What this calculator does
Cost per install is the average amount spent on advertising to generate one app install. It is the mobile equivalent of cost per acquisition for apps where the install itself, rather than a purchase, is the first measurable outcome of a campaign.
An install on its own is not a customer. CPI is most useful alongside a measure of what happens after install, such as how many users go on to register, subscribe or make an in-app purchase, since a cheap install that nobody opens again is worth very little.
The formula
Divide total ad spend by the number of installs it generated. The result is the average cost of one install over that spend.
| Term | Meaning |
|---|---|
| CPI | Cost per install: total spend ÷ number of installs. |
| Installs | The number of times the app was installed as a result of the campaign. |
| Spend | The total amount spent on the user-acquisition campaign. |
The inputs explained
| Field | What to enter |
|---|---|
| Total ad spend ($) | The total ad spend over the campaign or period being measured. |
| App installs | The number of app installs that spend generated. |
When to use it
Budgeting a user-acquisition campaign
Knowing CPI lets you estimate how many installs a proposed budget should deliver, before committing spend to a mobile campaign.
Comparing ad networks
Mobile ad networks and platforms often deliver installs at very different costs for the same targeting, and CPI is the standard way to compare them directly.
Judging install quality against cost
Combining CPI with a post-install metric, such as retention or in-app purchase rate, shows whether cheaper installs are actually worth acquiring.
Worked examples
Every figure in the tables below is produced by this page’s own calculator at build time, so the numbers and the tool always agree. Select any row to load that scenario.
How CPI changes as spend increases at a fixed number of installs
A fixed 400 installs, at a range of total spend levels.
| Total spend | CPI | Installs per $1,000 spend |
|---|---|---|
| $400 | $1.00 | 1,000.00 |
| $800 | $2.00 | 500.00 |
| $1,200 | $3.00 | 333.33 |
| $1,600 | $4.00 | 250.00 |
| $2,400 | $6.00 | 166.67 |
| $3,200 | $8.00 | 125.00 |
How CPI changes as installs increase at a fixed spend
A fixed $1,200 budget, across a range of install volumes.
| Installs delivered | CPI | Installs per $1,000 spend |
|---|---|---|
| 200 | $6.00 | 166.67 |
| 300 | $4.00 | 250.00 |
| 400 | $3.00 | 333.33 |
| 600 | $2.00 | 500.00 |
| 900 | $1.33 | 750.00 |
| 1,200 | $1.00 | 1,000.00 |
Questions
What is a good CPI?
It varies enormously by app category, platform, region and targeting, so there is no single number worth citing as a benchmark. Compare CPI against your own post-install value per user instead.
How is CPI different from CPA for apps?
CPI measures the cost of getting the app installed. CPA, applied to an app, typically measures the cost of a later action such as registration, subscription or first purchase, which requires the user to do more than just install.
Why do some installs cost more than others?
Cost per install reflects the competitiveness of the audience being targeted; users who are more likely to spend in-app, or who are harder to reach, generally cost more to acquire.
Should I always try to minimise CPI?
Not if it comes at the cost of install quality. Cheaper traffic sources sometimes produce installs from users who never open the app again, which can raise the effective cost per genuinely active user even as headline CPI falls.
To see the cost of a completed action further inside the app, treat that action as a conversion in the CPA calculator. For the impressions or clicks feeding the campaign, see the CPC calculator.