Paid UA
Predictive LTV isn't necessarily what the algorithm on the ad platform side can work with the best because you need to convey the variance.
"Usually kind of just predictive LTV is, I mean, first of all, it's hard to predict LTV, and then in the end, predictive LTV isn't necessarily what the algorithm on the Meta or like the ad platform side can work with the best. So, kind of bigger like signal engineering companies like Jurnney or like Day30 orientist, they, of course, kind of predictive LTV is the root of all of it, but then they tweak it in a way that they see the algorithm behaving, saying, for example, well, I don't know if my top payer is only 2x the LTV of my lowest payer, which in subscription apps like usually there isn't as much variance, then they kind of 10x these values or in the end kind of skew them in a way just so that the algorithm on the app platform side reacts better to it."
State of Web2App 2026: Funnels, Pricing & Subscription Growth at 43:41 · Featured in Growth Gems #140
Sylvain's take: Marcus mentioned companies like Churney, Day30, or Voyantis. And that this really only makes sense at a high scale.
Paid UA
If you can't build a full web funnel, use web redirect campaigns (landing page → app store) to access different ad inventory and older, higher-spending audiences and unlock placements like Facebook Feed that are difficult to reach with standard app install campaigns.
"Just using a web redirect is another way to access these benefits without even having a web funnel. So if you're saying, well, maybe a web funnel isn't a fit for my vertical, or we don't have kind of the resources in-house to build it, then this can be a really good way to basically run your app funnel but as a web campaign and then still access that different inventory, and for many, many apps, they tap into more relevant inventory because usually these campaigns are skewing a bit older, um, they have a more kind of their spend patterns tend to be a bit more diverse, so usually you're accessing additional placements like Facebook feed, which is hard to go after with app promotion campaigns."
State of Web2App 2026: Funnels, Pricing & Subscription Growth at 29:16 · Featured in Growth Gems #140
Sylvain's take: Several ways to do this nowadays. A straightforward one is to use Appstack.
Paid UA
Splitting out funnels and having a different pixel for each allows you to have very distinctive approaches/angles and help Meta find different audiences for different products or personas.
State of Web2App 2026: Funnels, Pricing & Subscription Growth at 20:20 · Featured in Growth Gems #140
Sylvain's take: Marcus did mention you want to still ensure that you send enough signal via each signal. Not a problem when you’re at scale.
Andrey mentioned training a general pixel for “general audience”. After you start getting some conversions
Paid UA
Think about the questions you could ask during users' first session (within 24 hours) that are likely to indicate which are a better fit (e.g., urgency of problem). If you see this materialize in conversion numbers, then you can use that for your "qualified" event.
"Um I haven't like tested much with it yet, but I also kind of I really like how like in B2B sales um usually leads are qualified by asking them about like the urgency and the severity of their problem. So in the end um in any kind of salesunnel there would be a question around like how long has this issue been going on? Have you tried other solutions looking into are. They like well informed high intent and then how severe is it? Like how much is this impacting your daily life? And I think kind of that's a way to also think about like this trial um um event engineering um through onboarding. Like really what are questions and events. You can input in that first session that will then likely tell you who the better audience is and then if you're actually seeing them. Convert. Better then start optimizing for them."
Signal Engineering: How to Optimize Ad Campaigns with Smarter Events and Value at 38:50 · Featured in Growth Gems #124
Paid UA
You can optimize an app promotion campaign on Meta and optimize for conversion API events (CAPI), but through incrementality tests there's evidence that there are less CAPI events attributed vs. using the Meta SDK.
"Thomas has already mentioned, but yeah, things just tend to be a lot more easy with like conversion API and if you can trigger server to server events and then you don't have to figure out all these kind of nitty-gritty when does the SDK how efficiently with scan even more so. Um I actually currently also validate with a client that you can even optimize for you can run an um app promotion campaign on meta and optimize for a conversion API event. However, in my incrementality testing, I've seen that in the end much less events get attributed. So the SDK caught like way more of the incremental value which in the end means more signal, better targeting. So yeah, even despite us sending as many identifiers as we could through conversions API, it didn't get there in terms of that kind of match rate. And of course that's going to be an issue in the long run. And we're going to choose the more efficient setup there. While also my client, of course, they they would love to send everything through conversion API, handle it on their end, and not even have to deal with the meta SDK at all. But at least with them, it didn't seem like it would be benefiting us. I wish."
Signal Engineering: How to Optimize Ad Campaigns with Smarter Events and Value at 53:25 · Featured in Growth Gems #124
Paid UA
On Meta, start with AEM rather than SKAN, especially if you're at a small scale. AEM doesn't have the same drawbacks as SKAN: delay in events arriving, privacy thresholds, etc. It's best to avoid using too many third parties as it makes the setup more complex and error prone, which mean the Meta SDK might be the best.
"Um generally I would advise to start with AEM just because it doesn't have all the like drawbacks of scan meaning like the um the delay in events arriving and then having to pass privacy thresholds and especially as you're starting out that can be really a pain if you're like at small scale and then figuring out like basically what campaign performance actually looks like um and enough signals arriving for your kind of campaigns in order to optimize then targeting properly. So my default is AM. While I do see um value in trying scan, especially now, I haven't actually tried the new version yet."
Signal Engineering: How to Optimize Ad Campaigns with Smarter Events and Value at 17:35 · Featured in Growth Gems #124
Sylvain's take: Meta doesn't report a “match rate” like they do for web events, and Marcus shared in this post what you need to do:
Paid UA
Sending back additional signal/events to the ad network allows you to check on user behavior throughout the funnel for different optimization events. Campaigns won't take into account these events for optimization, but it's useful for your analysis. Don't overdo it: send the key funnel events.
"So yeah, do you degrade signal by sending too many signals or yeah, what do you do, Marcus? I would usually say like send more because in the end that will allow you to do like deeper analysis on your campaign performance. Like there's always going to be these little quirks where you see. There's a campaign which has a good uh cost per action that you're optimizing for, but then the funnel seems to break in some other point compared to other campaigns. And then that's the stuff you're looking for in terms of like how you're going to um inform your creative strategy and what kind of you can look then again into like audience. Breakdowns looking and like I don't know do older. People drop off at a certain point um which again can inform like other parts of your like product development as well. So I'd say send more um. The campaigns are going to use the signal that you're optimizing for. So from that perspective that kind of shouldn't break anything.Um, yeah. So, my advice, send more, but then like don't overdo it."
Signal Engineering: How to Optimize Ad Campaigns with Smarter Events and Value at 1:05:27 · Featured in Growth Gems #124
Paid UA
When optimizing for trials, platforms like Meta often deliver young users who start free trials but cancel quickly because they're cheap to acquire. Use data breakdowns to identify these patterns, then either exclude in your targeting (e.g. 25+ targeting) or use a "qualified trial" event (e.g., only send the event for > 25yo).
"What I like, what I really like about Meta is that often times you can already find answers to why maybe your down-funnel metrics don't look as good in like data breakdowns. Oftentimes, what's the case when you optimize for a trial is Meta gives you very young traffic because they love to start a trial. It's free when they start, but then they cancel quickly. And young traffic is cheap. So, in the end, oftentimes you see very low cost per trials from young audiences."
Signal Engineering: How to Optimize Ad Campaigns with Smarter Events and Value at 36:25 · Featured in Growth Gems #124
Paid UA
Having a campaign optimizing for web events (Sales campaign with CAPI to send back signal) in addition to campaigns optimizing for app events derisks your setup because you can shift budgets in case of algorithm changes (like what was observed recently with Meta over-targeting Facebook reels).
"I think it's important to have at least a campaign optimizing for web events just as much as for app events. So, in the end, if you're running like a um a web quiz, then you will be sending that data through copy um and then that will allow you to run a sales campaign instead of a app promotion campaign. And things. Happen in the algorithm. I just recently had a case again where kind of placement targeting got messed up. I had like 10 15 apps reporting the same thing. I think I'm sure it happened to more uh prices spiked and that was just um issue on meta's end with the app promotion product. So if you then have your budget uh across two campaign types. Then you're more likely not to see it on both and you can easily kind of um derisk and then shift budgets. If all your eggs are on one basket, then sometimes that can be a chaotic week as experienced."
Signal Engineering: How to Optimize Ad Campaigns with Smarter Events and Value at 20:45 · Featured in Growth Gems #124
Paid UA
As you test new optimization events, also check the evolution of placements and demographics in Meta (e.g., age) to see if they do indicate you're targeting and acquiring the higher quality audience you're going after.
"Again what I like in meta is that also we can quickly crosscheck our assumptions there based on data breakdowns. For example, now I'm running campaigns against these goals and I can look at the audience targeting of these campaigns and I can see how my age targeting gets older the deeper the event is in the funnel already showing hey. Meta picked up on this going after a higher quality audience which I know because I also have data on older people converting better. So that's a very good early sign that we're actually driving the audience that we're trying to go after. So again, like that's that's why I'm just a big fan of meta and like you're able to like dig into this and figure out um at least for now still um yeah kind of what's what what's the traffic you're bringing in and do these assumptions actually make sense. Meta ads fanboy. I would wouldn't have expected that. That's why you build yourself as a meta ap"
Signal Engineering: How to Optimize Ad Campaigns with Smarter Events and Value at 52:15 · Featured in Growth Gems #124
Paid UA
There can be value in having multiple events different campaigns optimize for to broaden who you're acquiring and address a bigger market: some might convert and monetize quickly, other over the long run.
"Yeah. And I think even if you're not like multiplayer and like your app is more like classic just subscription app, then still there can be value in having multiple events to optimize for to like broaden like whom whom you're attracting and then some of them might um convert quicker and like monetize quickly. Others do it in the long run. So over time as you learn like which signals kind of um convert into which behavior then you can also make the decision of saying hey. Let's in the end optimize for both because in the long run that's going to help us address a bigger total addressable market."
Signal Engineering: How to Optimize Ad Campaigns with Smarter Events and Value at 29:32 · Featured in Growth Gems #124
Sylvain's take: On the same topic, Thomas later mention this:
It's not necessarily about finding the 1 perfect optimization event/signal, it's about utilizing them in different ways. There are cases where you might:
Want to have 2 campaigns optimizing for the same qualified trial event but with different targets depending on the audience (e.g., > 25yo and < 25yo), just with different target CPA
Use campaigns with different events to attract different audiences, each with their target CPA
I would add that you should also experiment with different events for your web vs. app campaigns (as mentioned last week if you missed it…that’s why you shouldn’t skip the sponsored insight 😂).
Paid UA
If you have a funnel/quiz with purchases on the web you need to be using Meta's Conversion API because the Meta pixel won't capture all the signal. Check your event match quality in the event manager and potentially send additional signal to help attribution.
"And if you have like a quiz and you convert people on the web. Then you should definitely have copy in place because the pixel is not going to capture a lot of these purchases. And you can actually check in events manager your kind of event match quality on on web. Facebook has a really good reporting around it and they tell you hey we think. This is your kind of match score and we uh recommend sending these additional things so that we can increase it and with a pixel you might get up to like a three out of 10. While with copy you can kind of go up to like nine. Um, so definitely like add those identifiers so that Meta knows what who's purchasing and can then uh attribute back to to the campaigns.All right, here's a here's a fun one that won't be controversial at all. What is your favorite MMP and I I'll add and why or maybe maybe a better question uh."
Signal Engineering: How to Optimize Ad Campaigns with Smarter Events and Value at 1:15:55 · Featured in Growth Gems #124
