Merge Gardens: keep a later receipt-validation change out of an earlier result
Futureplay's public case contains two useful milestones with different dates. Keeping them separate prevents a later data-quality improvement from becoming the explanation for an earlier campaign result.
Futureplay's public case contains two useful milestones with different dates. Keeping them separate prevents a later data-quality improvement from becoming the explanation for an earlier campaign result.
What the public case establishes
AppsFlyer's Futureplay and Mistplay case reports Merge Gardens acquisition outcomes for January 2026. It separately says receipt validation rolled out on iOS and Android in March 2026, with cleaner purchase signals afterwards. The account also describes channel and geographic expansion supported by attribution and revenue measurement. AppsFlyer case study.
This debrief analyses a public customer case published by a commercial supplier. It is not an original UA Ledger interview. The chronology below follows the supplier’s account; we did not inspect the advertiser’s dashboards or interview its staff.
The decision worth examining
The sequence matters: a March intervention cannot cause a January outcome. Both milestones may belong in a successful business relationship, but they answer different questions. The earlier result concerns campaign expansion against a target, while the later change concerns the quality of events entering measurement and optimisation. A debrief should preserve that causal boundary even when the headline joins them.
The proposed chronology gives every intervention its own effective date, affected population and result window. Target-relative performance is recorded separately from performance relative to another campaign. Exceeding a goal does not reveal the absolute return or whether the goal was demanding, so the target definition must accompany any decision based on the reported margin.
A usable next check
Build a dated intervention ledger for a current expansion project. Match every claimed outcome to changes already active during the relevant period, and identify later changes that cannot explain it. For receipt validation, replay accepted and rejected events and compare resulting signals separately. Keep geographic target attainment in its own analysis rather than using the validation exercise to certify it.
| Test element | Proposed specification |
|---|---|
| Control | Narrative combines all partnership milestones into one outcome |
| Variant | Each intervention is matched to its eligible time window and population |
| Readout | Chronology consistency and separate target-relative versus absolute outcomes |
| Stop or reject | Reject causal explanations whose intervention occurs after the result window |
Download the populated case worksheet. Its control, variant and decision rules are a proposed experiment, not an account of results already measured. Keep the source URL with the worksheet when circulating it so reported findings cannot become unattributed team benchmarks.
Evidence boundary
All advertiser outcomes remain supplier-reported. We did not inspect receipts, targets, budgets or campaign allocation. The case does not establish a universal benefit from rewarded traffic or a verified return forecast for another game.
For background, see reading an mmp dashboard without fooling yourself and cohort quality checklist ua spend. No first-hand campaign result is asserted.
