App Expert Reveals How to Improve Conversion w/ Julie Tonna

Kicking Off: Why Conversion Optimization Matters

Ariel welcomes everyone to a new live stream and introduces Julie Tona as today's expert guest. The topic is conversion optimization for apps, especially on the App Store.

Ariel is in New York City, already in winter mode, drinking peach ginger tea. Julie joins from the south of France, in Marseille, keeping it classic with water after a recent heat wave.

After a brief nod to the old "Zoom shuffle in" days and some quick chat about iced tea and espresso, Ariel shifts the focus to the main topic: how to turn all the traffic you work so hard to get—through App Store Optimization (ASO), Apple Search Ads, and other channels—into actual downloads and growth.

Why iOS and Apple Ads Are So Central

Ariel explains that a big theme across past live streams has been:

• App Store Optimization (ASO)
• Apple Search Ads
• Different ways of driving traffic to your app

Julie introduces herself more formally:

"I'm a growth expert. I've been working within the app ecosystem for around five years, in both big and smaller companies. I have a pretty good idea of what's working and what's not, especially on iOS."

Julie used to work at Apple, specifically on Apple Search Ads. From that vantage point, she repeatedly saw:

• iOS as the best-performing platform for many clients
• Apple Search Ads at or near the top of their best-performing channels

Ariel reinforces this with data previously shared in his newsletter: iOS is where the money is for apps and games. He also notes that many apps are iOS-only, largely for revenue reasons, even though Google Play is improving and becoming more like the App Store.

The Core Problem: Impressions Without Conversions

Ariel frames the key question: once a user lands on your App Store product page, then what? Are you sure they'll download the app?

Julie sees this problem often:

"Most developers want scale and volume. They get all these impressions and then you look at the downloads and there's a massive drop. You paid for and managed to get all those impressions, but you're not managing to get the downloads. It's quite a shame—you should be able to convert."

The goal of the conversation is to:

• Unlock growth tips and opportunities
• Improve conversion rate
• Make the most of both paid and organic traffic

Sometimes this means accepting fewer impressions at first but getting a much higher conversion rate, which still yields more installs and better ROI.

Custom Product Pages: A Huge Shift

Ariel has spent much of the year talking about:

• Product pages
• Screenshots
• How they impact downloads

He’s seen many viewers update screenshots after previous live streams and then report more downloads.

Then Apple introduced Custom Product Pages (CPPs):

• Initially extremely powerful for Apple Search Ads, allowing different product page variations per keyword.
• Now expanded to organic search in iOS 16, which Ariel calls one of the biggest App Store changes in a decade.

Julie:

"I've seen custom product pages being used a lot for Apple ads because you could improve your conversion rates. It's insane. You can run on the same keywords and have a product page variation just to align with this keyword. It's amazing."

Previously, that power was limited to paid traffic. Now it’s being linked to organic search terms (with some constraints, which they later unpack).

Ariel’s team at Appfigures has been tracking CPPs for months:

• 83 of the top 100 grossing apps use custom product pages.
• The 17 that don’t often have specific reasons (some simply don’t like Apple’s ecosystem constraints in various ways).
• Below the top 100, usage drops off sharply—but Ariel expects adoption to grow quickly now that CPPs work for organic.

Julie expects CPPs to:

• Make things more competitive
• Change rankings for some keywords
• Reduce irrelevant rankings, because everyone can now better align messaging with what users search for

A Powerful CPP Case Study

Julie shares a concrete example from her Apple Search Ads experience:

• A puzzle game app offering multiple modes: crosswords, sudoku, word and number games.
• They split campaigns and product pages by theme:
  • One custom product page dedicated to "crosswords" queries
  • Another for "sudoku" queries

Result:

• Tap-through rate increased by around 27%
• Conversion rate stayed stable
• Net effect: many more installs with the same traffic

Julie calls it "insane" and a massive win, and she hopes that team is now also using CPPs for ASO.

What Custom Product Pages Actually Do

For anyone who missed Ariel’s prior walkthrough, he summarizes CPPs:

• They let you change your screenshots (and creatives) based on the keyword a user searched.
• Example: If a user searches a specific use case, your product page can visually match that use case.

This is now available in iOS 16 (beta at the time of the conversation), and Ariel has already seen it live. His verdict: "wonderful."

Ariel wants to focus on:

1. What a custom product page is
2. How to set it up
3. How to make it actually help with conversion

He sees this as the biggest actionable tip right now: build custom product pages before iOS 16 fully ships, so you’re ready.

But there is a twist: the way Apple implemented CPPs for organic introduces significant limitations for optimizers. Ariel and Julie come back to that later.

Step One for Conversion Optimization: Go Deep on Data

Ariel asks Julie: where does she start when optimizing conversion for an app?

Julie’s process is heavily data-driven:

1. Look at actual conversion rate

   • Not just overall, but broken down by country/geo.
   • Example: high conversion in the US, very low in Germany.
   • Each country may respond differently to your app.

2. Prioritize by geo

   • Start where conversion is lowest and where you have meaningful volume.
   • Don’t try to optimize everything at once.

3. Within priority countries, analyze keyword performance

   • Where you rank highest
   • What your best-performing keywords are

4. Craft your product page around those keywords

   • Add keywords to screenshots
   • Refine app title and subtitle
   • Max out relevant, high-intent keywords in metadata
   • Use a descriptive description

This provides a roadmap: start with data, then focus your creative and copy around what users actually search for.

Ariel connects this to intent and segmentation:

• Keywords already segment your users by intent.
• That makes them a natural organizing axis for CPPs.

At Appfigures, Ariel’s team even built an AI agent that:

• Analyzes organic and paid keywords
• Identifies which are working best
• Suggests where to focus CPPs

It’s in beta, but has already made planning much simpler for their members.

Julie also uses AI for:

• Identifying high-volume keywords with lower conversion rates
• Surfacing huge opportunities where a modest conversion lift would yield thousands of extra installs

Grouping Keywords for CPPs: Use Cases First

Ariel asks how Julie segments keywords into groups for CPPs: by features, demographics, or use cases?

Julie generally groups by use case, though category and app type matter:

• Example: Travel app with flights, hotels, and car rentals:
  • One CPP focused entirely on flights
  • One on hotels
  • One on car rentals

Users searching for "book flights" don’t necessarily care about car rentals at that moment, so showing a page focused on flights makes the app feel relevant and tailored.

Ariel adds another example:

• A billing app for small businesses with many verticals: electricians, plumbers, etc.
• These users may search for "invoicing for plumbers" rather than "invoicing for small business."
• Showing a screenshot with a plumber and plumbing context instantly signals, "This is for me."
• But if you crowd that screenshot with plumbers, electricians, nurses, etc., the relevancy is diluted.

Competitors who show the exact niche ("for plumbers") will win those users.

Julie emphasizes that Apple Search Ads are valuable not just for installs, but also for insights:

"You pay for installs, but you also pay for insights about what people are really looking for and where there is volume and impressions available."

Now that CPPs work for organic, the flow can be:

1. Use Apple Search Ads and CPPs to test which angles and use cases convert best.
2. Bring that learning back into your organic strategy.
3. Update keywords, screenshots, and CPPs accordingly.

Testing and Product Page Optimization

A viewer (Leonardo) asks how to get A/B test results for low-install apps, and whether using Apple Search Ads with a default ad will trigger Store experiments.

Julie notes:

• From an Apple Search Ads perspective, CPPs there are not true A/B tests.
  • Most impressions go to the CPP variant when live.
  • The practical pattern is: default page for a couple of weeks, then CPP for a couple of weeks—more of a time-based comparison than simultaneous testing.

Ariel adds:

• Running Apple Search Ads just to accelerate testing is only effective if you’re willing to invest real money.
• In some categories, traffic is very expensive. He cites a Terra app where:
  • The Apple-recommended bid brought almost no impressions.
  • Doubling it did nothing.
  • Doubling a couple more times finally unlocked real traffic but quickly exhausted the budget.

For most low-install apps, Ariel recommends:

• Use the built-in Product Page Optimization (PPO) tools.
• Give experiments much more time.
• Make variations dramatically different, not subtle:
  • Don’t just tweak background colors or one word.
  • Test very different angles or layouts so changes are statistically significant.

He mentions a past live stream with an A/B testing expert (from companies like Google), who shared the same advice: big changes when traffic is limited; finer refinements only once you have real volume.

Apple Search Ads vs Meta Ads at Launch

Another viewer asks: during the first three months of an iOS app launch, which provides better conversion—Apple Search Ads or Meta ads?

Julie calls it a trick question but breaks it down:

Apple Search Ads

• Pros:
  • Very high-intent users (they’re already searching in the App Store)
  • Often better user quality and retention
  • Strong performance when aligned with relevant keywords
• Cons:
  • More expensive
  • Limited volume compared to broader social channels

Meta Ads (Facebook/Instagram)

• Pros:
  • Potentially greater scale
  • Can be more efficient on cost when optimized well
• Cons:
  • Less intent; you must "sell" the value more
  • Creative-heavy: you need many ad variations, which is resource-intensive

Julie’s guidance:

• If your goal is performance and scale early, Meta might be better.
• If your goal is high-value, high-retention users, Apple Search Ads shines, particularly after the initial launch phase.

Ariel adds:

• If users are actively searching for your solution (e.g., "calculator app"), Apple Search Ads are ideal—don’t make them jump from another channel into the App Store.
• If users don’t yet know they need your product (new tech, new category), you may need:
  • Broader channels (Meta, Reddit, etc.)
  • Non-keyword Apple placements like the Today tab or Search tab (brand billboard style)

Julie elaborates on Apple Search Ads placements:

• Today tab

  • Huge visibility and volume
  • Typically low conversion and high cost
  • Better for big budgets and brand visibility than direct performance

• Search tab

  • Often better CPI (cost per install)
  • Good for re‑engagement (e.g., targeting returning users)
  • Now supports CPPs, allowing more relevant creatives

For launch:

• Use search results and keywords if you have search intent.
• Consider Search tab and Today tab later, depending on budget and strategy.

Turning Keywords into High-Converting Screenshots

Once you know which keywords and geos to focus on, what’s next?

Julie’s approach:

1. Audit your existing product page against target keywords:

   • Do those keywords appear in your title or subtitle?
   • Do you feature them in screenshots?

2. Recognize screenshot importance:

   • On a small screen, screenshots can take up a third of the display.
   • They are often what users look at first and most.

3. Include keywords in screenshots:

   • Visually and textually show what people searched for.
   • Example: if "book cheap flights" is a key term, your screenshots should literally say and show that.

4. Test preview videos:

   • Helpful in some categories (especially games).
   • Less necessary for simple utilities where static screenshots can explain everything.

Ariel underscores the importance of placing search terms directly in screenshots:

• Seeing the exact phrase (or close to it) reinforces that the app is relevant.
• This is a major reason to have multiple screenshot sets and CPPs.

Julie notes that at Apple, one of the top best practices they promoted was:

"If you want to do a custom product page, add the same keywords you’re bidding on in Apple Search Ads into your screenshots. It’s a no-brainer. You’re looking for something, you see it on your screen, you know it’s what you’re looking for, and you download."

Many teams still miss this because they focus primarily on branding and design instead of direct alignment with user intent.

Ariel adds another layer:

• Apple now reads screenshots with AI.
• That means keywords in screenshots can help with indexing and discoverability.
• Failing to do this doesn’t get you penalized, but competitors that do will gain an edge.

Learn from Competitors (Without Copying)

Julie recommends:

• Look at competitors that rank higher or get more downloads.
• Analyze:
  • Their screenshots
  • Their copy
  • Their keyword usage
  • Their reviews

"It’s not copying, it’s just getting inspired."

Ariel strongly agrees. Whenever someone asks, "What do I do?", his default answer is:

• Look at the most successful apps in your space, even if they’re not direct competitors.
• Find what they’re doing right, then adapt it.

Useful competitor signals include:

• Reviews: goldmine for learning what users love, hate, or request repeatedly
• Download trends: to understand whether they’re growing or shrinking
• Where they’re featured
• Their Apple Search Ads keywords and creatives
• How they use CPPs

Ariel mentions he even built AI to analyze CPP strategies at scale and summarize them in simple, actionable insights.

Pitfalls of Audience Refinement in Apple Search Ads

They briefly dive into a more technical Apple Search Ads topic: audience refinement.

Julie cautions heavily:

• If you use audience refinement (age, gender, returning users, etc.) at the campaign level, you heavily rely on whether users allow personalized ads.
• Approximately 80% of users say no to tracking/personalization.
• If personalized ads are off, and you’ve enabled audience refinement, your ad often won’t appear.

This is particularly risky for keyword campaigns where volume is already limited.

Julie’s rule of thumb:

• Treat the keyword itself as your primary targeting.
• Avoid audience refinement for low-volume placements.
• Consider it only for:
  • High-volume contexts (Today tab, Search tab)
  • Very specific re‑engagement strategies

Ariel notes that this pattern—more performance by using less targeting—is similar to advice he’s heard in previous live streams about Meta ads: broad targeting often lets the algorithm do its job better.

Protecting and Targeting Brand Keywords

Julie also advises checking if competitors are:

• Running on their own brand terms in Apple Search Ads.

If they’re not:

• Their brand terms may be cheap to bid on.
• You can capture high-intent users searching for that brand.

However, Ariel offers nuance:

• In some categories (e.g., food delivery), users are simply hungry and indifferent to brand loyalty. Bidding on competitors’ brands can be very effective.
• In others, users are deeply committed to a specific brand; bidding on their name may be less fruitful.

Still, it’s worth testing—and the first step is always checking whether the competitor is actively protecting their brand with their own Apple Search Ads.

The Big Limitation: How Apple Tied CPPs to Keywords

A key constraint in Apple’s implementation of CPPs for organic:

• You must attach each CPP to specific keywords.
• You cannot type arbitrary keywords; you can only pick from a predefined list.
• That predefined list is derived from the keyword field of the main localization for the country.

Implications:

• Keywords in app name and subtitle, which are often the most important, are not reflected in that list.
• Keywords added via secondary/localized metadata hacks (like using Spanish-Mexico in the US) may not be accessible either.

Julie’s current approach (still experimental):

• She’s testing whether to keep traditional keyword lists or to change them to better match CPP needs.
• She’s trying to avoid repetitive keywords, instead using the field to represent distinct features/use cases.
• She’s reusing CPP configurations from Apple Search Ads and aligning them with organic keyword fields.

But she’s clear: it’s early, and no one has definitive answers yet.

Ariel shares two strategic ideas he’s considering (also early-stage):

1. Treat the default product page like just another CPP.

   • Don’t think of default as a catch‑all.
   • Choose your primary use case for the default page.
   • Reserve other use cases for CPPs attached via the keyword field.

2. Pack secondary use cases into the keyword field.

   • Example for a travel app:
     • Put "flights" and the primary positioning in title/subtitle.
     • Use the keyword field to cover "car rentals", "hotel deals", etc., creating hooks for related CPPs.

Both strategies must be tested, and Ariel notes that adoption isn’t yet broad enough to draw strong conclusions. Still, given that CPPs cost nothing to create, he encourages developers to set them up now:

"If it doesn’t do anything, okay. But if you get even a little bit, you did it. There’s no way to lose."

Deep Linking from CPPs to In-App Experiences

Julie brings up a crucial, often-overlooked tactic: deep links with CPPs.

• You can attach deep links to CPP campaigns.
• When a user taps install from a CPP and opens the app, they can land directly inside a specific feature or flow.

Example:

• A travel app with a CPP focused on flights can:
  • Show "Book flights easily" in screenshots
  • Deep link new installs straight into the flight booking flow

This improves:

• Post-install conversion (from install to action/subscription)
• Overall user experience consistency with what they saw on the product page

Julie also notes that you don’t always need completely new screenshot sets for tests:

• Sometimes just reordering screenshots (e.g., moving screenshot #5 to #2) can affect conversion.
• This is a low-cost, quick experiment if you don’t have time to redesign everything.

Ariel backs that up from his own teardown experience: often there’s a strong screenshot hidden near the end that, if moved forward, significantly improves performance.

Do Preview Videos Help?

A viewer asks if preview videos help for a workout app.

Julie’s guidance:

• Effectiveness depends heavily on category.
  • Games: preview videos often work very well.
  • Simple utilities: screenshots may be sufficient and more efficient.
• Check top players in your category:
  • If they use videos, consider testing one.
  • If they don’t, it might not be critical.

Ariel adds:

• If you decide to create a preview video, it needs to be good.
  • A mediocre or low-quality video can hurt conversion.
  • It’s better to have no video than a bad one.
• Focus on the first five seconds:
  • Users rarely watch longer unless instantly hooked.
  • Show your main feature or core use case right away.

Again, the theme returns: always test.

The Biggest Mistake: Misaligned Funnel Experiences

Ariel asks Julie the big question: what’s the one mistake that, if fixed, could change everything for conversion?

Julie’s answer:

"It’s really about aligning your full experience. Use your product page as a landing page. If you’re running ads, your ads and your product page need to make sense together. If a user clicks an ad, they need to see what they clicked on. Conversion will be better."

She expands the idea of alignment across the whole funnel:

• Ad → App Store product page → In-app onboarding → Core experience

If at any step users don’t see what they expected, they drop off.

Ariel breaks alignment down further:

• If someone searches a keyword and sees that keyword and use case clearly in the screenshots: good.
• If they then open the app and immediately see that use case in onboarding: better.
• If the app hides that feature behind a paywall or several taps: huge friction.

He notes that hard paywalls complicate this even more, but that’s a topic for a separate deep dive.

How to Actually Fix Alignment

Julie’s first step:

• Start with keywords: they often reflect specific features or use cases.
• Make sure what appears on the product page (copy + visuals) matches what people search for.

She also recommends user testing:

• Show people your App Store page.
• Ask what they expect from the app based on that page.
• Then see whether using the app actually meets those expectations.

Reviews and sentiment analysis are invaluable here:

• Look at what words and themes come up repeatedly.
• Identify what users love, what they hate, and what they wish the product did.
• This informs not just product decisions but also how you present features in screenshots and ads.

Ariel reiterates:

• Users are effectively telling you exactly what they need in reviews—your job is to listen.
• Do the same review analysis for competitors and steal the gaps they’re leaving.

Using Reviews and Reddit for Insight

Julie’s standard audit process includes:

• Data analysis (installs, conversion rate, keywords, rankings)
• Review mining for both the app and its competitors
• Sometimes also scanning Reddit for:
  • Unfiltered opinions
  • Pain points
  • Real-world use cases

She looks for:

• Repeated keywords and phrases
• Overall sentiment: how users feel about the product

"How you feel about a product is really important. That’s how people decide if they’re going to give you five stars or one star."

Ariel notes Reddit can be extremely useful, but also quirky—you may have to ignore some "Reddit-only" noise to get to real signal.

AI as a Selling Point: Use with Caution

A viewer asks whether to go all-in on "AI" keywords in screenshots, and what to do if some users are skeptical of AI.

Julie’s current position:

• She’s testing and hasn’t seen a clear, universal win yet.
• Many people search for "OpenAI" or "AI chatbot", but competition is fierce.
• She suspects people are not ready to see "AI" everywhere as the main selling point.

Her recommendation:

• Don’t just say "AI"—show what the AI actually does.
• Example: Instead of "AI-powered VPN", say "connects to the fastest server automatically" or "optimizes speed".

Ariel adds more nuance:

• If you slap "AI" on everything, your keywords become AI-focused.
• That throws you into competition with a massive number of AI tools and apps.
• Unless AI is truly the core, discovery may suffer.

He shares an example of an "AI VPN" where AI quietly optimized for speeds and server switching. Useful, but not something most users explicitly search for.

• For many people, "another AI app" can feel like noise.
• Only lean into AI branding when it concretely benefits the user in a way they understand.

Julie’s final advice on this:

"People don’t care if you use AI or not. They care what they’re going to get out of your app. Show what the AI does—that helps more than just saying you use some API."

The "Free" Keyword Trap

A question comes up about using the word "free" in CPPs.

Ariel notes:

• Apple often rejects submissions that lean on "free" too heavily.
• You might slip it through, but it’s an easy rejection trigger.

Julie points out a bigger strategic issue:

• Users who explicitly search for "free" are usually low-value users.
• In games, "free games" queries have high volume but tend to deliver:
  • Poor performance
  • Low monetization

Even if you monetize via ads, she reminds that many of those users want both "free" and "no ads".

In short:

• From both App Review and revenue perspectives, centering "free" is usually not worth it.

Dark Mode vs Light Mode Screenshots

A viewer asks if there’s a difference between using light-mode versus dark-mode screenshots.

Julie:

• She hasn’t tested enough to give a definitive answer.

Ariel’s take:

• This is exactly the kind of thing that must be tested.
• For some audiences, offering both might be ideal.
• Different keyword groups might respond differently, which can be reflected via CPPs.

At present, there’s no universal rule—but testing is straightforward using Product Page Optimization and/or CPP variations.

In-App Events as a Next Step

Julie mentions In-App Events as a later-stage tactic, once your core product page and alignment are solid.

• In-App Events show up in App Store search.
• Sometimes they appear more prominently than screenshots.
• You can now attach short videos to these events.

If done well, events can:

• Re-engage existing users
• Attract new ones with time-limited or special-value hooks

But she emphasizes: get your screenshot and CPP strategy aligned first; then layer in events.

What To Do After This Conversation

Ariel summarizes what he hopes viewers will do after watching:

1. Open Appfigures (or your analytics stack).
2. Analyze your keywords and your competitors’ keywords.
3. Read your reviews and competitors’ reviews.
4. Identify:
   • Main use cases
   • Secondary use cases
   • Repeated complaints and wishes
5. Use that information to:
   • Update screenshots (or at least reorder them)
   • Plan CPPs attached to your best-performing, high-intent keywords
   • Ensure everything from ad to app experience is aligned

Julie adds that once you’re confident in your core funnel, you can explore:

• In-App Events
• More advanced CPP setups
• Additional Apple Search Ads placements

But alignment remains the central theme.

Wrapping Up and Where to Find Julie

As the session ends, Ariel notes there were still more questions in the chat. He plans to answer additional questions later in the comments.

Julie and Ariel briefly discuss future ideas, like a teardown format focused on "Is your app aligned?", where they review real apps for funnel and messaging consistency.

Julie closes with a few resources:

• She runs a newsletter on Substack called Neo Ads, sending weekly breakdowns and use cases of ads that perform well from a growth perspective.
• She’s active on LinkedIn, where she shares tips, strategies, and observations on what’s working.
• She offers growth consulting and audits for apps and games, with a particular focus now on subscription models and full-funnel monetization.

Ariel promises to include links to:

• Julie’s newsletter
• Julie’s LinkedIn
• The recording of this live stream

He thanks Julie for joining and all viewers for their questions, and notes he’ll be back with another live stream in a few weeks, this time focused on open Q&A.

Julie thanks Ariel and the audience, and they sign off.

✨ This transcript was generated and enhanced by AI and may differ from the original video.