Smart+ optimised for First Request: +205% SKAN installs, CPI −44%
Commentary byROASArchiveOur analysis, not theirsBitaksi ran a Smart+ campaign optimised for the First Request in-app event next to one optimised for installs, and the event campaign got 205% more SKAN installs at a 44% lower average CPI.
Both ran in Istanbul from August to October 2025.
- SKAN installs: +205% more for the event-optimised campaign, a large volume gain.
- Average CPI: 44% lower, so installs got cheaper as well as more plentiful.
- Monthly installs: 3x more per month, per the source.
- No spend, install counts or First Request numbers are shared.
The change
Before
A Smart+ campaign optimised for app installs, targeting all ages and genders in Istanbul.
After
A parallel Smart+ campaign optimised for an in-app event, First Request, using Bitaksi's own UGC-style creative.
Our reading: why it likely worked
Our interpretation, not the advertiser's claim.
- Optimising for a deeper event gives TikTok a clearer signal of a valuable user, and those users still install.
- SKAN reporting is sparse, so a stronger event signal may help delivery more than install signals on iOS.
- It is odd that a deeper event lowered CPI; budget split or creative differences between the two campaigns may explain part of it.
Caveats: TikTok published this about its own Smart+ product, so it is vendor-reported. The two campaigns ran at the same time, but the source doesn't say budgets were equal or users were split, and no counts or spend are given.
How to run this in your account
- Hypothesis
- If we optimise an app campaign for a key in-app event instead of installs, then CPI and install volume will improve because delivery targets users who act after installing.
- Primary metric
- CPI
- Guardrails
- Cost per in-app event
- Event volume per week
- Budget split between campaigns
- Who should try it
- Apps with a frequent early event, like a first ride or first order, and enough volume for event optimisation. Less useful if the event is rare or slow to happen.
Who got the result
TikTok for Business · Bitaksi published this result (Platform success story). The numbers are theirs; the summary and commentary on this page are ours.
Questions
Should TikTok app campaigns optimise for installs or in-app events?
Bitaksi's event-optimised Smart+ campaign got 205% more SKAN installs at a 44% lower average CPI than the install-optimised one.
It is vendor-reported and may not hold for apps with rare events.
Can event optimisation lower CPI?
In this case it did, by 44% on average.
Normally you'd expect higher CPI for deeper events, so test it side by side.