Online Advertising
Cheap installs are easy to buy and mostly worthless. The number that decides your budget sits 30 days later.
Buying app installs is straightforward. Buying app installs that turn into users is a different exercise, and the gap between the two consumes most app marketing budgets.
Typical figures across categories: around 25 per cent of installs open the app a second time on day one; roughly 10 per cent are still active on day 7; and 3 to 5 per cent on day 30.
Those are averages of functioning apps. Buy installs on price alone and day-30 retention frequently falls below one per cent, which means an effective cost per real user a hundred times the headline cost per install.
Every major platform allows optimisation on post-install events. Use it.
| Optimisation target | Result |
|---|---|
| Install | many cheap, mostly dead installs |
| Registration | far better, still early |
| First meaningful action | best balance for most apps |
| Purchase | best quality, needs volume to learn |
The constraint is data volume. Optimising on purchase needs roughly 30 to 50 events a week to train properly. Below that, step back to an earlier event.
Between the ad click and the install sits the store page, where a large share of interest is lost. Its conversion rate typically runs 20 to 35 per cent, and it can be improved considerably.
Three things carry most of the effect: the first two screenshots, which are all most people see; the first three words of the subtitle; and the rating, where the step from 3.8 to 4.4 stars measurably lowers cost per install.
Improving the listing conversion rate from 25 to 35 per cent cuts cost per install by 29 per cent across every channel at once. No bidding change achieves that.
App install advertising has the industry's highest fraud rate. The common forms are click injection, where a click is faked just before an organic install, and install farms.
Two protections matter: use an attribution provider with fraud detection, and check the time between click and install. A distribution clustered under ten seconds is not human behaviour.
Set one figure: revenue per install over 90 days, measured by source. Everything else is diagnostic. If a channel delivers installs at £2 and £1.40 of 90-day revenue, it is losing money regardless of how good the cost per install looks.
Games £1 to £4, shopping apps £2 to £8, finance and insurance £8 to £30. The figure alone says nothing without a retention rate.
The earliest event that correlates with revenue, usually registration or first meaningful use within 24 hours.
More from Online Advertising
Online Advertising
Cost per click is the most widely misread number in online marketing. Look only at it and you will pay more per customer than the competitor with the pricier click.
Online Advertising
Most accounts grow wild. After two years nobody remembers why a given ad group exists. A structure that is still readable after a hundred changes.
Online Advertising
A campaign type that plays across every channel and shows you almost nothing. For some firms it is the strongest lever available; for others an expensive trap.