Application security feature

Click and ad fraud

Otukan separates genuine ad views, clicks and installs from emulator farms, scripted taps and tampered clients in your Android and iOS app, and flags them.

  • Android
  • iOS
  • Unity
Faked ad views and clicks is met at the session evidence boundary before it reaches the appFaked ad views and clicksSession evidenceYour app
  • Real devices, real people

    Otukan tells a physical phone in someone's hand from emulators, device farms and scripts that replay taps for reward or payout.

  • The view from inside

    Otukan sees your app's own code, input and environment at the moment of each view or click, a view that network-side filtering does not have.

  • Evidence for ad operations

    Each rewarded view, click and first open can carry a session verdict your ad operations team reconciles with network reports.

Ad fraud behaviour Otukan sees inside your app

Select a behaviour to see the signal, the app's response and what your backend records.

Emulator view farms

  1. Detect

    Otukan recognises emulators and virtual devices running your app to collect rewarded views or generate installs at scale.

  2. Respond

    Reward grants pause for the session while ads can still be shown, so farmed views stop paying out.

  3. Report

    The install and each view carry an emulator verdict your team can match against payout reports.

Scripted taps

  1. Detect

    On Android, Otukan notices accessibility services and input tools producing taps without a person touching the screen.

  2. Respond

    Clicks from scripted input are marked and can be excluded from rewards and paid events.

  3. Report

    An automation signal travels with the click event to your backend.

Faked ad completions

  1. Detect

    Otukan detects hooks placed on the code paths that report an ad as watched or clicked inside your app.

  2. Respond

    The completion is not honoured in the client, and the reward waits for your server's confirmation.

  3. Report

    Your server receives the hooking signal alongside the reward claim.

Ad slots hijacked in clones

  1. Detect

    Integrity checks recognise repackaged copies of your app in which someone replaced your ad identifiers with their own.

  2. Respond

    The altered copy can be stopped at launch, so it cannot monetise your content.

  3. Report

    Your team sees the rebuilt build and where its sessions come from.

Masked network origin

  1. Detect

    Otukan flags proxy and VPN conditions often used to make farm traffic look like it comes from high-value regions.

  2. Respond

    Region-priced rewards and campaigns can wait for extra checks.

  3. Report

    The network risk signal is recorded with the session for reconciliation.

Controls behind ad fraud signals

Each control answers one question about a click or view: who made it, on what, and through which route.

  • Mobile bot detection

    Tells a person using your app from a script driving it or calling its API directly, per session.

  • Accessibility abuse detection

    On Android, spots services that read the screen and tap on a user's behalf without the user.

  • Overlay detection

    On Android, finds windows drawn over your app that can steer or disguise where a user actually taps.

  • Proxy and VPN risk

    Adds network context to each session, so traffic routed to fake a location stands out.

You choose the response

Set a response for each user flow. Start by observing, then tighten flow by flow as the signals prove themselves.

Observe

Signal sent to your backend

Rolling out a new policy and measuring real-world signals.

Warn

This device looks modified. Continue with care.

Conditions some genuine users create, such as developer settings.

Restrict

Payments are unavailable on this device.

Payments, stored data and account changes.

Terminate

Session closed for your security.

Clear tampering on a high-risk flow.

How ad fraud checks are rolled out

  1. Map paid events

    List each rewarded view, click, install and conversion that moves money in or out.

  2. Attach verdicts

    Otukan session evidence is added to those events within your existing build and release pipeline.

  3. Reconcile first

    Compare flagged traffic with ad network and campaign reports before changing any behaviour.

  4. Withhold payouts

    Rewards and paid events from flagged sessions are held, while normal use continues.

Free for app publishers

Check your app's exposure

A passive review of your published app by an Otukan engineer. No files, no access to your systems. Findings go by email to the verified publisher.

You finish the request on the next page.

Evaluate click and ad fraud protection by asking where each fraudulent view, click or install is produced, and whether the product can see that place. Otukan works inside your Android and iOS app, including Unity games. It detects emulator farms, scripted taps, hooks on ad completion code and repackaged copies that swap your ad identifiers, and it attaches a session verdict to each paid or rewarded event your backend records. It does not scan other apps on the device or audit your ad networks. Fraud that happens outside your app stays with your attribution and reconciliation process, which Otukan’s evidence makes sharper.

Game-wide revenue risks are covered on game and revenue protection.

Attack behaviour

Ad fraud has two victims. When you sell ad space, fake views and clicks inflate your reports until networks claw revenue back. When you buy installs, fake installs and conversions spend your budget on nobody. The journeys differ, but the behaviour inside the app is familiar.

BehaviourVictim journeyWhat the fraudster gainsWhat Otukan observes
Rewarded-view farmingRewarded ad watched on dozens of emulated devicesIn-game currency to resell, or payout from a partnerEmulator and virtual-device signals per session
Scripted clickingTaps generated by a service, not a fingerPaid clicks and completed offersInput arriving from automation on Android
Faked completionAd reported as finished without playingRewards with no ad shownHooks on your ad completion path
Clone monetisationYour app rebuilt with the fraudster’s ad identifiersYour audience earning for someone elseRebuilt package and foreign signing identity
Install farmsMany installs and first opens from reset devicesPayouts for installs no person madeEmulators, masked networks and device resets

The underlying patterns are covered on mobile ad fraud prevention.

Detection limits

A security team evaluating this area needs a clear line between what an app can see and what it cannot.

Visible from inside your app

Your app can check its own code and signing identity, how touches reach it, the hardware and system it runs on, and the network conditions of its session. On Android, it can also see accessibility services and overlays that interact with it. On iOS, the sandbox exposes less about the device, so Otukan relies on the app’s own process, loaded libraries and environment signals there. Each platform’s signals are designed and tested on their own devices, and an Android result is never assumed for iOS.

Outside the app’s view

Some fraud never touches your code. Another app can claim credit for an install it did not cause, and fraudsters can send fabricated events straight to an ad or attribution server. Confirming those cases takes your attribution partner’s data, server-side reconciliation and, for a live incident, investigation beyond one device. Otukan does not scan the whole device, and it does not replace those checks. It gives them an independent in-app view to compare against.

Response

Proportionate controls

Ad fraud responses should cost the fraudster money, not cost your users access. Good defaults are to keep serving ads, withhold rewards from flagged sessions, exclude flagged clicks and installs from payouts, and stop only rebuilt clones at launch. Test each response three ways: reproduce the farm or script in an authorised lab and confirm the signal, repeat the same journey on a clean supported phone and confirm nothing changes, then re-run install, update, sign-in and your rewarded-ad flow.

What to measure

Judge the rollout on money, not detections. Track rewards withheld, paid clicks and installs excluded from payouts, network clawbacks over time and support contacts about held rewards. The result to aim for is fewer clawbacks and lower acquisition cost per retained player, with support volume flat.

Support paths for legitimate users

Some flagged sessions will be people: a player on a PC emulator, a user whose assistive tool taps for them, a commuter on a corporate VPN. Tell them plainly why a reward is on hold and give support a way to release it. Otukan’s verdict names the signal behind each hold, so support can explain it rather than guess.

Otukan’s response options are covered on threat response.

Common questions

  • Can Otukan see click injection from other apps on the phone?

    Otukan works where your app has authority: its own process, how input reaches it and the device environment it runs in. Fraud carried out by another app, such as claiming attribution for an install it did not cause, or events spoofed straight to an ad server, is confirmed by your attribution partner and server-side reconciliation. Otukan's in-app evidence gives those checks an independent view to compare against.

  • Does this protect rewarded ads in our game?

    Yes. Rewarded views are where farms earn most, because every completion grants something of value. Otukan detects emulators, scripted input and hooks on the completion path, and the reward can wait for your server to confirm. Players who watch ads normally receive their reward as usual.

  • What happens to a genuine user on an emulator?

    Some people play on PC emulators for comfort or streaming. Otukan reports the emulator as a fact, and your policy decides what it affects. Many teams keep the game fully playable, pause only ad rewards or paid campaign events for those sessions, and show a clear message so support can answer questions.

  • How is this different from what our ad network already filters?

    Ad networks judge traffic from the network side, after the event. Otukan adds evidence from inside your app at the moment of the view or click: whether a person touched a real device, and whether the app itself was altered. Joining the two views catches cases neither side sees alone.

On the mobile engineering team?

How protection fits your build, signing and release, platform by platform.

Pay for people, not farms

Tell us the platforms, whether you earn from ads, buy installs or both, and where the numbers stop adding up. We never ask for ad account logins, builds, signing keys or user data.

Or request a technical demo