← All posts
November 21, 2026

Working Backward From Target CPI to a Max CPT Bid: The Break-Even Math Indie Devs Skip

Apple Search AdsASOindie iOSunit economicsCPT biddingapp marketing

Most indie developers set a max CPT bid by feel. They start at a number that seems reasonable, watch what happens for a few days, and nudge it up or down based on whether taps show up. That works in a loose way, but it leaves you guessing about the one number that decides whether a keyword can pay for itself. There is a simpler approach: start from the install cost you can afford, then work backward to the most you should ever pay per tap.

The idea is small, but it changes how you read every keyword afterward.

Why start from installs, not taps

Apple Search Ads charges you per tap, not per install. A tap is only a step toward the outcome you care about, which is an install, and eventually a paying user. So the chain looks like this: you pay a cost per tap (CPT), some fraction of taps become installs, and the cost of one install is the cost per tap divided by that fraction.

That relationship is the whole trick. If you know how many taps it takes on average to get one install for a given keyword, and you know the most you are willing to pay per install, you can compute the highest CPT that still lands inside your limit.

The formula

Two definitions first:

  • Tap-to-install rate is installs divided by taps for the keyword or ad group. Some people call this conversion rate.
  • Target CPI is the highest cost per install you can accept and still make the campaign worthwhile.

The break-even ceiling for a max CPT bid is:

Max CPT ceiling = target CPI x tap-to-install rate

Think of it as a ceiling, not a goal. If you pay exactly the ceiling on every tap, your average cost per install lands right at the target. Pay less and you come in under it. Pay more and you are buying installs you have decided are too expensive.

A worked example

These numbers are purely illustrative, not benchmarks.

Suppose you have a utility app and decide that an install is worth paying up to $2.00 for, based on what a typical install is likely to become in revenue over the months you expect to keep it. You look at a keyword group and see a tap-to-install rate of 40 percent, meaning two out of five taps turn into installs.

Max CPT ceiling = $2.00 x 0.40 = $0.80

So a max CPT bid around $0.80 keeps your expected cost per install at or below $2.00 for that group. If the same group had a tap-to-install rate of 20 percent, the ceiling would drop to $0.40 for the same target. The keyword did not change. The page converted worse, or the searchers were less ready to install, and the affordable bid shrank with it.

Where the math comes from in practice

You will rarely have a clean tap-to-install rate on day one. Use what you have and be honest about the gap.

  • Start with a rate from a comparable keyword group in the same storefront, or from your own product page's overall tap-to-install rate as a rough stand-in.
  • Treat the first few days as a range, not a fact. A keyword with twelve taps tells you very little about its true rate.
  • Recalculate once you have enough taps that a single install does not swing the number. Wait for volume before you trust the division.

Choosing the target CPI

The hardest input is the target itself, and the formula cannot fill it in for you. Your target CPI should come from what an install is worth to you, not from what competitors seem to pay. If you do not yet know installs-to-revenue, estimate it conservatively and revisit it as cohort data arrives. Revenue attributed through the install-to-purchase chain, for example via RevenueCat, is the honest way to check whether your target was too generous.

A few guardrails help:

  • Do not set the target so high that the ceiling is above the real average CPT on that keyword. If the ceiling is far above what you actually pay, the math is not limiting you and the keyword is not really being tested against your economics.
  • Do not set it so low that the ceiling sits below the minimum bid Apple will accept for competitive terms. Then the keyword simply never serves, which teaches you nothing.
  • Remember that the bid is a maximum. Your actual CPT often comes in below it, so your realized CPI can beat the ceiling. That is the point of Apple's auction, not a bug.

What to do when the ceiling and reality disagree

When a keyword's actual CPT sits near or above its ceiling and it is still converting poorly, the answer is usually not a higher bid. Check three things in order:

  • Is the tap-to-install rate falling because of the product page, the screenshots, or the price shown on the page? Fix the page before you touch the bid.
  • Is the keyword too broad for the intent, pulling in searchers who tap out of curiosity? Look at the search term report.
  • Is the target itself wrong? If real revenue per install is lower than you assumed, lower the target and let the ceiling fall with it.

Only after those checks should you move the bid. Changing one thing at a time keeps the read clean.

Keep the habit small

You do not need a spreadsheet with a dozen tabs. Keep one line per ad group: the target CPI, the current tap-to-install rate, the resulting ceiling, and the bid you are currently running. Update it when you review the account. Over a few weeks, it becomes the fastest way to see which keywords are paying for themselves and which ones only look cheap on paper.

Taken together, the break-even ceiling gives you a principled reason for every bid, which is much easier to defend later than a guess. If you would rather have this recalculated for you against your own ASA and revenue data each day, with each change still waiting for your approval, that is the kind of work AdsBuddy is built to do. Either way, start from the cost of an install you can afford, and let the bid follow from it.

Run Apple Ads with AdsBuddy

Connect Apple Ads and RevenueCat, then get a short prioritized list of changes to review and apply. Plans from $29/mo.

Get started
Get new posts in your inbox
Practical Apple Search Ads tactics for indie iOS devs. No spam, unsubscribe anytime.