How to Optimize User Acquisition Across Major Ad Channels — Shane Ly, AppsFlyer

David Barnard:

Welcome to the Sub Club Podcast, a show dedicated to the best practices for building and growing app businesses. We sit down with the entrepreneurs, investors, and builders behind the most successful apps in the world to learn from their successes and failures. Sub Club is brought to you by RevenueCat. Thousands of the world's best apps trust RevenueCat to power in-app purchases, manage customers, and grow revenue across iOS, Android, and the web. You can learn more at revenuecat.com. Let's get into the show.

Hello, I'm your host, David Barnard. Today's conversation is shorter than usual and will be featured in RevenueCat's State of Subscription Apps report. Each episode in this series will explore one crucial metric and share actionable insights from top subscription app operators. With me today, Shane Ly, Solutions Architect at AppsFlyer. On the podcast, I talk with Shane about the state of major acquisition channels in 2025, how improved measurement is increasing confidence in ad spend and why Reddit might be your next top performing user acquisition channel. Hey Shane, thanks so much for joining me on the podcast today.

Shane Ly:

Thanks, David. Thanks for having us. Very excited to join. My name's Shane. I'm a Solutions Architect here at AppsFlyer. Typically, work on pre-sales side of things, but I work on a variety of clients including subscription apps itself of all sizes, so super happy to join and give insights itself.

David Barnard:

Awesome. So I did want to talk through what you're seeing and I think working at AppsFlyer, you do have a really unique perspective on the industry. I mean, you all work with tens of thousands of apps across all different verticals and see what's working, what's not working. You get to see a lot of data that probably a few people in the industry get to see. You get to work hands-on with a lot of people in the industry who get those unique perspectives I think few folks in the industry get. So yeah, super excited to dive into what's working today, 2025 with acquisition for subscription apps. And so I wanted to go through all the major channels, but let's just kick it off with an overview. What's going on with user acquisition today? What's working what's not?

Shane Ly:

In the past several years on iOS signal loss has been the biggest challenge itself. So how do we know that a ad is actually driving subscriptions itself? So as with iOS, with SKAdNetwork, we've got limited measurement windows. So essentially in order to do measurement, we have to be very creative with how we do this measurement itself today. Android is business as usual. For iOS, there's several new developments in 2024, which we can also take advantage of in 2025.

David Barnard:

Let's go through just channel by channel and what's working, what's not. So Apple Search Ads, I mean this is like the starter kit of what I and many people recommend. If you're not at least starting with search ads, you're probably leaving money on the table. So what do you see working in search ads and then anything new or different these past couple of years?

Shane Ly:

Typically, search ads are still working pretty well. Any measurement partners still getting reports of the keyword data. So what are you buying? Et cetera. Where are your placements? You're still able to lifetime value or LTB these users. It's easy to see a user coming in on Apple Search Ads and then ultimately, did they subscribe? Did they cancel? Did they renew? These are all signals that still were on Apple Search Ads itself.

David Barnard:

One thing that did change super recently, so I imagine you and the team are still sorting through exactly how this is going to impact things, but I know Apple did change view through attribution with Apple Search Ads. I don't even fully understand it, but as somebody who does this day in day out, maybe you can explain it to me and the audience a little better and what it will and won't impact moving forward.

Shane Ly:

Let's do a little bit of historical as well. So historically with Apple Search Ads, we were able to get click-through attribution or essentially as long as someone's clicked on the ad and we do attribution with them by way of their tokens, we do have the ability to attribute on these clicks. So very recently Apple Search Ads announced that they are also doing view through attributions, and essentially what that means is let's say they've searched for a keyword and they've seen promotion or seen the ad itself. Essentially what's happening here is they're able to also tell us that, hey, there was a view through attribution for this particular device, and how that's going to help is even though they saw a keyword that they searched on, even though they didn't click on it, we are able to still report on the fact that there was a view of that ad itself.

David Barnard:

Gotcha. Is that going to over attribute to search ads? Is that, I think I saw a few people on Twitter commenting on maybe over attribution or is that just an additional signal that you're going to be able to sort differently than the click through attributions?

Shane Ly:

Our comment here, or my comment here is going to be more that they are leveling the playing field. So Meta, TikTok, Google all do view through attributions today. It was something that was asked for by various advertisers itself, but now they're able to see the performance of the view through attributions. Regardless, you're still able to filter these within the dashboard, so you see view throughs and you see click throughs anyways. So I would not call that over attribution. I would essentially say that it's an additional signal itself.

David Barnard:

Cool. Yeah. Yeah, so it sounds like more helpful than anything else. It's great to see Apple evolving on things. I think there's many fronts where we would hope Apple and Google will continue evolving. Let's move on to some of the bigger networks. TikTok has been in the news a ton this year and a ton of people are finding more success. I mean, I feel like it's really growing as a user acquisition channel, not just for trying to go viral and organics, but also actually running campaigns on TikTok. What do you see going on in TikTok?

Shane Ly:

So for 2024, we'll talk about 2025 in a second here. 2024, there were quite a lot of developments itself. Number one, they became a self-reporting or self attributing network. So essentially they're responding back to us with, hey, we did see attribution itself. Typically, when that happens with a self-reporting network, there's a device ID that needs to be transmitted. IOS, the issue there is that you now have to prompt on both sides. So essentially we have signal loss there. For 2024 going into 2025, we now have them as a advanced SRN. So essentially what happens here is there are additional signals that are sent over to them to make attribution itself. So we're seeing quite a big lift in terms of attribution through that method. Essentially, TikTok was the first one that did it with us. Following suit, Snapchat also did that with us and then Meta as well, and hopefully Google pretty soon. But essentially with the advanced SRNs, we are starting to get more signals itself by sending additional parameters over for attribution.

David Barnard:

Gotcha. And so you mentioned that getting those additional signals is helping drive more efficient spend and being able to get the signal quicker and more accurate from that SRN integration?

Shane Ly:

Correct. We're getting the signals quicker because we're not reliant on SKANs 24-hour or delay mechanism. In addition, you're able to LTV the users. So we do abide by any of the user data retention policies. They all have different retention policies, but you're able to LTV the users for a subscription app. What that means is now we could see that, hey, they did subscribe seven days later and then they did renew month two, month three, month four for instance.

David Barnard:

I imagine that's super helpful. Moving on to Meta, I feel like Meta seems like the network that is most quickly and most effectively recovered from ATT over the past few years, and I just hear so many apps who are doing well on Meta these days that the signals are getting stronger, the algorithms like figuring it out. What are you all seeing on your end as far as the effectiveness of Meta and any changes to that?

Shane Ly:

On the Meta end itself, they are one of the larger ones aside from Google itself. So essentially what we're seeing here is just more attribution methods. There are Web to app plays itself, so sometimes advertisers do send them to a web landing page first and then ultimately into their app for subscription. Sometimes they do a subscription first and then into the app, and then they do send them into the app itself. So there's various methods of doing this outside of SKAN. You could still run SKAdNetwork. With SKAdNetwork, you do have that limited window itself. So we do see advertisers taking advantage of multiple ways of doing this. So just to recap, it is app install campaigns with Advanced SRNs. In 2025, that's a very big signal recovery. There still are web to app plays with meta itself, so essentially sending to a web landing page and then ultimately converting and in parallel, there's still SKAdNetwork that can be ran as well.

David Barnard:

Probably should be a completely separate topic, but what's going on with SKAN? Is it getting more effective? It does seem like more of the big networks are adopting it. Yeah. Tell me about what you're seeing work and not work as far as SKAN goes.

Shane Ly:

SKAN, SKAdNetwork, we are seeing a lot of networks coming on. It's over a hundred now itself. Usually the larger ones, Meta Google, TikTok for instance. Essentially that's still the deterministic way of getting, hey, where did your users come from? The limitations here are just going to be that you're only able to measure up to a certain amount of time. So networks on SKAN 3, you're able to measure 48 hours essentially of data. And then for networks on SKAN 4, you're able to measure up to 35 days of data. However, on your second and third post backs, what's happening there is you have less fidelity. So for a subscription app it might be, did they subscribe or not? There's not much more details that can be put into those post backs.

David Barnard:

Do you see any networks where that 35-day window does make a meaningful difference? Does that first renewal on a monthly or continued engagement, does that get fed back into the algorithm in a way that actually helps better train the algorithms? Or is it one of those nice to have but maybe doesn't actually impact things?

Shane Ly:

For the larger networks, we have not seen them using that signal on SKAdNetwork. We do see it on our dashboard. We do see performance itself, but we haven't seen that actually fed into the algorithms, possibly due to the fact that when you are looking for signals downstream, there are privacy thresholds itself. So you don't fully have accurate subscription data. You have very accurate install data. So users did come from a particular network. However, when it actually comes to did they subscribe? You could be missing 10% of that traffic. We do have dashboards that show that with a null conversion rate. However, that is something that's still an issue today with SKAdNetwork.

David Barnard:

Has the industry solidified around particular best practices or schema around SKAN or any tips about apps trying to use SKAN?

Shane Ly:

It hasn't been a standard around a template. It's still been here are six bits or 63 conversion values. Let's configure that. Where measurement partners come into play here is that we make it easy to actually configure those itself. So you can see at least within AppsFlyer UI, you have the ability to configure, do we want to measure revenue? Do we want to measure just the event? Do we want to measure a funnel? Et cetera. So instead of having developers try to figure this out, essentially we're having a marketer/advertiser come into our UI and essentially configure that in our conversion studio itself.

David Barnard:

Well, moving on to the next big player in the space, Google. I know Google's actually been making a lot of moves the last couple of years. So tell me about what's working with Google.

Shane Ly:

Probably the biggest thing that Google's done in the last year from a attribution perspective is that YouTube now serves an ATT prompt. Very exciting for us because historically Google did not do any prompting. So we're not able to, any measurement partners not able to make measurement with Google using traditional methods. And reason for that is there's no device ID on the Google side. Now that YouTube has it, that's been a huge advantage. We are now starting to see attributions to YouTube itself, and you're able to get the same signal. So essentially if we do attribution, you're able to lifetime value the users itself.

David Barnard:

Very cool. What kind of opt in rates are you seeing? I know you all share data and it's been cool. I've seen a few reports here and there over the years of opt in rates and double opt in rates. But is it meaningful? And I guess even if the percentage is low, you at least have a sample to help inform other decisions. But what are you seeing as far as opt in rates and how that affects efficiency?

Shane Ly:

So a couple of things that we typically advise is showing the prompt. Opt in rates itself, it can be anywhere from somewhere low, 10% to 50 to 60% depending on app. We have seen really high acceptance rates on certain apps itself. So we would still recommend prompting, and ultimately it's another data point that you are able to capture and possibly reuse. So if you're using, maybe you might be doing re-engagements at some point in the future. So that would be another signal that you could send over to those networks to re-engage this user.

David Barnard:

So we've talked about the three biggest players in the space. I mean, putting aside Apple Search Ads and Google Play Ads, which Google Play ads run through Google anyway, but with those three biggest networks, TikTok, Meta and Google, are you seeing any particular trends of increasing spend on Google versus Meta or TikTok versus Google? Any trends in shifts of spend or apps being able to scale on one network but not another?

Shane Ly:

We're just seeing overall more spend going into all three itself. So it's nice to see, and a reason why is because now they're a little bit more confident in terms of, hey, is this actually driving performance. When iOS 14 first came out, spend was held back, even though you put money in, they don't know. Is it driving results itself? So now, 2024, 2025 with additional data points, we now are seeing advertisers invest a little bit more confidently itself.

David Barnard:

That makes sense. As you're more able to more confidently attribute that spend, of course you're going to spend more. So it makes a lot of sense. And it is interesting to hear that you're seeing that just across the board versus any one network running away with share of spend. Reddit is another interesting topic. And I don't hear a lot of apps talking about it, but it feels like that could be another huge opportunity and maybe between Reddit, Snap and maybe there's a few other places, any thoughts on some of these smaller networks where there might actually be some kind of arbitrage and opportunity to spend where maybe others are overlooking?

Shane Ly:

So Reddit has pretty much kickstarted their ad network this year, so they've spent a lot more time on it, particularly app install campaigns. So they've grown quite a bit. They became an SRN in terms of doing attribution. They are investing pretty heavily in that. They have a large user base. So it's definitely one of those networks that are up and coming and something that should be considered for subscription apps. I mean, people are going in there, they're reading the subreddit itself, they're allowing for promotions on those subreddits and measurement. With a measurement partner, you're going to be able to see was that an effective spend?

David Barnard:

So pulling all this together, we've got a lot of different channels, a lot of opportunity, but folks listening to this podcast got to figure out, where do we spend money? How do we spend money? What are the tactics that we need to be thinking about to help drive that efficiency? So any kind of specific advice for folks as they navigate those waters.

Shane Ly:

I think the important part here is going to be measurement overall itself. So having a place where you're able to see that performance in an unbiased way. So essentially we want to be able to compare Meta with Google, with TikTok, we also want to be able to see your influencers or your programmatic buys, et cetera. So having all that in one place is very important. Also, any re-engagements, like let's just say you acquire these users, but they haven't made that subscription. Maybe there's an email push that gets sent out to them, maybe they re-engage on another network, but those are also important tactics just to make sure that we first acquire the user and then we get them into a subscription itself.

David Barnard:

Any other top tips, especially for subscription apps, which is going to be the majority of our listeners, to make sure that you're driving the ad spend effectively?

Shane Ly:

Probably the biggest one is just to make sure that the data pipelines are set up. So essentially having a measurement, a mobile measurement partner's SDK, essentially that will help us track the installs itself. Having a good pipeline for how to notify that a subscription was actually done back into this measurement partner. Reason for that is we need to effectively know where did that user come from and did they actually convert or did they actually subscribe? So typically here we see trial starts and then subscription converted. We do also want to monitor channels where subscribers are churning as well. So those are all effective signals, but essentially the biggest tip here is to make sure that we have that data pipeline set up so you're able to see this measurement end to end. So where did they come from? Did they actually subscribe? And is there a particular channel that maybe we should defocus because a majority of the users are churning itself?

David Barnard:

I mean, it's a huge problem and we've talked about this a lot on the podcast in other ways, but sometimes you'll drive a campaign where it looks amazing and free trial starts are off the charts, and then people are canceling that trial. So having that data pipeline set up well so that you see if a channel is driving those low intent users who are just trying to use your free trial and aren't actually intending to subscribe, it's just so important to get that measurement piece right, is to have that data fed in there. All right, well, as we're wrapping up, anything else you wanted to share?

Shane Ly:

That's largely it. Just wanted to mention, AppsFlyer is the leading mobile measurement partner. We're in most majority of the apps out there today, so definitely if want to learn more, hit us up at appsflyer.com and we'd be happy to dive into details. That's all.

David Barnard:

Awesome. Well, thanks for joining me today. I really appreciate how much data AppsFlyer does share with the industry. You all regularly share reports and blog posts, and part of the reason we're doing this podcast is that you all are sharing some data that's going to go into RevenueCat's State of Subscription Apps report. So definitely worth checking that out. If you're listening to the podcast and haven't seen the report yet, checking out that data, it's going to be really interesting and then keeping an eye on what AppsFlyer shares to make sure you're paying attention to trends in the industry and stuff like that. So do appreciate AppsFlyer sharing the way it does.

Shane Ly:

Thank you so much, David. Thank you for having us. Lovely being on the podcast with you all.

David Barnard:

Thanks so much for listening. If you have a minute, please leave a review in your favorite podcast player. You can also stop by chat.subclub.com to join our private community.