Custom Product Pages: matching the listing to the ad that earned the tap

Someone taps an ad about your app's offline mode, lands on your product page, and the first screenshot is about collaboration. They bounce, and you paid for that tap.

That gap is what Custom Product Pages exist to close.

What a CPP actually is

An additional App Store product page for the same app, with its own URL, its own screenshots, its own app previews and its own promotional text. The app name, subtitle, icon and description stay the same — those belong to the default page.

You send traffic to it deliberately: Apple Search Ads campaigns, your own paid social, an email, a partner link. It does not replace your default page and it does not compete with it in search.

This is different from Product Page Optimization, which is an experiment to find a better default. A CPP is a permanent destination. People conflate them constantly.

	PPO	CPP
Purpose	Find a better default page	Match a page to an audience
Lifetime	Runs, concludes, ends	Permanent
Traffic	Split from existing	Sent by a link you control
Own URL	No	Yes

The one principle: message match

Whatever earned the tap should be the first thing on the page.

If the ad said "works offline", panel one is offline. If the keyword was "receipt scanner", panel one is scanning a receipt. If the referral came from a privacy-focused newsletter, panel one is about privacy.

This sounds obvious and is routinely not done, because the default page was written for everyone and nobody revisits it per campaign.

Splits that are worth it

Apple allows far more of these than anyone should build. The practical limit is how many audiences you can write genuinely different screenshots for — usually three or four.

By feature intent. Your app does five things; people search for one. A page per major use case, leading with that use case.

By campaign creative. The ad's visual and claim echoed in the first screenshot, so the page feels like a continuation rather than a different product.

By audience. Someone evaluating for their team needs different proof from someone evaluating for themselves — sharing and permissions versus speed and simplicity.

By seasonality. A page for a specific moment, retired afterwards.

Splits that are not worth it

Rewording the captions. If the only difference is phrasing, you have made maintenance for nothing. That is a PPO test, not a CPP.

One per ad group. Twenty near-identical pages is twenty things to update every release, and nobody updates them.

Anything you cannot maintain. This is the real constraint. Every CPP is a full screenshot set that goes stale when your UI changes. A page showing a two-version-old interface is worse than no page.

Measuring it honestly

Apple reports impressions and conversion per CPP. Two cautions:

Compare like with like. A CPP fed by a tightly targeted campaign will out-convert your default page regardless of its screenshots, because the traffic is better qualified. That is not evidence your screenshots worked. The honest comparison is the same campaign pointed at the default page versus pointed at the CPP.

Give it real volume. The same statistical discipline applies here. A CPP with two hundred impressions tells you nothing.

Producing the sets

The reason CPPs stay a slide in a strategy deck rather than a thing teams do: each one is another full screenshot set at every size, and if a set is a multi-day design job then four pages is a month.

The workable model is forking. Start from your default set, change the first one or two panels to match the campaign's promise, re-render only those. Everything else stays identical — which is also correct editorially, since the pages should feel like the same product.