Cycling through all six. Tap any point to stop.

Measured on six things

Reverse photo identity check across public web pages

Security

Probe scored 1/13: HTTPS not enforced, no HSTS, CSP, frame protection, or secure cookies; DMARC missing. Claims 24-hour media retention but publishes no SOC 2, pen-test, or bug bounty.

Screenshots

1 / 1

SherlockSearch

About this app

I recently stumbled upon Sherlock Search and it's quickly become one of my go-to tools for verifying online identities. In a world where a single photo can end up in a dozen places, Sherlock Search lets you upload one face photo and then checks public web indexes to find possible matches. It's not about spying on private profiles; it's about uncovering where a face might appear on publicly accessible pages. The process is straightforward: you upload a clear JPG, PNG, or WebP, adjust the crop, confirm free access (no registration needed), and then it runs a visual comparison against public sources. You get a confidence score—like 91% Strong Match or 34% Unlikely—but the real value is that every match comes with a link to the original source. You can then inspect the context yourself: names, dates, captions, and so on. It's a huge help for anyone worried about impersonation or just curious about their digital footprint. The interface is clean and the privacy policy is reassuring: medi...

Key features

1Upload one face photo (JPG, PNG, WebP) and search public web indexes for possible visual matches — no registration required
2Confidence-scored results (Strong/Likely/Unlikely Match) with a source link on every match so you can inspect the original page
3Privacy framing: browser-side photo setup, up to 24-hour media retention, no card or subscription during free access

Target customer profile

Drawn from the product itself, not from a survey.

Demographic

Privacy-conscious individuals

Pain points

Worried about photo misuse or impersonation online

Primary needs

Quickly find where their face appears publicly

Demographic

Online daters

Pain points

Concerned about catfishing or fake profiles using their photos

Primary needs

Verify if someone is using their image on dating apps

Demographic

Social media users

Pain points

Want to check if their profile photo is reused elsewhere

Primary needs

Detect unauthorized use of their photos across platforms

Demographic

Journalists & OSINT researchers

Pain points

Need to verify identity of subjects in public images

Primary needs

Trace image origins and corroborate identities

Scout Review

Written by AI from measured evidence, scored out of 100.

SherlockSearch is a face-search tool that checks public web indexes for possible visual matches from a single uploaded photo, returning confidence scores and source links. The pitch is responsible: no registration, currently free, 24-hour media retention, and a repeated warning that a score is a lead, not a verdict. The homepage is well-crafted, the speed numbers are strong (mobile 87, desktop 97), and the accessibility audit is clean apart from one contrast failure. The problems are structural. Security scored 1/13 on the header probe — HTTPS not enforced, no CSP, no HSTS, no secure cookies — which is hard to square with asking users to upload their face. There's no help center, no third-party reviews, and no compliance signals. The category is crowded: PimEyes for facial search, Google and TinEye for reverse image, face-check.net as a direct peer. SherlockSearch's edge is honesty and presentation, not technology. It's a promising early-stage tool with a real security gap and a differentiation problem.

Measure by measure

Usability

No registration, currently free, and a four-stage flow spelled out on the homepage with imperative verbs. The FAQ covers six real questions, but there's no help center or docs beyond it.

Design

Dark navy hero with a purpose-built upload card, color-coded match badges, and a tight 12-platform grid. Repeated light-blue sections flatten hierarchy and OnlyFans/Tinder logos get equal weight to mainstream apps.

Speed

Lighthouse mobile 87/100 (LCP 3.9s, CLS 0, TBT 16ms, FCP 1.2s) and desktop 97/100. Layout is stable and low-blocking; the 3.9s mobile LCP is the one real drag.

Security

Probe scored 1/13: HTTPS not enforced, no HSTS, CSP, frame protection, or secure cookies; DMARC missing. Claims 24-hour media retention but publishes no SOC 2, pen-test, or bug bounty.

Accessibility

Lighthouse accessibility 96/100 with one failing audit for insufficient color contrast. Semantic headings and a responsive layout are present, but there's no WCAG statement or multi-language support.

Growth

Real problem — photo misuse and catfishing are growing — with a specific wedge: source-linked matches and confidence bands instead of raw image dumps. But PimEyes already owns facial search.

Improvement areas

Security headers are almost entirely absent (1/13 probe score): HTTPS not enforced, no HSTS, CSP, frame protection, or secure cookies
No help center, docs, or third-party reviews beyond the homepage FAQ — hard to verify the flow before uploading a face
Differentiation is presentation, not capability — PimEyes, Google, and TinEye already cover facial and reverse image search at scale

Design & Usability

SherlockSearch's homepage is genuinely well-built for an early-stage tool. The dark navy hero with a centered upload card feels purpose-built rather than templated, the 12-platform grid is tight, and the color-coded match badges (green 91%, yellow 68%, red 34%) make the report preview scannable at a glance. The four-stage process cards use imperative verbs that reduce ambiguity, and the FAQ answers six real questions about access, timing, and retention. The weak spots are visual: repeated light-blue section backgrounds flatten the hierarchy, and placing OnlyFans and Tinder logos at the same weight as LinkedIn and YouTube creates unintended emphasis. Usability is solid on the surface — no registration, currently free, clear CTAs in three places — but there's no help center, no docs, and no third-party reviews to confirm the flow holds up in practice. It's a competent, honest landing page that stops short of being memorable.

Speed & Security

Speed is the strongest measured dimension. Lighthouse mobile performance lands at 87/100 with CLS 0 and TBT 16ms, and desktop hits 97/100 — the layout is stable and the main thread stays clear. The one real drag is mobile LCP at 3.9s, which means the upload card takes a beat to arrive on a phone. Security is the opposite story. The HTTP probe scored 1/13: HTTPS is not enforced, and there's no HSTS, CSP, frame protection, X-Content-Type-Options, Referrer-Policy, Permissions-Policy, or secure cookies. SPF is present but DMARC is missing. For a tool that asks users to upload a face photo and claims 24-hour media retention, the absence of transport enforcement is a genuine problem, not just a missing badge. No SOC 2, ISO, pen-test, or bug bounty signals were found. The privacy copy is reassuring; the infrastructure behind it is not.

Accessibility & Growth

Accessibility scores well on automated checks: Lighthouse gives 96/100, with the single failing audit being insufficient background/foreground color contrast. That's a real issue because the color-coded confidence badges are the core of the report UI, so the one thing failing contrast is the thing carrying the meaning. There's no accessibility statement, no WCAG conformance claim, and no multi-language support, so the 96 is a floor, not a ceiling. On growth, the idea holds up: photo misuse, catfishing, and impersonation are real, growing problems, and the target segments — privacy-conscious individuals, online daters, OSINT researchers — are specific. The wedge is source-linked matches with confidence bands and an explicit 'score is not a verdict' charter, which is more responsible than raw reverse-image dumps. But PimEyes already owns facial search with a real index, Google and TinEye own reverse image search at scale, and face-check.net is a direct peer. SherlockSearch's differentiation is presentation and caution, not capability — a defensible niche, not a moat.

Conclusion

The honest read: SherlockSearch is a well-presented, fast, free face-search tool with a genuinely responsible framing and a security posture that doesn't match its own privacy copy. If you're curious about your digital footprint and want source links rather than a black-box score, it works and costs nothing. If you're uploading a photo you care about, the missing HTTPS enforcement and absent security headers should give you pause until they're fixed. The growth path is real but narrow — PimEyes and Google aren't going anywhere, and 'we show you the source' is a feature, not a moat. Fix the transport security, publish a real privacy/security page, and this becomes a credible niche tool. Leave it as-is and it's a nice demo with a screen door.

Suggestions for improvement

Enforce HTTPS site-wide and add HSTS, CSP, frame protection, and secure cookies — 1/13 on the security probe is the biggest fixable gap.
Publish a security/privacy page detailing media handling, retention, and deletion, plus a DMARC record to match the 24-hour retention claim.
Fix the color contrast on the confidence badges — they're the core of the report UI and the one failing Lighthouse audit.
Add a help center or docs beyond the FAQ, and seed third-party reviews so users can verify the flow before uploading a face photo.

How we score

Alternatives comparison

Named competitors, point by point. Nobody paid to appear here or to be left out.

	SherlockSearch	Google Reverse Image Search	TinEye	PimEyes	face check
Search method	Upload one face photo; checks public web indexes for possible visual matches	Upload image or paste URL; finds visually similar images and pages	Upload image; finds where it appears online with tracking over time	Face search across the web for your own or a public figure's face	Face-based search for public image matches
Confidence scoring	Yes — Strong/Likely/Unlikely Match bands (85-100, 50-84, 0-49)	No explicit confidence score; visual similarity ranking	No confidence score; exact and modified match results	Yes — returns matches with confidence scores	Not specified in collected evidence
Source link on every match	Yes — every match links back to the original public page	Yes — links to pages containing the image	Yes — links to pages where the image appears	Yes — links to source pages, some behind paywall	Not specified in collected evidence
Registration required	No — no registration, currently free	No — free without account	No for basic use; account for advanced features	Account required for full results; paid tiers	Not specified in collected evidence
Pricing model	Currently free; existing paid credits remain saved	Free	Free tier; paid plans for API and advanced search	Paid subscription; limited free preview	Not specified in collected evidence
Media retention / privacy	Up to 24-hour media retention; browser-side photo setup claimed	Google privacy policy applies; no explicit retention window stated	Privacy policy applies; no explicit retention window stated	Privacy policy applies; opt-out available for indexed faces	Not specified in collected evidence

Google Reverse Image Search

Google's reverse image search is a free tool that lets you upload an image or paste a URL to find visually similar images and pages that contain them. It's widely used but may not focus specifically on facial recognition.

TinEye

TinEye is a reverse image search engine that allows you to upload an image and find where it appears online. It's known for its accuracy and ability to track image usage over time.

PimEyes

PimEyes is a face search engine that lets you search for your own face or a public figure's face across the web. It's more focused on facial recognition and can return matches with confidence scores.

face check

Comparing options? See SherlockSearch alternatives, scored side by side

View all Development & Engineering