What the Mobile App Attribution Debugger Checks
The debugger has 29 tools in three groups. Eight read a specific type of payload, link or file and compare it, field by field, with the documented requirements. Eleven start from a problem you can see, such as low match quality or numbers that do not agree, and work back to the cause. Ten are utilities that help you plan, build and look things up.
Check a payload
- Event payload: AppsFlyer S2S, Adjust S2S (JSON or query string), Meta Conversions API or App Events API, GA4, Snapchat Conversions API, Branch, Singular and Google Ads app conversion requests. You can also paste a cURL command, a HAR export, a raw request from Charles, Proxyman or Fiddler, a Postman collection, or a fetch(), axios or Python requests call, and the payload is read out with any keys dropped. JSON copied from code, with comments or trailing commas, is read too. The payload type is detected automatically.
- SKAN postback: a raw SKAdNetwork postback or an AdAttributionKit postback as Apple sends it, including conversion values and postback windows. Apple’s signature is verified in your browser.
- Ad revenue: an AdMob, ironSource or AppLovin MAX revenue callback, or an AppsFlyer ad revenue event, including unit and currency problems.
- Attribution link: a OneLink, AppsFlyer, Adjust or Branch link, or a postback or callback template. Checks the parameters, limits, redirects, unfilled macros and shut-down Firebase Dynamic Links against the vendor documentation, for example a missing or invalid media source.
- Deep link files: checks the files iOS and Android read to open your app from a link: the apple-app-site-association file and the assetlinks.json file, including the app IDs and fingerprints.
- App config: your Info.plist, PrivacyInfo.xcprivacy, GoogleService-Info.plist, an iOS .entitlements file, AndroidManifest.xml, google-services.json, build.gradle, a Podfile or an Expo app.json. Checks SKAdNetwork and AdAttributionKit keys, the postback copy URLs, the tracking prompt text, Firebase consent defaults, Meta SDK keys, the AD_ID permission, Associated Domains and App Links against the Apple and Android documentation.
- Install referrer: a Google Play install referrer string, a Play Store link or the Install Referrer API values. Decodes it and checks encoding, click IDs and timestamps.
- Attribution result: AppsFlyer conversion data, the Adjust attribution object or the Apple Ads (AdServices) attribution record, explained field by field.
Diagnose a problem
- Match quality: scores the customer information fields in a Meta Conversions API event that affect Event Match Quality, such as missing identifiers or values that are not correctly hashed.
- Duplicates: compares the same event sent by two routes, for example the app SDK and your server, to show whether it will be deduplicated or counted twice.
- Privacy and consent: finds readable personal data in any payload and checks it against the consent signals inside.
- Why is it broken?: you pick the symptom, such as installs that look organic, an event that is not showing, or revenue that is wrong, paste any payload, and get the likely root causes with fixes.
- Error lookup: paste an API response or a log line from Meta, AppsFlyer, Adjust or GA4, including a GA4 validation answer or a success answer, and get what it means, what it does not prove, and the fix.
- Log timeline: paste several events at once and see order, duplicate and time window problems across them.
- SDK log: paste the device log (Xcode console or Logcat) with AppsFlyer, Adjust, Firebase, Meta or Branch debug logging on, and find the launch request, the server answer, and the install and event lines.
- Event mapping: check how AppsFlyer or Adjust events map to Meta and Google Ads, including Meta event names, the parameter that becomes the value, the currency and the Google Ads import rules.
- SKAN CV decoder: enter a conversion value from 0 to 63 and, optionally, your bit schema, and see what each field decodes to. It can also encode a value from your fields and check a SKAN 4 schema for overlaps and gaps.
- Number gap: enter installs or events as reported by the MMP, the ad network, SKAN and the store. The tool explains the usual reasons for each gap and which one to check first.
- Bulk check: many payloads at once, or a CSV export such as AppsFlyer raw data, each checked like Event payload, with the findings counted so you see which problem is most common.
More tools
- Compare payloads: two versions of one payload side by side, with every field that differs listed.
- Timestamp check: any timestamp read and measured against the documented Meta, Adjust and GA4 time windows.
- TCF consent: an IAB TCF v2 consent string decoded into purposes and vendors, with the spec checks.
- Hash values: SHA-256 for Meta user data fields, with the documented normalization, in your browser.
- ID check: tells you what an ID is (IDFA, GAID, AppsFlyer ID, app_instance_id, client_id, fbp or a Firebase app ID) and what an ATT status or an all-zero advertising ID means.
- Field glossary: what a field means, where it belongs, its source, and which rules check it.
- CV schema planner: turns your funnel events and revenue steps into a SKAN 4 or AdAttributionKit conversion value schema and checks it.
- SKAN ID list: merges your network IDs with Google’s published SKAdNetwork ID list, removes duplicates and gives you a block to paste into Info.plist.
- Postback builder: builds an AppsFlyer or Adjust postback template from each platform’s verified list of macros.
- Tracking plan: the actions you want to measure, in your own words, become event names for AppsFlyer, Firebase and GA4, and Meta, as a CSV.
Each finding points to the exact line in your input, explains the problem in plain language, and links to the page the rule comes from, with the date it was last checked. When the documentation does not state a rule outright, the finding says it is a judgment call.
What the Mobile App Attribution Debugger Cannot See
The debugger reads only what you paste. It cannot see:
- whether the request reached the endpoint, or what the endpoint returned
- SDK start-up order and timing inside your app, unless you paste the events into Log timeline or the device log into SDK log
- settings in your MMP dashboard, Meta Events Manager, Google Ads or App Store Connect
- how a platform decided to attribute an install
A failed check points to a rule your data breaks, so it is the right place to start. A clean result means something narrower: the input itself follows the rules. It does not prove that your tracking works end to end.
How to Use It
Using the mobile app attribution debugger takes a few seconds:
- Choose the check that matches what you have. If you are not sure, use Event payload and let the tool detect the type, or start from Why is it broken?
- Paste your input. For payloads, a test event works best, since it shows the same structure without real customer data.
- Press the button under the box and read the findings, starting with the ones marked FAIL.
- Use Share report, Save as PDF or Save as image to pass the result to your developer or team. The image shows finding titles only, never your payload values.
When the Fix Depends on Your Setup
Some findings are fixed by changing one field. Others depend on where the event is built: the app, the SDK configuration, the MMP, or a server-side pipeline. In those cases, the fix plan button at the end of the report requests a written fix plan that maps each finding to the exact place in your setup to change. For a full review of your whole setup, see the
mobile app attribution audit. Everything is handled asynchronously, in writing.
Related Guides
For the platform rules behind the checks, see Meta’s
Conversions API server event parameters and Apple’s
SKAdNetwork documentation.
Frequently Asked Questions
Why are my Meta iOS installs showing as organic in my MMP?
A common cause is that Meta attributed those installs through SKAdNetwork or Aggregated Event Measurement, which report in aggregate and are not matched to individual devices, so user-level MMP reports show them as organic. Other causes are a missing or misconfigured Meta integration in the MMP, and events sent without the identifiers Meta needs to match them. The payload checks can rule out the last cause. The first two need a look at your MMP and Meta settings.
Why is my SKAN conversion value null?
Apple withholds the fine conversion value when an install does not reach a high enough crowd anonymity tier. In SKAdNetwork 4, the first postback can carry a fine value only in the higher tiers, the second and third postbacks never carry a fine value, and at the lowest tier there is no second or third postback at all. A coarse value of “none” is not the same as null. Paste the postback into the SKAN check to see which case applies.
Why is my AppsFlyer S2S event not recorded?
The most common payload causes are a missing or wrong appsflyer_id, missing required fields, and timestamps or event values in a format the API does not accept. The app ID goes in the URL path (/inappevent/{app_id}), not in the request body, so a wrong or empty app ID in the URL is another common cause. Paste the full cURL command so the debugger can check the path too. If the payload passes these checks, the problem is often in authentication or in how the request is sent, which the debugger cannot see. If the API returned an error, paste it into Error lookup.
Why is my Meta Event Match Quality low?
Low Event Match Quality usually means Meta received too few customer information parameters, or received them without the required normalization and hashing, so it could not match the event to a person. The Match quality check lists which identifiers are present, which are missing, and which do not look correctly hashed.
Why do my MMP, ad network and store numbers not match?
Each source counts differently: different attribution windows, different definitions of an install, SKAN data that arrives late and in aggregate, and store counts that include users who never opened the app. Enter the numbers into Number gap to see the usual reasons for each gap and which one to check first.
Is my payload uploaded or stored anywhere?
No. The mobile app attribution debugger runs every check in your browser and stores nothing on any server. An optional list of your last 10 checks can be kept in your own browser, and it is off by default. If you choose to contact me from the fix plan page, a link to your report is added to your own email or WhatsApp message, and nothing is sent until you press send.
Does a clean result mean my tracking works?
No. A clean result from the mobile app attribution debugger means your input follows the documented rules. Delivery, SDK timing, dashboard settings and attribution decisions happen outside the payload and need to be checked separately.
Is this tool affiliated with Apple, Meta, Google, AppsFlyer or Adjust?
No. It is an independent tool. Product names are used only to describe which payloads it checks.
About the Author
Built by Al-Fahad Hridoy, an Attribution & Tracking Consultant working with AppsFlyer, Adjust, SKAdNetwork, Firebase, GA4, Meta Conversions API and server-side tracking. Every finding links to the source behind its rule and says when a rule is a judgment call rather than a documented requirement, so you can verify each one yourself.Apple, SKAdNetwork, AdAttributionKit, Meta, Google, GA4, Firebase, AppsFlyer, Adjust, Branch, Singular, Snapchat, AdMob, ironSource and AppLovin are trademarks of their respective owners. This is an independent tool and is not affiliated with or endorsed by any of them.