Skip to main content
DigeeSell logo

Car buying and selling app, UAE · United Arab Emirates

iOS installs cost 5.8 times more than Android on the same campaigns

5.8x
iOS cost per install versus Android
$0.93
Android cost per install
$5.39
iOS cost per install
40.9%
Registrations reaching activation

The starting point

The app was being judged on cost per install, which is the metric app campaigns hand you first and the one that matters least. Installs were arriving. What nobody was measuring was how many of those installs became users who did anything.

Measured — full-funnel export

App install campaign

Roughly $10,000 of media, tracked from install through registration to a deep in-app action. The platform split is stark, and it is not a creative difference — the same campaigns ran across both in the same period.

$1.70
Blended cost per installmeasured
$0.93
Android cost per installmeasured
$5.39
iOS cost per installmeasured
$4.43
Cost per registrationmeasured
$10.84
Cost per activated usermeasured
~$10,000
Total mediameasured

Where the money actually goes

The funnel underneath

Cost per install was $1.70. Cost per user who actually did something was $10.84 — six times higher. Two drop-offs explain the gap, and the second one is the expensive one.

38.4%
Installs that reached registrationmeasured
40.9%
Registrations that reached activationmeasured
6.4x
Cost per activated user versus cost per installmeasured

What actually drove it

Weight budget to the platform you can actually measure

Since Apple's tracking transparency changes, iOS attribution is delayed and aggregated. Android reports cleanly. Weighting acquisition budget to Android buys measurable performance; iOS is then judged on blended lift rather than on a cost per install nobody can verify.

Stop optimising on cost per install

It is the cheapest number to move and the least connected to revenue. An install that never registers costs the same as one that does. Buying against activation changes which audiences and placements survive.

Find the second drop-off, not the first

Everyone looks at install to registration. Here the larger loss was registration to activation — only two in five registered users reached the action that made them worth acquiring. That is a product and onboarding problem that media spend cannot fix.

Report the whole funnel to the client, every time

A dashboard that stops at installs will always look better than one that continues to activation. Showing both is what makes the budget conversation honest.

How these numbers were produced, including where they are estimates

  • Figures come from the campaign's own funnel reporting for the period stated.
  • Costs are in United States dollars, as billed by the ad platform, for a client operating in the United Arab Emirates.
  • This is a single reporting period, presented as a snapshot rather than a trend. No claim is made about performance direction over time.
  • Cost per activated user reflects the campaign's own definition of a deep in-app action, not a purchase. No revenue or return on ad spend figure is claimed.
  • Client name withheld.
All case studies