Cost Per Install (CPI) Calculator

Take the five inputs from the same campaign, the same date range and the same attribution window, usually from your ad network and your mobile measurement partner. Mixing a 7-day install count with 30-day spend is the most common way to get a CPI that is simply wrong.

Enter $1,000 total ad spend, 200,000 impressions, 2,000 clicks, 100 installs and 40 day-30 retained users. The headline result is CPI $10.00.

The funnel block explains where that $10 came from. CTR is 1.00% (2,000 ÷ 200,000), CPC is $0.50 (1,000 ÷ 2,000), click-to-install rate is 5.00% (100 ÷ 2,000) and the effective CPM is $5.00, the price of 1,000 impressions. These multiply back to the CPI: $0.50 per click divided by a 5% chance each click becomes an install is $10.00.

The retention block adds the quality check. Day-30 retention is 40.00% (40 ÷ 100) and cost per retained user is $25.00 (1,000 ÷ 40). The chart plots impressions, clicks, installs and retained users side by side so you can see which step loses the most people. Here it is the click-to-install step: 95% of people who clicked did not install, which points at the store listing or a mismatch between the ad's promise and the app.

Leave the retained-users field at 0 if you do not have retention data yet; the retention figures then show 0 rather than an estimate.

All inputs below zero are treated as zero, and any ratio whose denominator is zero returns 0 instead of an error.

• CPI = Spend ÷ Installs
• CTR = Clicks ÷ Impressions × 100
• CPC = Spend ÷ Clicks
• Click-to-Install Rate = Installs ÷ Clicks × 100
• Effective CPM = Spend ÷ Impressions × 1,000
• Day-30 Retention Rate = Retained Users ÷ Installs × 100
• Cost per Retained User = Spend ÷ Retained Users

Two identities are worth knowing because they show how the pieces connect. First, CPI = CPC ÷ click-to-install rate (as a decimal), and CPC itself is CPM ÷ (1,000 × CTR), so CPI is driven by three levers: what the network charges for attention, how often the ad gets clicked, and how often the click becomes an install. Second, Cost per Retained User = CPI ÷ retention rate. A 40% retention rate turns a $10 CPI into $25; a 10% rate would turn it into $100. That is why a cheap install with weak retention can be the most expensive user you buy.

Suppose a $500 test campaign delivers 60,000 impressions, 900 clicks, 180 installs and 27 users retained at day 30. The calculator returns CTR 1.50%, CPC $0.56, click-to-install 20.00%, CPM $8.33, CPI $2.78, retention 15.00% and cost per retained user $18.52.

Encouraged, the team scales the same campaign to $40,000, which delivers 5,000,000 impressions, 50,000 clicks, 8,000 installs and 800 retained users. The audience is broader and less interested: CTR falls to 1.00%, CPC rises to $0.80, click-to-install falls to 16.00% and retention falls to 10.00%, even though CPM is roughly unchanged at $8.00. CPI rises to $5.00 and cost per retained user rises to $50.00.

The headline CPI went up 80%, which a team might accept as the normal cost of scale. The cost per retained user went up 170%, from $18.52 to $50.00, because the weaker click-through, install rate and retention compound. Judging the scaled campaign on CPI alone would hide more than half of the deterioration, which is why small test results should be re-checked at each budget step rather than extrapolated.

CPI only becomes a decision when it is compared with lifetime value (LTV), the margin a user generates over their time in the app. The fair comparison is like for like: either cost per retained user vs LTV per retained user, or CPI vs LTV per install. Because installs that churn before day 30 generate little value, LTV per install is roughly retention rate × LTV per retained user, so the break-even CPI is retention rate × LTV per retained user.

Assume, purely for illustration, that a retained user is worth $40 in lifetime margin. In the test campaign, 15% × $40 = $6.00 of value per install against a CPI of $2.78, and $40 against a $18.52 cost per retained user is a ratio of 2.16 to 1. In the scaled campaign, 10% × $40 = $4.00 per install against a CPI of $5.00, a ratio of 0.80 to 1: every retained user costs more than they return.

Subscription businesses often use the guideline from David Skok's SaaS Metrics 2.0 that LTV should exceed 3 times acquisition cost and that acquisition cost should be recovered within about 12 months. Neither campaign above clears 3 to 1, a signal to improve retention or monetisation before adding budget. For ad-funded or one-off-purchase apps, estimate LTV from your own revenue per retained user rather than borrowing a subscription rule of thumb.

Two campaigns can reach an identical CPI for opposite reasons. Campaign A spends $1,000 for 200,000 impressions, 2,000 clicks and 100 installs: CTR 1.00%, CPM $5.00, CPI $10.00. Campaign B spends $1,000 for only 50,000 impressions but the same 2,000 clicks and 100 installs: CTR 4.00%, CPM $20.00, CPI $10.00.

Campaign A buys cheap impressions and gets few clicks, so the lever is the creative or the targeting. Campaign B's ad is clicked four times as often, but it is paying four times as much per thousand impressions, so the lever is the bid, placement or audience cost. Both share the same 5% click-to-install rate, so the store listing is not what separates them. Only the funnel breakdown tells you which fix applies.

The calculator divides the numbers you give it; it cannot check that they are consistent. Installs attributed by an ad network, installs counted by an app store and installs measured by an attribution partner often differ, and privacy frameworks such as Apple's App Tracking Transparency mean some installs cannot be attributed to a campaign at all. CPI will be only as accurate as the least reliable input.

It treats every install as paid. It does not separate organic installs that a campaign may have prompted, re-installs, or fraudulent installs. Retention uses a single day-30 count, so it cannot show the shape of a retention curve or distinguish a user who opened the app once on day 30 from a daily user. It does not compute LTV, return on ad spend or payback period; use the related tools for those, and treat any LTV you compare against as an estimate carrying its own uncertainty.

The store listing is often the weakest step, and the ASO & Conversion Calculator breaks that stage into impression-to-page-view and page-view-to-download rates. When the goal is a purchase or subscription rather than an install, the Marketing CPA Calculator measures cost per acquisition on that later event. If the app earns money from advertising, the Ad Revenue Calculator projects revenue from views, fill rate and RPM, a starting point for the value side of the CPI-vs-LTV comparison. The Paid Media Metrics guide explains how CPM, CPC, CTR and CPA fit together across channels.