When Viktor Orlov stepped on stage at Business of Apps Berlin 2025, he opened with a small admission. Two years earlier, in the same room, his workshop had tried to explain too much. Too much theory. Too much showing what he knew about Apple’s systems, not enough guidance on what practitioners could actually do differently the next morning.
This time, the ambition was narrower and more practical. Apple Search Ads, Orlov argued, is not a mysterious channel. It is a constrained system with clear incentives. Teams struggle not because it is opaque, but because they optimise for the wrong things.
Why Apple Search Ads still behaves differently
Orlov’s first point was almost unfashionable in a market obsessed with novelty. Apple Search Ads remains one of the strongest acquisition channels in mobile.
The reason is not efficiency, automation, or scale. It is intent.
Search is explicit. A user typing a keyword in the App Store is already close to a decision. “You are working with direct search intent,” Orlov said. “That gives you the highest purchase intent across the whole marketing mix.”
“You are working with direct search intent. That gives you the highest purchase intent across the whole marketing mix.”
That intent creates stability. Prices fluctuate less violently than in social channels. Volumes do not collapse overnight. Creative fatigue exists, but it is not the dominant constraint. Apple Search Ads rewards correctness more than novelty.
At the same time, the channel has not evolved in ways that make it easier to operate. There are still no native post-install integrations. There is no meaningful automation. Everything that matters is manual.
Which is precisely why discipline matters.
Control is not optional
Rather than starting with bids or budgets, Orlov began with structure.
The default way many teams set up Apple Search Ads accounts bundles dozens or hundreds of keywords into a single ad group, tied to a single bid. It feels efficient. It is not.
The alternative, single-keyword ad groups, feels tedious until the logic becomes clear. Control is not an abstract benefit. It is the only way to understand what is actually happening inside the auction.
“The more control you have, the better it works,” Orlov said. “Control is the biggest success factor in Apple Search Ads.”
“The more control you have, the better it works. Control is the biggest success factor in Apple Search Ads.”
Control enables relevance. Relevance determines whether Apple allocates impressions at all.
Apple Ads strategy for stable growth and higher ROI
Source: Business of Apps via YouTube
The competitor most teams misunderstand
When Orlov asked the room who their biggest competitor was in Apple Search Ads, the answers followed a familiar pattern. Other advertisers. Big brands. Aggressive bidders.
They were all wrong.
“The most important competitor is the number one organic result,” he said.
“The most important competitor is the number one organic result.”
Apple Search Ads auctions are not fought in isolation. Paid ads compete directly against organic listings for attention, credibility, and clicks. If the top organic result is more relevant than your ad, Apple has little incentive to show you frequently, regardless of bid.
This framing shifts the optimisation question. The task is not to outbid other advertisers, but to outperform the best organic alternative for that query.
Why IPM quietly governs scale
The centrepiece of Orlov’s talk was not CPI, ROAS, or any familiar executive dashboard metric. It was IPM: installs per thousand impressions.
IPM is not something Apple exposes as a first-class metric, but it sits at the heart of the auction. It combines tap-through rate and conversion rate into a single signal of relevance.
“Apple doesn’t use IPM directly,” Orlov explained. “But it uses the components of it to decide who deserves impressions.”
“Apple doesn’t use IPM directly. But it uses the components of it to decide who deserves impressions.”
Below a certain threshold, scale stalls. Impression share caps out. Campaigns feel artificially constrained. Cross that threshold, and something changes. Impression allocation accelerates. Volumes grow faster than spend. CPIs begin to fall instead of rise.
Internally, Orlov described this threshold as an “invisible wall.” The auction does not reward incremental improvement evenly. It rewards relevance disproportionately.
Custom Product Pages as relevance engines
Improving IPM requires changing what users see, not what teams bid. This is where Orlov was most direct.
“Custom Product Pages are the most powerful tool in Apple Search Ads,” he said. “And almost nobody uses them properly.”
“Custom Product Pages are the most powerful tool in Apple Search Ads. And almost nobody uses them properly.”
Despite Apple expanding the limit to 70 Custom Product Pages per app, most teams treat them as a nice-to-have. Design resources are scarce. Roadmaps are full. Gains are assumed to be marginal.
In practice, CPPs are where relevance becomes tangible. Matching messaging, visuals, and onboarding flows to specific keyword intent changes how the auction evaluates your ad. Conversion improves. IPM rises. Impression share follows.
Deep links extend this logic. Sending users directly to the feature, offer, or flow they searched for shortens the path to value. It also signals alignment to Apple’s system.
“You’re not just improving conversion,” Orlov noted. “You’re improving how Apple understands your relevance.”
“You’re not just improving conversion. You’re improving how Apple understands your relevance.”
Pricing is an outcome, not a lever
On auction pricing, Orlov pushed back against the idea that competition alone drives CPI.
For apps with in-app purchases, Apple has near-perfect visibility into revenue. Keyword prices reflect average profitability across the category. For apps without IAPs, Apple estimates. And it estimates well.
“If a keyword is not profitable for you,” Orlov said, “it means it’s profitable for someone else.”
“If a keyword is not profitable for you, it means it’s profitable for someone else.”
Apple Search Ads functions as a performance benchmark. When campaigns fail, the explanation is rarely that the channel is broken. More often, the execution does not yet match the economics of the category.
Knowledge as the missing layer
Tools, Orlov argued, have converged. Most platforms offer similar features, similar dashboards, similar promises. What does not converge easily is understanding.
“The knowledge is the most important factor,” he said. “Understanding how the auction works is what creates results.”
“The knowledge is the most important factor. Understanding how the auction works is what creates results.”
Across the case studies he shared, spanning fintech, utilities, mobility, and subscriptions, the pattern was consistent. When relevance improved, prices fell. When IPM rose, scale followed. Category mattered far less than execution.
Apple Search Ads is not forgiving, but it is predictable. Teams that learn how its incentives align are rewarded not with tricks, but with compounding performance.
A channel that rewards seriousness
Apple Search Ads does not reward haste. It does not reward shortcuts. It rewards teams willing to treat it as a system worth understanding.
Orlov’s argument was not that Apple Search Ads is easy. It was that it is honest. If results disappoint, the feedback is already there, embedded in impression share, relevance, and conversion.
The work is not glamorous. It is also difficult to fake. And for teams willing to engage with it properly, that may be its greatest advantage.
Watch the recording to discover all of Viktor’s insights. You can also watch all the sessions from Business of Apps Berlin 2025 here.