Soundtrack:

Alright, we’re back, picking up where we left off after last week’s Part One of my look at AppLovin.

Now that we’ve talked about AppLovin’s acumen in the world of mobile game ads and what that LITERALLY LOOKS LIKE, let’s talk about where they’re going next (e-commerce) and if they’ll make it.

Table of Contents

• 100% Wrong, 35% Right

• Churning, Muddy Waters

• General List of Other, Well-Worn Grievances

• Trailing, or Trouncing, The Trade Desk

• Must CTV? Bust: A-P-P?

• Becoming A Household Brand

• What Should E-Com Advertisers Do?

• What Should Ad Tech People Do

• Is This Good For A Healthy, Free Internet?

• Firing My First Shot At “Independent Ad Tech”

100% Wrong, 35% Right

To be fair, let’s start by seeing how AppLovin themselves are appraising the progress of their foray into e-commerce.

Ah, well, this is awkward…

Trying earnestly to take them at their word kind of backfires for them, because this means we have to talk about a statement from their CEO that is a huge problem.

If you don’t immediately see the problem here, it’s daring to utter the phrase “nearly 100% incrementality.”

No channel is 100% incremental for a major e-commerce brand, and none are even close to that. This isn’t an acceptable “fudge” or an understandable minor exaggeration.

If you come to know me or just read this blog regularly, one thing you will learn about me is that I am very old fashioned in one regard: I think carelessly making statements that clearly couldn’t be true is bad.

I think that so many tech CEOs do it so much these days is actually having negative consequences on our tech industries.

I think it’s warping expectations and creating an environment where potentially great products are being smothered by fake hothouse flowers blocking all their sun, only to wilt themselves after choking out better ideas with shade and overabundant venture capital.

Once we move beyond the false claim, though, things become nuanced, and it’s not quite a black-and-white picture in terms of if AppLovin have an incrementality problem or not.

The MuddyWaters report says that they think the actual percentage of sales that are incremental is 25%-35%.

In the long history of fraud-laden digital marketing platforms, those are actually good numbers.

Not only do they not get you on the Mount Rushmore of Attribution Gamers, they get you laughed out of the Credit Claimers Clubhouse.

35% is an incrementality number that, especially if it’s just a start before incrementality optimization really happens, a brand could really work with.

As KnoCommerce cited in its own case, it’s at least more incremental than Google Search usually is when a brand first measures it.

MuddyWaters glosses over this figure actually being decent, instead focusing on the 65%-75% gap in reality and the CEO’s claims.

And I get it! Like I said, I think AppLovin’s CEO making that wild incrementality claim damages their credibility in regards to any and all claims they make about their e-com beta results going forward.

That being said, this is kind of like if a baseball player said they were going to be the greatest player of all time, and then hit .247 with 29 home runs. Obviously they overestimated their own talent, but plenty of teams will take that performance.

Even my beloved Atlanta Braves need that kind of performance, sometimes:

Churning, Muddy Waters

There’s another element to the MuddyWaters report that may seem loosey goosey to some folks, but that I think is actually a case well made.

They surveyed all the e-commerce sites where the AppLovin pixel had been placed, and watched for when these pixels were either removed, or were still present but broken (implying nobody cares if they work anymore).

The idea here is that this shows which e-commerce customers have churned. They found that 22.6% of customers likely churned in Q1 of 2025, according to this survey.

There’s a lot to be said about this method, but I can attest to one thing that MuddyWaters says being true in my extensive tracking implementation experience:

Clients basically never remove pixels.

You cannot do an audit of any large, long-established website without discovering they have dozens of active tracking tags for technology they no longer use, and often have “active” tracking pixels for companies that literally no longer exist.

So when MuddyWaters say that 22.6% is extremely conservative, and the number is likely much higher, I think that’s true.

There’s an argument that the AppLovin e-com beta advertisers are inherently more vigilant about these kinds of things than the average advertiser, and so maybe the real number isn’t much higher to the extent it would be if we used this methodology for churn tracking to look at the general population of all digital advertisers.

That being said, if the pixel dropped from 22.6% of sites, I could still believe real churn is more like 40%.

I can’t even tell you if that’s particularly great or terrible for a major digital ad platform, and the truth is, I don’t know that anyone could say so. There isn’t a ton of precedent-not many platforms have launched an e-com specific offering after already massively scaling a general mobile ad business.

Here is where we run into a familiar problem that may give us concerns, whether we think 22% or 40% is a good or bad churn number or not.

Here we have a “reportedly” and not the citation of an official earnings call, as for the “near 100% incrementality” figure, so I am not as convinced this is something we can say anyone at AppLovin has officially claimed.

That being said, this is another whiff of irrational exuberance and grand claims, after we’ve already smelled plenty of that coming from AppLovin.

General List of Other, Well-Worn Grievances

• AppLovin allegedly violates the TOS of every app store and ad platform it interacts with by creating a PIG (persistent identity graph) stitched together from other platforms’ identifiers it’s not supposed to be using, and user PII. It’s about as bad a privacy mess as any major digital company might be messing with right now, but the concern isn’t regulation. The concern is that if this is what they’re doing, and they need to do it to succeed, Meta and Google and others can use this dependency to destroy most of their business. In fact, that’s what happened to Cheetah Mobile when they crossed Google.

• There’s a class action lawsuit against AppLovin alleging that they misled investors by claiming their success to date and expected growth was due to their own technology, when in fact it was due to the aforementioned misuse of other companies’ data and technology.

• There are widespread reports of large ad credits being used to lure in these e-com beta customers. I know a lot of you are going to wonder how this is even a problem because it’s a widespread practice, but look: there’s no disputing that it artificially inflates advertiser participation in a platform beyond what performance of said platform would normally dictate. Even if you think it’s fair play, you can’t deny that AppLovin’s e-com beta has wider participation than it normally would due to these incentives.

I know I’ve relegated these to a lightning round, and so I want to belabor a point for a moment:

All of this looks a lot like stuff said in the past about other companies that did, indeed, end up having bad things to hide!

It’s weird to acknowledge all this kind of stuff is bad, and things like this are the reason MNTN has a steaming pile of poop for a future, and then act like it’s not a big deal for AppLovin to deal with!

Speaking of CTV and grim futures…

Trailing, or Trouncing, The Trade Desk

I said earlier that choosing the road that leads away from building as much owned and/or controlled inventory as possible makes AppLovin more like The Trade Desk. Let’s see how they’ll compete.

To give AppLovin some credit, they’ve succeeded almost entirely in a Venn Diagram of things TTD has never really succeeded at: “performance advertising,” and mobile app inventory. It has been noted that it’s almost comical how much TTD is a literally unknown quantity to “performance” advertisers and non-whale-sized brands in general.

If AppLovin flourished in this realm fairly, and not through any of the trickery that has been alleged, that’s a huge long term advantage for them.

The Trade Desk, however, has done three things that no other piece of independent ad tech has succeeded at.

1. Unapologetically made the “open web” and all of its (historically low value) inventory part of their core mission and vision, and succeeded in becoming one of the principal arbiters of what is quality open web inventory, profitably.

2. Hedged this bet magnificently by bringing programmatic buying of a lot of CTV inventory, coupled with hot items like retail media data, to the enterprise advertiser set.

3. Baked measurement tools for outcomes like Brand Lift and proxies like attention much more deeply and effectively than anyone besides Google, and Google have the massive advantage of tons of proprietary tech and the world’s largest search engine to help them out with that.

The open web inventory is where I could see AppLovin being most competitive, but even there, I see two problems.

One is that this is the smallest prize. As much as Jeff Green and company wave the banner about saving journalism and preserving the non-walled garden web, every quarter that goes by this seems more and more like a symbolic commitment than TTD’s real future cash cow.

The other issue is that TTD have done this not through pixel or API based online conversion data, but through the notions of Curation and Quality. This is the polar opposite of AppLovin’s approach to things, and historically a “pure performance” approach to the open web has been disastrous, ESPECIALLY when combined with an (alleged) proclivity for attribution gaming.

Must CTV? Bust: A-P-P?

On CTV, AppLovin’s problem is basically the same one of not being equipped to help “curate” “quality” for human buyers, although not everyone understands this yet, because there are still a lot of people pushing the idea that CTV should be as “open” and “real time” or “purely programmatic” as the open web, or large social media platforms.

This is a load of horse hockey sold by people who have sunk cost into technology to “make CTV programmatic” in the most open-web sense of The P Word, and by now even these people have realized that isn’t realistic, but they’re on an F Series fundraising round, as in they are F’d if anyone gets even the slightest whiff that this isn’t a sure thing, and so they will say whatever they need to in order to keep the roulette wheel spinning.

There’s not as much CTV inventory in general as anyone thinks there is, there’s even less of it if you want to use any kind of reliable targeting data, and all the best CTV campaigns out there are run using lots of Guaranteed Deals and custom inventory packages and full-on human negotiated inventory blocks.

So AppLovin showing up and muttering “AI magic, do your thing” at the CTV ecosystem, without the massive amount of partnerships and infrastructure TTD has, is not going to be such a tremendous opportunity.

Becoming A Household Brand

On the third element, the Brand and Attention and survey measurement suite-this is all completely anathema to AppLovin’s entire philosophy. Asking them to do this would be like asking Meta to give more granular control back to advertisers.

Animals have a nature and AppLovin’s is classic “pure performance” on mobile, not CTV with a Lumen survey about Brand Favorability and a Tracksuit data integration.

There’s an obvious compounding problem and a sneaky compounding problem here.

The obvious one is that mobile advertising powerhouse Meta is, while mostly singing the performance gospel, also building out their “longtermism” and “brand building” use cases, which will eventually even be turned on the mobile gaming market.

The less obvious one is that Google’s UAC is plugged into YouTube, and while it might take a minute, someone is eventually going to have the spark that leads to YouTube’s long term brand building power turned onto Google’s general, more performance marketing oriented customer base.

As I reminded people in my first piece, Gabe Laydon saw a lot of things about mobile game advertising way back in 2016, and one thing he grasped is that TV ads are actually really effective for mobile games at a certain point, but that it’s hard to see this and capitalize on it properly.

I don’t have a timeline for when the three companies above get their Mobile Game Brand Building Stacks together, but I’m pretty sure I know what the podium rankings will be.

P.S. I have linked to different parts of that YouTube video in BOTH of my posts about AppLovin because if you care about mobile game user acquisition and you haven’t watched that video, you need to do that right now. ESPECIALLY BECAUSE it’s from 2016…it will haunt you how it seems like it can’t possibly be that old.

What Should E-Com Advertisers Do?

If you’re an advertiser in a sector that is NOT mobile gaming, I would say you should have AppLovin on your testing roadmap exactly where you would have a couple years ago. No change needed amid the recent hype.

Where exactly that is, I would say, depends on a few things I’ll briefly try and help you navigate.

If AppLovin has an insanely low barrier to entry for you, it’s fine to make it a top ~3 priority after Meta, Google, Amazon, and your other top tier advertisers.

The trap I would advise you to look out for here is to not overlook other channels that have a very low barrier to entry, that you’re not thinking about simply because they’re less hyped in your world right now, i.e. Snap, Pinterest, Reddit, etc.

If, for whatever reason, there is a moderate or high barrier to entry for you getting up and running on AppLovin, that’s a great opportunity to test another channel that has a high barrier to entry but also has better total long-term impact than mobile app video ads.

Like what? That’s what you have to pay me, or someone else, to figure out for your specific case. Maybe ask your cute little MMM.

What Should Ad Tech People Do

Honestly, cool your jets.

I don’t think there’s actually anything terribly new about this AppLovin arc; it’s only interesting because the stock price is oddly high, and that is also not a new phenomenon for an ad tech company.

“But Lee, you say that as you write like 7000 words on the topic!”

Yes, and as I note in my first piece, I am writing these AppLovin pieces for two sequential reasons:

1. A record number of people have asked me, directly, for my opinion on the company and its products.

2. I very quickly got tired of talking about this and wanted to produce the definitive document I could just send the link to if I am ever asked about this again.

To put my money where my mouth is, I will not write another blog post about AppLovin in the year 2025, on penalty of buying anyone who calls me out a coffee or beer if I do so.

Now that you know I’m not going to keep harping on this, don’t you want to subscribe?

Is This Good For A Healthy, Free Internet?

This is the first time I ever get to address the question: could any ad buying platform ever good for the internet?

Of course, I’m going to kind of sidestep it, and define the scope of how we’ll discuss ad platforms for now.

There simply are not, at this time, enough examples of ways to monetize making things and putting them on the internet, without ads, to write off advertising as a business model entirely.

Believe me, I dream of the day we can say “this is stuffed with ads and so it should be relegated to the dust bin of history” but we aren’t there.

Firing My First Shot At “Independent Ad Tech”

Here’s where there’s probably an expectation to say “well, it’s not Google or Meta, and it’s kind of inventory agnostic, so as far as ad platforms go, it’s good for the internet!”

This is where you’re about to learn about a distinct axe I have to grind with the American Democratic Party and “independent ad tech.”

If we want independent ad tech to win out over the megaplatforms and improve the world, it must:

1. Be positioned to actually defeat said platforms.

2. Actually improve the world once it has attained supremacy.

Before anyone gets too worked up about my little image above, please know that I think some independent ad tech companies are on the right path and trying their best and generally good.

I’m far from the biggest $TTD ( ▲ 0.07% ) fanboy but as you can probably tell from my writing, I think they’re generally a net good and mostly honest in their aspirations to do good.

There are other people doing good, or the best they can. If you know me, you know if I think this about you and who else I think it about. Feel free to reach out if you’re looking for clarity in any way.

That being said, independent ad tech is generally a trash heap.

What that trash heap does not need is another five alarm fire to spread stink far and wide so that it can be smelled by market analysts, retail investors, or even the general public.

That’s the potential downside of this AppLovin story-another Rocketfuel but bigger, blowing a hole in the market capitalization and credibility of an entire industry.

The potential upside is another legitimate contender to Meta, Google, and Amazon that gives buyers another viable option to spend a lot of money with, and in turn, potentially gives lots of game developers and publishers another good option to monetize.

I think another non megaplatform option is ALWAYS GOOD, so I would generally love to see it, although I will note that I am not sure how good an actor AppLovin would be in a seat of power like that.

Power corrupts, absolute power corrupts absolutely, moderate power leads you to believe you can do things like make your core UX look like the periodic table of elements and challenge one of the most famous acronyms in the history of the world for ownership of a semantic space.

Disclaimers

This is not investment advice, nor is it legal advice. I am neither a certified financial professional of any kind, nor an attorney. Everything in this article is opinion, conjecture, speculation, etc.

I held no stock in any companies discussed in this article at the time of its writing, nor do I hold any financial positions against any of these companies, i.e. shorts and the like.