Bermuda Adventures: a retention case is not proof of a winning creative
Belka's Bermuda Adventures story is often easier to read as a creative success than its evidence permits. It is chiefly a useful warning to distinguish the promise shown from the audience objective used.
Belka's Bermuda Adventures story is often easier to read as a creative success than its evidence permits. It is chiefly a useful warning to distinguish the promise shown from the audience objective used.
What the public case establishes
TikTok's Belka Games case describes Bermuda Adventures, a farming simulation with advertising and in-app purchases. It says Pangle retention optimisation in Japan and Korea was combined with purchase and install campaigns. Its reported comparison is against install-only optimisation, not against an identified alternative video. TikTok for Business case study.
This is a source-led creative casefile, not a frame-by-frame viewing or a campaign we ran. The diagram below is our own schematic. It does not reproduce the advertiser’s footage, and we have not established reuse permission for that footage.
The decision worth examining
The ad team needs to know whether a farming promise attracts players who want to return, but this source cannot answer that creative question by itself. Changing the delivery objective can change the selected audience even when the file remains identical. A production brief that asks for more of the winning creative without naming the file has confused a media result with an asset diagnosis.
Use the case as a briefing boundary. The source describes customisable characters and farming, so those are legitimate product themes to investigate. It does not supply a viewed sequence, timing map or asset-level retention breakdown. Our schematic therefore maps the missing relationship between theme, delivered audience and return behaviour rather than inventing an ad that supposedly generated the results.
A usable next check
First obtain authorised campaign files and map each to a stable creative identifier. Compare the same creative across declared optimisation objectives, and compare different themes within an unchanged objective where assignment permits. Keep Japan and Korea separate in the first readout. If those cells are absent, report the missing comparison instead of assigning the campaign-level gain to the visual treatment.
| Test element | Proposed specification |
|---|---|
| Control | Same identified asset under install optimisation |
| Variant | Same asset under a declared return objective |
| Readout | Matched cohort return behaviour with market and media conditions recorded |
| Stop or reject | No creative verdict when the asset ID or comparison cell is missing |
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
This note is a source-led casefile with no original footage inspection. Vendor-reported improvements are not independent estimates; the public source does not isolate creative, disclose cohort counts or prove current product availability.
For background, see the creative brief is a hypothesis document and post campaign creative debrief process. No first-hand campaign result is asserted.
