Cost per install

Cost per install (CPI) is the average price paid to acquire one new app installation through paid marketing channels. If a campaign spends five thousand pounds and generates two thousand installs, the CPI is two pounds fifty.

CPI is the mobile equivalent of cost per acquisition for web businesses, but with an important distinction: an install is not a customer. Users who install an app may never open it, may use it once and abandon it, or may never convert to a paying user. This makes CPI a necessary but insufficient measure of acquisition efficiency. It must be evaluated alongside post-install metrics like activation rate, retention, and lifetime value.

Despite this limitation, CPI remains the primary unit of measurement for mobile user acquisition because it is the point at which attribution is most reliable. Mobile measurement partners (MMPs) like Adjust, AppsFlyer, and Branch can attribute an install to a specific campaign, ad creative, and channel with high confidence. Post-install events are harder to attribute cleanly, making CPI the anchor metric for campaign-level optimisation.

CPI also directly determines the scale of growth. Given a fixed budget, lower CPI means more installs, which means more users entering the funnel. For a subscription app that converts 5% of installs to paid subscribers, reducing CPI from three pounds to two pounds increases the number of subscribers acquired by 50% for the same spend.

Divide total campaign spend by total attributed installs. The calculation is straightforward, but accurate attribution is not.

Install attribution relies on matching a user who clicked an ad to the same user who subsequently installed the app. This matching is performed by MMPs using a combination of device identifiers, probabilistic fingerprinting, and self-attributing network (SAN) reporting from platforms like Google, Meta, and Apple Search Ads.

Since Apple introduced App Tracking Transparency (ATT) in iOS 14.5, deterministic attribution on iOS has become significantly harder. Many users opt out of tracking, which means a portion of installs cannot be attributed to specific campaigns. This creates a measurement gap where reported CPI may be higher than actual CPI because some campaign-driven installs are counted as organic.

To account for this, many teams track both attributed CPI and blended CPI. Attributed CPI divides spend by attributed installs only. Blended CPI divides spend by the incremental increase in total installs (both attributed and organic) during the campaign period, giving a more realistic view of true cost efficiency.

CPI sits between ad spend and the post-install funnel. In a metric tree, it decomposes upward into the factors that determine how much each install costs, and connects downward to the metrics that determine how much each install is worth.

The tree reveals two distinct paths to improving acquisition economics. The left branch reduces the cost of each install through better creative, smarter targeting, and higher app store conversion. The right branch increases the value of each install through better onboarding, activation, and monetisation. The most effective growth teams work both sides simultaneously.

The tree also surfaces a key insight: app store listing conversion rate affects CPI even for paid campaigns. Many ad formats drive users to the app store listing page, where they decide whether to install. A higher-converting listing page reduces the number of clicks (and cost) needed per install.

CPI varies enormously by platform, geography, app category, and ad format. iOS CPIs are typically higher than Android due to smaller inventory and higher user quality. Developed markets cost significantly more than emerging markets. Gaming apps have a wide CPI range depending on genre.

1. 1

   Invest in ad creative quality and testing

   Creative is the single largest lever for CPI. High-performing video ads, playable ads, and interactive formats generate higher click-through rates, which reduce the cost of each click and therefore each install. Test new creatives weekly and retire fatigued assets before performance degrades.

2. 2

   Optimise the app store listing for conversion

   A higher conversion rate on the app store listing page means more installs per click. Test screenshots, app preview videos, icon design, and the first few lines of the description. Even a 10% improvement in listing conversion directly reduces CPI by 10%.

3. 3

   Diversify acquisition channels

   Relying on a single channel (typically Meta or Google) concentrates spend and increases competition for the same audience. Explore emerging channels like TikTok, Apple Search Ads, connected TV, and influencer partnerships where competition may be lower and CPIs more favourable.

4. 4

   Refine targeting and use exclusions

   Exclude audiences that install but never activate. Suppress users who already have the app installed. Focus on lookalike audiences modelled on high-LTV users rather than broad reach. Every install from a user who never opens the app inflates effective CPI.

5. 5

   Leverage organic growth to reduce blended CPI

   Invest in ASO, referral programmes, and viral loops to increase organic installs. Since organic installs are free, a higher organic percentage reduces blended CPI even if paid CPI remains constant. The strongest mobile businesses generate 60% to 80% of installs organically.