agentsclimarketplace
Sign in

Ad attribution

Skill rshankras/claude-code-apple-skills/skills/app-store/ad-attribution

Privacy-preserving ad measurement with AdAttributionKit (SKAdNetwork's successor) — install and re-engagement attribution, conversion-value strategy under crowd anonymity, and end-to-end postback testing. Use when running paid acquisition beyond Apple Ads, measuring re-engagement campaigns, designing conversion values, or migrating from SKAdNetwork.From its SKILL.md

Install
npx -y skills add rshankras/claude-code-apple-skills --skill ad-attribution

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

SKILL.md

5.9 KB, ~1.3k tokens by cl100k_base, as published. Nobody here has run it

Ad Attribution (AdAttributionKit)

You can't optimize paid traffic you can't measure. AdAttributionKit is Apple's privacy-preserving attribution framework — interoperable with SKAdNetwork, and the go-forward path (it also works in alternative marketplaces). This skill covers the advertised-app side: what an indie running paid campaigns must implement and how to design conversion signals that survive privacy thresholds.

Note the seam: Apple Ads attribution uses the AdServices framework, not AdAttributionKit — campaigns there are measured in the Apple Ads console (see app-store/apple-search-ads). This skill is for every other paid channel (ad networks, social, publishers).

When This Skill Activates

  • Running or planning paid user acquisition outside Apple Ads
  • "How do I know which campaign drove installs?" / measuring re-engagement ads
  • Designing or debugging conversion values / postbacks
  • Migrating from SKAdNetwork (fully interoperable — registered networks need no re-enrollment)

How it works (the 60-second model)

Ad network signs an ad (compact JWS impression) → publisher app shows it → user installs (or re-engages) → after a privacy-delayed measurement window, a postback with your conversion info goes to the ad network (and optionally to you). Attribution requires no user tracking; the price is coarser data, governed by crowd anonymity.

Conversion-value strategy (the part that's actually yours)

  • Two granularities: fine value 0–63 and coarse low/medium/high. Low crowd sizes downgrade fine → coarse and can trim the 4-digit campaign source-identifier to as few as its first 2 digits — so encode what matters most (campaign family, geo tier) in the first two digits, detail in the rest.
  • Map values to revenue-predictive early behavior (completed onboarding, trial start, first purchase), not vanity events. 64 slots go fast; reserve ranges (e.g. 0–15 activation, 16–47 monetization ladder, 48–63 reserved).
  • Update with updateConversionValue(_:) / PostbackUpdate; set lockPostback when the value is final to schedule transmission (still privacy-delayed, never instant).
  • Re-engagement postbacks are separate from install postbacks: update them independently via conversionTypes (.install / .reengagement); the first re-engagement update must land within 48 hours.
  • Multiple simultaneous re-engagement campaigns need conversion tags (iOS 18.4+): read the tag from the re-engagement URL's query items, store it, and pass it in PostbackUpdate — without a tag, updates hit only the most recent conversion.

Publisher-side mechanics (if your app also shows ads)

  • Click-through: UIEventAttributionView over the tappable ad + appImpression.handleTap() within 15 minutes of creating the impression.
  • View-through: beginView()/endView() on the same instance; ≥2 seconds on screen to count; one open view-through per network per advertised app.
  • SKOverlay / SKStoreProductViewController: attach the impression — they count view on present, click on tap.

Attribution windows & cooldowns (iOS 18.4+ configurability)

  • Defaults: 30 days click-through, 1 day view-through (install ads).
  • Override per ad network and interaction type in Info.plist (AdAttributionKitConfigurations → AttributionWindows); you can ignore "view" or "click" (not both) per network.
  • Set cooldowns (install-cooldown-hours, reengagement-cooldown-hours) so a re-engagement tap minutes after an install isn't double-attributed.
  • Country-level postback data (iOS 18.4): storefront country at install, gated behind an extra crowd-anonymity tier — a bonus signal when volume allows, never guaranteed.
  • Overlapping re-engagement conversions require opting in: EligibleForAdAttributionKitOverlappingConversions = YES in Info.plist.

Testing — never ship blind

  • Developer Mode → Settings → Developer → Ad Attribution Testing: removes time randomization, shortens all conversion windows, sends postbacks on demand — full end-to-end rehearsal in minutes instead of days.
  • Point development postbacks at a separate dev endpoint, never production: they're signed with a different key (kid), the network ID is always development.adattributionkit, and the advertised item ID may be 0 from Xcode.
  • ✅ Verify the postback pipeline before spending a dollar. ❌ Debugging attribution after a live campaign burns budget on unmeasurable installs.

Checklist

  • Conversion-value map documented (0–63 + coarse fallback + first-2-digit priority)
  • lockPostback fired once value is final; re-engagement first update < 48h
  • Conversion tags wired if >1 re-engagement campaign runs at once
  • Cooldowns configured; windows tuned per network where defaults don't fit
  • Dev-endpoint postback rehearsal done in Developer Mode
  • Campaign readout joined with growth/analytics-interpretation (source-type funnel) — attribution tells you which ad, App Analytics tells you what they did after

References

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Gives 0 of the 12 instructions most license compliance skills give in ~1.3k tokens

Counted across 99 of the 137 authors here whose files we hold, read 2026-09-06

  • Generate a structured attribution reportin 7 of 99, across 4 files
  • Include alternative hypotheses and false flag considerationsin 7 of 99, across 4 files
  • Collect evidence across all six attribution categoriesin 7 of 99, across 4 files
  • Use ATT&CK technique IDs for TTP comparisonin 7 of 99, across 4 files
  • Justify attribution confidence levelsin 6 of 99, across 3 files
  • Favor the hypothesis with the least inconsistent evidencein 6 of 99, across 3 files
  • Analyze infrastructure overlap between campaignsin 6 of 99, across 3 files
  • Score each evidence item as consistent, inconsistent, or neutral per hypothesisin 6 of 99, across 3 files
  • Override transitive versions when no direct fix existsin 5 of 99, across 2 files
  • Install the Snyk CLI and authenticate with SNYK_TOKENin 5 of 99, across 2 files
  • Scan package manifests and lockfiles in the CI/CD pipelinein 5 of 99, across 2 files
  • Preview fixes with snyk fix --dry-run before applyingin 5 of 99, across 2 files

Said here and by no other author read

  • Encode key campaign info in the first two digits
  • Map conversion values to revenue-predictive early behavior
  • Reserve conversion-value ranges across the 64 slots
  • Set lockPostback when the conversion value is final
  • Update install and re-engagement postbacks independently
  • Send the first re-engagement update within 48 hours

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 325,949. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.