Attribution for game studios that buy players, not installs
See which campaigns and creatives bring players who finish the tutorial, come back on day one and make a first purchase. Map SKAN and AdAttributionKit values to progression, and filter click spam before a network gets paid for it.
What Jumbo Gaming changed with Linkrunner
Social, puzzle and learning game teams trust Linkrunner to measure installs to first purchase
Customers include Jumbo Gaming, Matiks, Kulfi, Virbhumi.
A first purchase is worth more than a thousand installs
Most players never pay, so the campaign that wins is the one whose installs convert to a first purchase, even a small one. Tutorial completion, D1 return and early level milestones arrive on day 0 and day 1, before most first purchases, so they are the signals you optimise toward while purchase data builds. For ad-monetised games, use a session or ad-view threshold you define, because Linkrunner does not see the ad revenue itself.
- Install
- 1tutorial_completeSame sessionFirst sign the player understood the core loop; the cleanest early optimisation signal
- 2level_5 / level_10 (or chapter_1 for stories)Day 0 to day 2Progression depth separates curious installs from players
- 3d1_return (session on day 1)Day 1Retention is the earliest predictor of payers in casual and puzzle games
- Optimise on thisfirst_purchase (with value via capturePayment)Day 0 to day 7 for most payersThe event that pays back the campaign; only completed payments count as revenue
- 5ad_view_threshold (for example 10 rewarded ads)Day 0 to day 3Stands in for ad-monetised value when ad revenue sits in your mediation platform
Why game attribution is harder than counting installs
SKAN squeezes your progression signal into 64 values
Most game value appears in the first few days, which is exactly what SKAN's first postback window (0 to 2 days) captures, but only as one fine value from 0 to 63. If tutorial, levels and first purchase are not ranked well, the postback tells you little. The same mapping choice applies whether you run SKAN or AdAttributionKit.
How SKAN mapping works →
SKAN 4 postbacks
Copies to Linkrunner
- Postback 1Day 0 to 242Fine value
- Postback 2Day 3 to 7MediumCoarse value
- Postback 3Day 8 to 35HighCoarse value
ROAS is only half the picture for ad-monetised games
Linkrunner tracks in-app purchase revenue at D0, D3, D7 and D30 per campaign, but it does not measure impression-level ad revenue. Hybrid games should read Linkrunner ROAS as purchase ROAS, and join ad revenue from their mediation platform using exported cohorts.
How revenue tracking works →- type
- FIRST_PAYMENT
- payment_id
- ord_88213
- user_id
- user_2041
Credited toIN_Prospecting_Sep+$4.99A winning creative dies in days
Game creatives fatigue fast, and network-level ROAS hides which video carries the campaign. The Creatives tab shows full metrics and D7 and D30 retention per Meta creative; Google reports at ad-group level, so Google creatives show content, status and Ad strength.
See creative reporting →| Creative | ROAS | |||||
|---|---|---|---|---|---|---|
Onboarding walkthrough, 15s Meta, video | Active | - | 4,812 | $0.92 | 38.4% | 1.84x |
Festive offer, square Meta, image | Ad set paused | - | 2,165 | $1.31 | 29.7% | 1.12x |
Feature tour, 5 cards Meta, carousel | Active | - | 1,408 | $1.07 | 33.2% | 1.46x |
App campaign, responsive  Google, videoAd strength Good | Active | Good | - | - | - | - |
Sample data. Google rows show Google's Ad strength rating, since Google reports metrics at ad-group level.
High volume attracts click spam
Casual campaigns run millions of clicks, which hides click flooding and click injection. Managed fraud rules run on every project from day one, and "Tag and withhold postbacks" stops fraudulent installs from training the network's algorithm or triggering affiliate payouts.
How Fraud Control works →SKAN 4 postbacksCopies to Linkrunner
- Postback 1Day 0 to 242Fine value
- Postback 2Day 3 to 7MediumCoarse value
- Postback 3Day 8 to 35HighCoarse value
Where game studios find paying players
- Meta App Promotion campaigns: fast creative testing at casual-game volume; send tutorial_complete and first_purchase through the Conversions API so Meta optimises toward players who progress.
- Google App campaigns: reach across Play, YouTube and AdMob inventory for puzzle and word games; import first_purchase as a conversion action.
- Apple Ads: search intent for puzzle, word and board games on iOS, matched through AdServices with no ATT prompt.
- TikTok: gameplay clips suit casual and story games; run through campaign links with ttclid and send events back to TikTok.
- ShareChat and Moj: Android-first, regional-language reach for ludo, card and casual games in India, with S2S postbacks.
How Jumbo Gaming tied every transaction to the campaign that brought the player
“Implementing Linkrunner was seamless. Their analytics have helped us make data-driven decisions that significantly improved our user acquisition strategy.”
Challenge
Jumbo runs social games across many campaigns and needed to see which ones brought players who transact, not just players who install.
What changed
Jumbo moved its campaign links and in-app transaction tracking to Linkrunner, so every transaction ties back to the campaign that acquired the player.
Outcome
30K+ transactions tracked to source, and an 18% uplift in click-to-install rate.
Frequently asked questions
Find the campaigns that bring paying players
Track tutorial, levels and first purchase per campaign and creative, with your first 25,000 attributed installs free.