Score every visitor in 5 ms.
Block the invalid ones before the ad fires.
PubSentry is the AI Security Engineer for ad publishers. It judges every visitor in real time and, if they’re invalid, the ad never serves — so the bad impression or click never reaches Google, and your AdSense account never sees the spike.
prebid
prebidInvalid traffic doesn’t just waste money.
It gets accounts banned.
Bots, click-bombs and policy violations drain your revenue and put your AdSense account one strike from suspension. Most tools report the damage after it’s on your account. PubSentry stops it at the door.
Account suspension
One invalid-click spike can flag your AdSense account. PubSentry intercepts abusive clicks before they ever reach the network — the spike never happens.
Wasted spend & clawbacks
Invalid impressions and clicks erode RPM and trigger clawbacks. Every block is framed as the revenue it protected — a dollar twin on every metric.
Sophisticated bots
Residential-proxy mimicry slips past simple filters. The cross-site reputation network and the ML moat close the gap as they scale — and we’re honest about today’s boundary.
The only honest place to stop fraud is before the ad fires.
A local gate scores every visitor in under 5 ms — before any ad loads. Invalid ones are suppressed across GPT, AdSense, Amazon and Prebid, with a universal fallback. A blocked impression was never served; a blocked click was never billed.
- Under 5 ms, locally — no network round-trip in the critical path
- Suppressed before the slot paints — no captcha, no reader friction
- Fails open — any error serves the ad, so the page is never harmed
It reads you the answer, then shows the data.
Most tools dump you into a table and make you assemble “am I okay?” yourself. PubSentry opens with a written brief from your AI Security Engineer — what it scored overnight, what it blocked, the revenue it protected, and the one thing that needs your eye.
- A narrated daily “Today” brief, signed by the Engineer
- Recommend-then-apply: it proposes a counter-rule, you approve it
- Every number links back to the one request that produced it
A renewal report your boss will understand.
Sentry counts errors. PubSentry counts dollars. Every metric carries its money twin, and the Reports screen turns a month of protection into a board-ready proof of value — revenue protected, account kept clean, zero readers challenged.
- Money-protected framing on every metric (estimated at your real RPM)
- A composite “how close am I to a ban?” Account-Safety score
- The FPR = 0 covenant — 0 readers challenged, front and center
Built like a security product, not a counter.
Sentry is to code what PubSentry is to ad revenue — except Sentry counts errors and we count dollars. Six things a single-tenant reporting tool structurally can’t copy.
Block-before-serve
The verdict happens before the ad loads — the only honest place to stop fraud. Suppression across every major ad stack.
One detection engine
The same scoring engine runs in the tag, the edge, and the server — a parity test fails the build if they diverge. Rules decide; ML only narrates.
Verdict-as-spine
Open any request and read its decision trace: the signal vector, each weight, the reputation context. Auditable like a stack trace.
FPR = 0, by covenant
Blocking a real reader is the worst failure. Conservative defaults, a calibration gate in CI, and a blast-radius preview before any rule goes live.
Reputation network
An entity flagged on any PubSentry site is pre-flagged on yours — a cross-site signal a single-tenant tool can’t build.
Radical honesty
No fake scores, no “100%”, an “awaiting traffic” state when there’s no data. We publish exactly what we can and can’t catch.
The same brain in the tag, the edge, and the server.
No drift between “what the browser thinks” and “what the server thinks.” A parity test fails the build if the three diverge — so a verdict means the same thing everywhere, and the LLM never invents a number or runs in the gate.
Flagged once, blocked everywhere.
When a fingerprint or IP turns invalid on any PubSentry site, it’s pre-flagged on yours before it costs you a thing. The network gets smarter with every publisher — the kind of moat a single-tenant plugin can never build.
Put a security engineer on your ad stack.
One tag. Live verdicts in minutes. Start free with 500 pageviews — no card — and only pick a plan once your real traffic is scored.