Ariel: Welcome to a brand new AF Chat. It’s been a while since I had a guest on, but I’ve got a really cool one today: Marcus Burke.
Marcus: Hi Ariel. Nice to be here. Hello everyone.
Ariel: Today we’re talking about Meta ads, but not just the basics. Marcus already covers a ton of that on his social channels.
We’re going deeper: the bridge between Meta ads and app store optimization (ASO) — because that’s absolutely critical if you don’t want to waste money. We’ll talk about:
But first, a tiny bit of tradition.
Ariel: Let me know in the chat: what’s in your cup and where are you from?
I’m drinking Perfect Peach tea, which I’ve fallen in love with over the last few months. I keep drinking it and stocking our office with it. It’s in my new Appfigures mug with our logo on it, which I’m very happy about. I’m in (surprisingly) sunny New York City today. The weather here is wild.
Marcus: I only have a cup of water.
Ariel: Water is great.
Marcus: I should have caught that. I know you always do the tea thing. I should have gotten one. I’m also into fruity sweet teas more than Earl Grey or anything.
I’m based in Berlin. Not sunny at all. It’s cold by now, but otherwise a beautiful city.
Ariel: Perfect. Sun, not sun, tea and water. I also have water here.
Marcus and I were just at the RevenueCat conference last week. We both did so much talking that my throat is still feeling it. We actually pushed this live stream back once because I had a cold and completely lost my voice. Apparently my throat is always on display.
The chat is full of people with coffee in Berlin, coffee in Brazil, German beer, Earl Grey… a lot going on.
Ariel: Marcus, share with the world who you are, what you do, and why you’re here.
Marcus: I’m Marcus. I work as a consultant mainly focused on Meta ads and upper-funnel optimization. People know me more for Meta ads, but I always try to teach clients that you’re not going to win this game if you only focus on your ad account.
You need to streamline the whole funnel:
Only then can you build a truly scalable account. That’s a big part of what we’re going to talk about today.
I’ve been consulting for the past two or three years. Before that I led the growth team at Tandem, a German language-learning social network.
That’s a fun category but hard to work on because you need multiple users talking to each other for a delightful experience. You need to match them well, and you’re always dependent on them actually doing the job. You can’t fully control that with just tech. Hard job, but a lot to learn.
Ariel: That’s plenty of experience to get started. Language learning is such a fun topic, especially now with what Duolingo has done to the space and the million other language apps no one talks about because they’re not as “sexy.” But that’s a different conversation.
Today we’re focusing on the meat and potatoes of Meta ads: how to really extract more from Meta, beyond the basics.
You can follow Marcus on LinkedIn — we’ll link his profile in the description. That’s how I found him originally: talking about Meta ads on LinkedIn. Then we met at the conference last week.
Ariel: We’re talking about:
Because if you get a click from an ad, that’s just the beginning.
What is the bridge between Meta ads and app store optimization? How do these actually link up?
Marcus: The mechanics of the stores work in a way where apps that already have a ton of traffic are the ones winning on search.
You need to drive enough volume for the app store to recognize you as a viable player so you can rank in positions that bring organic search volume.
So you need a traffic source to get the whole machine started.
There are some hacks:
But I don’t want to talk about hacks. I want to talk about building a scaling brand and how to make sure everything you do in ASO pays into what you’re doing on Meta.
Once you scale your ads and the store algorithm starts recognizing you because you’re driving:
—which are the two most important factors—
…you start winning search.
I learned a lot of this from the Appfigures channel back in the day when I was doing ASO for Blinkist.
Let’s assume you’ve done basic ASO:
After that, the main way to win and outrank others is volume.
And the way you buy traffic on Meta has a big impact on that, because prices vary hugely by audience. For the same dollar you might:
On Meta there’s this mantra: “creative is targeting.”
But because Meta’s targeting happens under the hood, many advertisers don’t even track where their ad money is spent. And that’s a huge mistake because:
…all change how much volume you can drive, and therefore how much ASO impact you get.
I want to dive into how to leverage that for your brand and how to make sure ASO becomes a sparring partner to UA so that your blended costs come down as you scale.
Ariel: You said “creative is targeting.” I’ve been talking about something similar with screenshots and how you present yourself.
When you do paid ads, you’re getting installs that could convert into ratings if you optimize. If you don’t, you just get “dumb installs.”
Those don’t:
When we say blended cost, we mean the overall cost to acquire a new user — organic + paid together.
You’re essentially saying that if people listen to this and apply it, they can bring that cost down.
Marcus: That would be nice. Let’s try to make that happen.
Ariel: If you take any of this advice and apply it, I’d love to hear how it goes. People come back to my LinkedIn posts and YouTube videos and share results — usually positive. If that happens for you, let us know.
Ariel: Let’s go high-level. You put an ad on Facebook/Instagram. The user taps, lands on the App Store. Then what?
Marcus: If Meta is your first major paid channel — which is common — you want whatever you do on Meta to feed directly into your optimization strategy on the app stores.
If you’re driving volume through certain messages in your Meta ads, you want people to find those same messages on the store.
You want a clear link between:
That doesn’t necessarily mean setting up custom product pages (CPPs) from day one, but at least:
I’ve seen wild mismatches between ads and store pages. But the store page is your landing page. If people click into one thing and then see another, they’re gone.
Ariel: Let’s define CPPs quickly.
Marcus: CPPs, or custom product pages, are Apple’s way to let you create specific landing pages you can deep-link to from channels like Meta.
I see a lot of advertisers tempted to start with CPPs because it’s a cool feature. And yes, a perfect match between an ad and a product page can improve install performance.
But creating new screenshot sets is time-consuming:
Meanwhile, on the ad side you usually move much faster.
So in the beginning, I would:
The uplift from a perfectly aligned CPP is usually marginal compared to finding a winning creative angle. A winning ad can 3x performance. A good CPP might give you +5–10%.
Once you know what works, you can:
Ariel: So you’re saying: find the winning ads first, then build CPPs that match. Don’t do everything in parallel.
Marcus: Exactly. Don’t over-index on CPPs. They’re a good tool, but they won’t make or break you if your default screenshot set is at least average.
Once you know which ads are working and see clearly different angles or audience archetypes, then it becomes obvious what to put on CPPs.
Ariel: Let’s say my app serves three to five different audiences. Not wildly different, but different enough that I’d want separate product pages.
Would you:
Marcus: Finding that second or third angle is usually the bigger lift.
I’d still:
Once one angle clearly performs best, double down and really nail it:
If you always drive a generic mix, it’s hard to know how to optimize anything because it “could be anything” behind your blended metrics.
Later, as you scale and add more channels and audiences, experiments become harder:
So early on, there’s value in being quite specific. Find traction with one audience, learn it deeply, then expand.
Ariel: That makes sense. And for Meta specifically, it’s also much harder to “max out” like you do with Apple Search Ads.
Marcus: Exactly. Meta has massive volume. More than 3–4 billion monthly active users across Facebook and Instagram. Half of humanity.
But creative still dictates where you deliver.
Marcus: A simple example: media type.
If you only do short-form 9:16 videos, Meta will deliver mostly on:
The users who spend time there skew younger. Someone like my mom hangs out on the Facebook feed. She’s not on Instagram; definitely not on TikTok.
If you want her, you need ad formats that work there:
So just by picking a media type, you implicitly choose placements, and by choosing placements, you implicitly choose audiences.
As you scale, you need to figure out:
That’s what keeps you from hitting a ceiling.
Ariel: You mentioned reading reviews and turning them into angles. That’s such a clever idea.
Reviews are a treasure trove of information:
The sweet spot is competitors’ users complaining about something you do have.
Then you:
If those users see your ad, they think, “I’ve got to try this.”
But if they land on your store page and don’t see that benefit, they think it was clickbait or false advertising. That’s the importance of bridging ad → store → app.
Marcus: Totally. And beyond store reviews, you can look at:
With AI it’s easy now:
It’s so accessible that I do a lot more research now than a few years ago because it’s so cheap in terms of time.
Ariel: You also wanted to talk about web-to-app and web-to-web, since it’s a hot topic.
Marcus: More and more apps push:
For a few categories, this works tremendously well. There are big players spending 80–90% of their budget on web instead of app.
They look like small brands on the App Store but are huge on Stripe.
However, that absolutely cripples your install volume.
If:
Then 95% of that traffic never hits the app store.
To the store algorithm, you look tiny, even if your install rate from store page to install is 90%+ (because everyone already paid).
If you go all-in on web-to-app, you need to:
If you’re in a hyper-competitive vertical and you’re sure you’ll never outrank the top three anyway, this might be fine.
But if you want store search as part of your strategy, you must send some traffic to the app store directly.
As you scale, a hybrid is usually best:
Ariel: You have an example you wanted to show with Appfigures.
Marcus: I’m using Appfigures to look at the keyword rank history for Cali, one of the major AI calorie trackers.
They haven’t been around very long, but they’re making waves in the health & fitness / nutrition category.
If we look at big terms like:
…they’ve climbed into very strong positions.
They’re competing on "calorie" (popularity 51 in the US) with:
On "calorie" they even outrank Lose It!, despite being a relatively young challenger brand.
If we look at total ratings in the US:
Revenue estimates for Q3 (from Appfigures):
Cali is far away from MyFitnessPal in revenue, but when we look at installs:
Now look at downloads per rating (DPR):
So from the same amount of installs, Cali generates more ratings. That’s one of the biggest signals for the store. To the algorithm, ratings growth says:
“These guys are important.”
And that’s how they end up in top positions on very competitive terms.
Why is Cali able to do this?
That brings in very cheap volume at low CPIs.
MyFitnessPal, by contrast, seems more focused on converting store search rather than pushing big UA from Meta:
So challenger brands that:
…can attack established players that have been around since basically the origin of the App Store.
I’d encourage everyone to do this kind of analysis:
Maybe you can:
Ariel: It’s fascinating to see the two lines cross: installs first, but then especially ratings.
Cali is relatively small compared to MyFitnessPal, which is huge, old, everywhere, and very “default” in the space.
But because Cali:
…the store starts rewarding them. DPR comes down, ratings accelerate, and rankings follow.
Ariel: You mentioned young vs. old audiences. How should we think about demographics for Meta ads?
Marcus: Definitely think about it.
The young vs. old split is one of the first that matters.
In general:
Younger audiences (18–24, roughly):
Older audiences (30+, 35+):
On Meta, if you optimize for start trial, Meta learns:
“Find me people who start trials cheaply.”
That’s going to bias strongly toward younger audiences.
You need a clear understanding of:
Both can be viable strategies, especially as you scale, but:
Meta gives you:
Use those to rate creatives by audience, not just by blended numbers.
Ariel: And all of this feeds back into ASO.
If your blended install rate on the store is trending down over time, an ASO person in isolation might panic.
But that might actually be:
So the install rate goes down, but revenue could be going up.
Marcus: Exactly. When you look at store analytics, Meta traffic usually shows as:
Younger users especially:
So if your Meta spend grows and you see search traffic going up, don’t instantly celebrate that as “pure organic.” Part of it is driven by your paid campaigns.
Also note:
Again: young and old behave differently, and you have to keep that in mind when looking at store metrics.
Ariel: You said if you’re attracting a lot of young users who won’t convert, it might still be worth it to capture ratings from them.
So should you show the rating pop-up as early as possible?
Marcus: I’m a big fan of asking early.
I’ve seen many apps significantly increase ratings per download just by asking during onboarding.
Why?
I’ve seen early prompts beat more sophisticated strategies where teams try to:
If that “delight” happens on Day 3 and almost everyone has churned, it doesn’t matter if conversion on that prompt is 100%. You’re just not asking enough people.
So early on, just:
People are getting used to it because many apps do it.
However, don’t just throw the pop-up on the launch screen. Make it intentional:
Then put the rating request after that.
Some apps also do a custom “social proof” screen first:
Then trigger the native rating pop-up on top of that.
And don’t underestimate the role of the ad in this experience. If the user thought the ad was cool, that’s already part of the value journey. Asking for a rating after a strong first-run + ad combo isn’t unreasonable.
Ariel: I’ve seen apps that blindly ask right on launch. That feels bad to me.
Your framing is interesting: the whole thing is an experience:
If you do it right, by the time you ask for a rating, they feel like they understand and like the app.
And realistically, if they’re going to churn after one use anyway, you might as well ask. If they dismiss it, you didn’t lose anything.
Marcus: Exactly. Just make it intentional.
Ariel: You talked earlier about Cali’s pricing and how it’s tuned for the younger audience they attract.
Marcus: Right. Cali’s yearly pricing is around $30. They have discount flows that bring it down to ~$15. That’s a fraction of what other apps charge.
This lets them:
But it makes it harder to convert older, more expensive-to-acquire users at a high enough price point to get a good ROAS.
So the best player in a space would ideally:
Ariel: So onboarding and pricing should start at the creative level:
Marcus: Pretty much.
“Creative is targeting” goes further than most people think.
You can make the wrong ads for your price point.
If you:
Your conversion is going to be terrible.
That doesn’t mean “wrong audience.” It might mean:
Sometimes just cutting price in half can quadruple conversion and suddenly your economics work.
So there’s no single “right” audience. You just have to:
Marcus: Let’s talk about app icons, one of my passion topics.
Most apps use basically the same patterns:
Probably 80%+ of apps fall into those buckets.
From a user’s perspective, that’s confusing. If they’re browsing on the store and see 10 circles, how do they choose?
In many placements, they only see:
No screenshots, no long description. So if your icon is just another generic circle or letter, you don’t stand out. Clicks become almost random.
I think app icon optimization is more important than screenshot optimization, or at least a bigger lever, because:
So if you increase click-through from impression to product page, that’s a conversion rate uplift. And that’s huge for ASO.
Most teams:
Then when you do try to change it, it becomes a heated brand debate:
Meanwhile, from the user’s perspective, they’d prefer a descriptive icon that:
I’ve seen big improvements from icons that look more like mini banners than logos:
Instead of a “clever” circle that only brand designers understand.
Ariel: Earlier this year we released an icon research tool at Appfigures. You can:
Or start from a keyword like “flower” and see all flower icons.
I asked people to drop their apps and I’d show them how many others look exactly like them.
With a small handful of exceptions, almost nobody is unique.
Everything on the App Store is so crowded. If I search for “calendar,” I see infinite calendars:
As a user, I have no clue who to pick. I probably:
And like you said, the icon is a banner.
To me, experimenting is the key. The teams that don’t test icons because brand wants purity are just leaving growth on the table.
The App Store and Google Play are not places to show off brand aesthetics. They’re places to drive discovery and downloads. Show off fancy branding on shirts, walls, or press; in the store it’s about conversion.
We’ve seen examples where apps used more descriptive “banner-like” icons and conversion improved significantly. That goes against common belief, but most people simply don’t test icons.
Testing was impossible for a long time; now with Product Page Optimization you can test. Is it perfect? No. But it’s far better than guessing.
Marcus: Exactly. Is Product Page Optimization perfect? I’m not sure. I always double-check with before/after data, and on Android I like to test on the Play Store too.
On iOS, the reported uplift is based on:
But a lot of impressions are from placements where users never click anyway, which can dilute results.
Still, I’d much rather:
…than have the icon be purely brand-driven with no data.
It’s a design challenge:
Not a license to ignore performance entirely.
Ariel: I see the same dynamic with app names.
Brand people say, “Don’t put keywords in the app name, it makes it dirty; just use the subtitle.”
But then competitors take those keywords and outrank you.
Beyond ASO, keywords in the title help humans.
If my mom sees “Dropbox” alone, she has no idea what it does. If she’s searching for “photo backup” and sees two results:
She’s going to tap the second one, especially if the icon also communicates photos + backup.
So even a huge brand like Dropbox could have a discoverability problem purely from naming.
The same logic applies to icons.
I once saw the Mailchimp chimp on a building in NYC — just the mascot, no name, no tagline. My Uber driver saw it and said, “That’s cool, they must sell something with chimps.”
I told him it’s an email platform. He was completely deflated.
Mailchimp had this massive banner and did nothing with it for regular people.
The App Store is exactly that, except users are actively searching for a solution. If your icon doesn’t tell them what you do, you’re squandering that intent.
You can:
If your store page doesn’t convert — icon, screenshots, description — you might as well have spent nothing.
Marcus: For Meta traffic specifically, the icon doesn’t matter as much after they land — they’re not judging you on that when they’re already on your product page.
But for organic uplift and blended CAC, the icon is huge.
You want the store algorithm to notice you. One of the best ways is to:
Ariel: Toward the end, you made a subtle but important nuance. You recommend being specific in learning about audiences — but not by over-constraining targeting.
Marcus: Right. I still recommend:
But then:
Don’t tell Meta “only show to X age and Y interest” if you can avoid it. Narrow targeting usually means you pay more for that constraint.
Instead:
That way you:
Be aware of what you’re buying, but don’t force it artificially unless you have a strong reason.
Ariel: We’ve talked a lot about ASO and funnel alignment. Let’s zoom out.
What’s next for Meta ads and display in general?
Marcus: Meta will keep increasing ad load over time:
At the same time, there’s been some widely shared data that social media usage might have peaked, especially for younger users.
Usage seems to be trending down slightly for the first time. If that holds, overall growth in available attention will stall.
Now add AI:
Ad creators will jump on this, too. That raises philosophical questions:
I’m not super optimistic about that vision.
Ariel: My take is slightly more optimistic, but I agree about the short term.
In the next 1–2 years, everyone will:
Right now, it’s cheap and fast. As a person who creates a lot, I can feel the generic tone and neutrality, but it’s so efficient that people will use it anyway.
But then:
At some point, being “hand-made” might be a differentiator:
We could see new cycles where human creativity comes back as a premium layer.
In the meantime, though, these tools are an enormous advantage:
Marcus: I could see networks pop up that explicitly:
If that’s even technically feasible at scale, people might prefer to hang out there.
On the other hand, I’m also excited about these tools personally.
In the old static-ad days, I used to make my own ads and I loved that loop:
I’d encourage anyone running ads to make at least some of their own creatives. You’ll fail a lot, but you learn quickly.
With tools like Sora and others:
That lowers the barrier to experimenting, which is great.
Ariel: The pace is insane. Sora is moving fast, Runway dropped V3, 3.1… it’s hard to keep up.
In the short term, this is a huge opportunity:
In a year, it might become table stakes and the economics will change, but right now the advantage is real.
Ariel: We covered a lot:
There were a lot of great questions in the live chat. Many of them are more like the things Marcus talks about on LinkedIn.
I’m thinking of collecting those questions, turning them into a list, and maybe collaborating with Marcus on a guide or blog post.
If you asked something and we didn’t get to it live, we’ll try to get answers out through other channels.
Marcus: If you’re not following me on LinkedIn yet, do that. I talk about all this stuff there.
Lately I’ve been very focused on AI-driven creative production. I want to get back into making my own ads and I’m partnering with some freelancers.
If you’re thinking about AI creative production, feel free to hit me up. I’m happy to discuss and share learnings. Everyone’s still early here.
Ariel: We’ll link Marcus’s LinkedIn in the description.
We’ll also figure out what to do with the chat questions — maybe a blog article, maybe something we send out via email. So if you’re not subscribed to the channel yet, make sure to do that.
This is where we talk about everything related to getting more downloads and more revenue. And if you haven’t already, give the video a like so we can optimize for the YouTube algorithm.
Marcus, thank you very much for joining. This was super exciting. I’ve been looking forward to this all week since we met at the conference.
We’ll be back in a few weeks with another live stream.
Marcus: Thanks a lot, Ariel.
Ariel: Thanks everyone for watching. See you soon.
✨ This transcript was generated and enhanced by AI and may differ from the original video.
Learn how to legally reverse-engineer your competitors’ app strategies using real data: performance, audience, keywords, tech stack, and more.
Live teardown of 10 app product pages, focusing on screenshot mistakes, keyword strategy, and concrete tips to improve ASO and conversion.