Apple Search Ads Custom Product Pages: How Free2move Cut CPA by 14–16%

Most Apple Search Ads accounts run every keyword against the same default App Store product page. That page was designed to convert everyone — which means it converts no one particularly well. Custom Product Pages (CPPs) let you show a different set of screenshots and messaging per ad, matched to what the user actually searched for.

We tested exactly that for Free2move (ex SHARE NOW), the carsharing app, in Germany — on brand keywords and on competitor keywords. Same ad groups, same bids, same 30-day window; the only variable was the product page. The CPP won both times, for different reasons.

The setup

In two ad groups (search results placement, Germany) — one on brand terms from returning users, one on competitor keywords — two ads ran side by side over the last 30 days:

• Default Ad — Free2move's standard App Store product page
• Custom product page — localized for the German market

The results on brand keywords: −14% CPA

Default page → custom product page:

• Tap-through rate (TTR): 36.97% → 39.65% (+7.2%)
• Conversion rate (CR)*: 146.64% → 158.41% (+8.0%)
• Avg. cost per tap (CPT): €2.53 → €2.35 (−7.1%)
• Avg. CPA: €1.72 → €1.48 (−14.0%)

*CR includes total downloads (new + redownloads). On a returning-users ad group, redownloads attributed on both taps and impressions push CR above 100% — that's expected, and it applies equally to both ads, so the comparison holds.

The CPP improved the funnel at both ends. More of the people who saw the ad tapped it (+7.2% TTR), and more of the people who tapped it installed (+8.0% CR). Compounded, that's what drives the 14% CPA reduction — same auction, same bid, cheaper installs.

What 14% lower CPA is worth

The percentage is the headline; the budget math is the point.

• At the default page's €1.72 CPA, €10,000 buys ~5,800 installs. At the CPP's €1.48, the same €10,000 buys ~6,750 installs — 16% more volume for identical spend.
• Shifting all of the ad group's traffic to the CPP-level CPA would have cut spend by ~10% over the period for the same install count.
• None of this required a bid change or a new keyword. It's a one-time asset that keeps paying on every impression.

CPA improvements from bid optimization are fragile — the auction moves and they evaporate. CPA improvements from conversion rate are structural: you're extracting more installs from the traffic you already pay for.

Why CPPs matter on non-brand keywords too

The instinct is to treat CPPs as a brand-campaign tool. The data argues the opposite.

On brand terms, users already know the app — the default page is "good enough," and the CPP still beat it by 14%. On non-brand keywords — generic terms and competitor names — the user has no relationship with your brand yet. The gap between a one-size-fits-all product page and one that mirrors their search intent is far larger, which means the CPP upside is larger too.

The same account proves it. In the competitor ad group — users searching for rival carsharing apps — the same German CPP ran against the default page over the same 30 days:

• Tap-through rate (TTR): 6.97% → 6.63% (~flat)
• Conversion rate (CR): 69.8% → 83.5% (+19.7%)
• Avg. cost per tap (CPT): €2.44 → €2.46 (~flat)
• Avg. CPA: €3.50 → €2.94 (−16.0%)

Look at where the win comes from. TTR and CPT barely moved — the ad renders similarly in search results, and the auction charged the same price per tap. The entire 16% CPA reduction happened after the tap: someone searching a competitor's name is actively comparing options, and a product page built for that comparison converted ~20% more of the taps into installs than the default page did.

On competitive terms you often can't out-bid your way to a cheaper tap — but you can out-convert everyone after the tap.

And notice the CPA gain on competitor keywords (−16%) was bigger than on brand (−14%), on users who didn't search for Free2move at all. Per €10,000 on competitor terms: ~2,850 installs at the default page's €3.50 CPA vs. ~3,400 at the CPP's €2.94 — 19% more volume for identical spend.

The structure that makes it work: 1 campaign, 1 ad group, 1 keyword

For your big keywords — the handful of generic terms that drive most non-brand volume — we recommend isolating each one in its own campaign: one campaign, one ad group, one keyword, one CPP. It sounds heavy. It's the opposite:

• A dedicated CPP per keyword. "Carsharing" and "rent a car" deserve different screenshots. Isolation is the only way each search intent gets its own page — inside a shared ad group, every keyword is stuck with the same creative.
• Full bid and budget control. Your biggest keyword no longer competes with twenty small ones for the same daily budget, and its bid reflects its own economics, not an ad-group average.
• Clean data, faster decisions. One keyword per campaign means the search term, the CPP, and the CPA all line up — no blended averages hiding a winner or a loser.
• Faster statistical significance. All the volume for that term flows through one A/B comparison (default vs. CPP) instead of being diluted across a keyword list.

Keep your long-tail and discovery keywords grouped as usual. Reserve the 1:1:1 structure for terms big enough to fund their own learnings.

How to replicate this test

1. Build one CPP per intent theme (brand/returning, top generic terms, competitor terms), localized per market — screenshots and promotional text matched to the search, not to a generic feature list.
2. In each ad group, run the CPP ad alongside the default ad. Don't change bids mid-test.
3. Judge the full funnel, not just TTR. On brand terms the CPP won at both ends; on competitor terms it won entirely on post-tap conversion rate while TTR stayed flat. A flat TTR doesn't mean the test failed.
4. Break out your biggest generic keywords into 1:1:1 campaigns with their dedicated CPP.
5. Refresh CPPs seasonally — creative that references the moment outperforms evergreen creative, and dated names keep iterations auditable.

The takeaway

A localized custom product page beat Free2move's default App Store page everywhere it ran: −14% CPA on brand keywords (winning on both tap-through and conversion rate) and −16% CPA on competitor keywords (winning entirely on post-tap conversion). Together that's roughly 16–19% more installs per euro — from a creative asset, not a bid change.

If you're spending on Apple Search Ads and every keyword still lands on your default product page, you're paying a relevance tax on every install. Scalebay manages Apple Search Ads for 70+ apps — if you want us to audit your CPP setup, get in touch.