pairwise platform overlap as Jaccard (shared over union) and the three-way union.

Counts as the tools report them. OSINTsearch reports 770 endpoints, fingerprint.to 759 on davidbombal and 618 on the other two, Revealer 813. Those pools are not the same list, so hit rates are context, never a ranking. OSINTsearch's own Sep 5 post on this comparison reported 59, 67 and 57 on davidbombal with 47 of 77 platforms shared by all three; it counted fingerprint.to's 69 rows as 67 profiles. Our count keeps all 69 rows, and our union is 80 with 46 shared.

The scoreboard

fingerprint.to returned the most rows on all three handles and was the fastest sweep in every file that records one, 20.4 to 20.7 seconds against OSINTsearch's 28.0 to 29.2 and Revealer's 30.4. OSINTsearch and Revealer land within two rows of each other on davidbombal and srhoe, the first hint of the overlap result below.

Two row types here are not profiles. fingerprint.to's RegisteredEmails row (all three handles) and its Discord row (davidbombal and srhoe) carry no profile URL. OSINTsearch's instagram_last_active row (srhoe morning run, deadcatx3) is a second Instagram row with an activity signal. Both stay in the counts because the tools count them.

The Revealer deadcatx3 export needs a warning label. Of its 28 rows, 17 are not profiles: 14 have platform names ending in _sitemap.xml and point at a site's sitemap, and 3 end in _f_20260510_rts-srb, a fragment of some other query that leaked into this one. The 11 real rows are Archive.org, Periscope, MySpace, Twitter, Telegram, Fragment, LevelBlue, Twitch, Minecraft (NameMC), Malpedia actors and Hugging Face. Every table below uses 11 for Revealer on deadcatx3.

Field richness

Rows carrying the fields an analyst reads first. davidbombal / srhoe (OSINTsearch 09:24 run) / deadcatx3.

The last three lines need care, because the wording matters.

Nobody found a former username on these handles. OSINTsearch's former-username row on all three is NameMC, and its raw.name_history holds exactly one entry: the current handle. fingerprint.to's NameMC row on davidbombal holds the same single entry. On srhoe, both tools return an osu profile with an empty former-username list (previous_usernames: [] in OSINTsearch's raw block, former_usernames: [] in fingerprint.to's extras). That matches what both tools promise: earlier names where sources still provide them. Our OSINTsearch review covers handles where those fields are populated.

Nobody found a ban either. Every is_banned, is_suspended, is_disabled, locked and is_deleted value in all nine files is false. What differs is how many rows carry the field at all. fingerprint.to's extras block carries an account-state key on 17 davidbombal rows (is_banned on Instagram, Roblox, Kick and TradingView, is_suspended on Twitter and SlideShare, is_disabled on Patreon and Rumble, is_locked, account_status, status and is_active on the rest), 16 on srhoe and 6 on deadcatx3. OSINTsearch's raw block carries one on 8 (Kick is_banned, Truth Social locked, Tellonym is_active, four status rows, Twitch deleted_at). Revealer's export schema has no such field. That is a real gap in ours.

On the shared fields, fingerprint.to leads on join dates, avatars and user IDs on every handle, by 6 to 13 rows on davidbombal. OSINTsearch and Revealer tie on bios (23 and 23 on davidbombal, 12 and 12 on srhoe), both ahead of fingerprint.to's 16 and 13. On deadcatx3, Revealer's Sep 4 file does not have the columns.

Platform overlap: OSINTsearch and Revealer see the same map

Normalise the platform names and compare.

OSINTsearch and Revealer agree on 87 percent of platforms on davidbombal and 90 percent on srhoe. In the meme, cia and richard benchmark, the two leaders sat at 0.50 Jaccard. Here, all 57 of Revealer's davidbombal platforms and all 35 of its srhoe platforms are found by OSINTsearch or fingerprint.to. Revealer contributed zero exclusive platforms on the two handles where it has a same-day export. On deadcatx3 its 11 real platforms are all in the OSINTsearch file too.

fingerprint.to is the engine that looks different. It contributed 18 platforms nobody else found on davidbombal and 17 on srhoe, and they are the big consumer networks: LinkedIn, Reddit, Discord, Bluesky, Patreon, Roblox, Rumble, Scribd, SlideShare and Fiverr on davidbombal; Facebook, LinkedIn, Reddit, Discord, Quora, Roblox, VSCO, Redbubble, Beacons and Bigo on srhoe.

Only OSINTsearch found, per handle:

• davidbombal: Internet Archive user contribution search, LevelBlue (AlienVault OTX), Periscope.
• srhoe: TikTok, plus the second Instagram row. The TikTok row supplied the username variant 5rh0e in the identity block.
• deadcatx3: Halo Infinite, NameMC, plus the second Instagram row. (Revealer's minecraft row on this handle is the same NameMC host.)

Only fingerprint.to found, top five per handle: Bluesky, Discord, LinkedIn, Patreon, Reddit on davidbombal; Facebook, LinkedIn, Reddit, Discord, Quora on srhoe; DeviantArt, Quizlet, Redbubble, Reddit, Roblox on deadcatx3.

Only Revealer found: nothing on davidbombal or srhoe; on deadcatx3 only the 17 artifact rows.

The honest reading: on small handles OSINTsearch and Revealer check a very similar list of endpoints, and fingerprint.to is the second opinion that adds coverage. OSINTsearch plus fingerprint.to reaches 80 platforms on davidbombal against 69 for the best single tool, a 16 percent gain, and 55 against 46 on srhoe, a 20 percent gain. Adding Revealer as a third on these handles adds bios and a breach lookup, not platforms.

What each export is for

OSINTsearch is the only one of the three with an identity block. On davidbombal it gives a probable name at 0.31 confidence, a probable location at 0.05, two username variants (davidbombal with a 58-platform source list and davidbombal.eth from ENS), two websites, 8 profile pictures and 8 profile links, and 59 platforms agreeing. On deadcatx3 it surfaced a second variant, nervequake, from Archive.org. The schema 3 export records found, not_found and could_not_check per run (59, 662 and 47 on davidbombal), the only per-run failure accounting in the set. Every row keeps a raw block with the untouched source payload, which is what makes the empty-list finding above checkable. Free plan: 15 searches a week, per osintsearch.org/pricing.

fingerprint.to is the widest single sweep on these handles, the fastest, and the richest on avatars, user IDs and join dates. Its extras block is populated on 67 of 69 davidbombal rows and carries the account-state fields, plus the platforms the other two miss. It has no identity summary and no raw payload; extras is already a normalised view.

Revealer.US ties OSINTsearch on bios and is the only export here with a breach section: 3 breach rows on davidbombal, 0 on srhoe, 0 stealer-log rows on either. Thin, on these handles, and the username scan added no platform the other two missed. The deadcatx3 file is the weakest export in this article, and it is ours. The breach and stealer side runs on data breach lookup and stealer logs, a different product from the username scan.

Revealer is not a consumer reporting agency. Its data may not be used for employment, tenant, credit, or insurance eligibility decisions.

Verdict, by dimension

No single winner. Here is the OSINT search result per dimension, on these three handles and nine files.

One tool on a small handle: fingerprint.to. Two: add OSINTsearch for the identity block and the raw payload, and because its union with fingerprint.to is 16 to 20 percent larger than either alone. Revealer's case on these handles is bios and the breach section, not platform reach, and its Sep 4 deadcatx3 export should not have shipped with sitemap rows in it.

Frequently asked questions

Is OSINTsearch better than fingerprint.to? Not on row count, speed or exclusive platforms. fingerprint.to returned 69, 46 and 22 rows against OSINTsearch's 59, 37 and 21, finished each sweep in about 20.5 seconds against about 29, and found LinkedIn, Reddit, Discord and Facebook where OSINTsearch did not. OSINTsearch leads on bios, the identity block and the raw payload per row, and has a free plan. They share 51 to 60 percent of their platforms, so run both.

Is OSINTsearch better than Revealer? On platform coverage the two are nearly the same tool on these handles: 0.871 and 0.895 Jaccard, and Revealer found nothing OSINTsearch or fingerprint.to missed. OSINTsearch adds the identity block, the raw payload, per-run failure counts and account-state fields; Revealer adds a breach section and matches OSINTsearch on bios. OSINTsearch wins on export depth.

Why does the OSINTsearch srhoe count show two numbers? We ran it twice on 2026-09-05, at 09:24 and 21:20, and got 37 and 31 rows. The evening run uses a newer export schema with an endpoint total and failure counts; six platforms in the morning file were absent in the evening one. Both are reported.

Why do your davidbombal numbers differ from OSINTsearch's own post? OSINTsearch's Sep 5 post reported 59, 67 and 57 with 47 of 77 shared. OSINTsearch counted fingerprint.to's 69 rows as 67 profiles, dropping the two rows with no profile URL; we keep all 69, and our normalised union is 80 with 46 shared by all three. The OSINTsearch and Revealer figures match exactly.

Did any tool find a former username or a banned account? No. The fields exist in the OSINTsearch and fingerprint.to exports and every one is empty or false on these handles. NameMC's name history holds only the current handle for both tools, and the osu former-username lists are empty in both. Those fields populate only where a platform publishes history.

Can I use these tools for background checks? Not Revealer. It is not a consumer reporting agency, and its data may not be used for employment, tenant, credit, or insurance eligibility decisions. Use an FCRA-compliant provider.