Thinking · Newsletter
An ad buyer ran AppLovin next to six other channels for one client and shared every number. Here's what he found, and what it means for the ads you put inside a game.
Source: AppLovin Axon ads, I tested these so you don't have to by Guaranteed PPC on YouTube. The test and the numbers are his. We've pulled out the lessons.
The scoreboard
He used to think app traffic was junk too. Then he saw enough advertisers converting on it to stop excluding it on sight. His client, a student loan forgiveness service, is exactly the kind of offer that can pull someone out of a game.
One month of data for one client, pulled from Google Analytics on a first-click basis. AppLovin had been running for 3 to 6 months at this point. Treat it as a case study, not a benchmark.
Before you buy
AppLovin is the biggest mobile ad network, and it runs video inside game apps. That reach comes with a high entry price and a tough audience, so the first four lessons are about whether to go in at all.
AppLovin asks for $20,000 a month to start. Google Display Network can reach game apps with no minimum, so you can see whether that traffic converts for you first.
The fixPut game-app placements in their own Google Display campaign with a capped budget. Scale on AppLovin only if it converts.
Accidental taps happen more on phones, but as he puts it, you're not really paying for a click. You're really just paying for an impression.
What counts is whether the people who see the ad engage and buy. The average mobile game player is 40.
The fixJudge the placement on conversions and ROAS, not on how many taps look accidental.
An ad that interrupts play mostly annoys people. It works for things they haven't seen before, like a toothbrush that brushes your teeth for you
, products up to about $500, or a new service such as student loan forgiveness.
The fixIf your product wouldn't surprise someone mid-game, leave this channel for later.
AppLovin converts below the big platforms, so it pays when they already beat your break-even by a wide gap. His example: you need 500% ROAS to profit, and you're getting 800% on YouTube and 1,000% on Facebook.
The fixIf you're only just breaking even on your main channels, make those more efficient first.
The ad itself
A player can close your ad in a second. Once they've left the game, you finally have time to talk. These two lessons are about getting them there.
Ads that explained the new law behind loan forgiveness did worse than a simple line: you can get your loan forgiven, and others already have. The vague version made people want to find out more.
The fixMake the click the ad's only job. Do the explaining on the page it lands on.
Retargeting people who already visited is quite hard to not make money
on, even inside apps where cold traffic is weaker. Most buyers need several touches before they buy.
The fixPut warm audiences in front of more ads before you chase new reach. Frequency beats reach.
What this means for your creative
“Once you got them off of the app, away from playing the game, now you have time to communicate with them.”
Guaranteed PPC
Most of these lessons land on the creative. The ad gets a player to tap, and the page they land on does the explaining. That's why we build the ad, the end card, and the landing page as one connected test, then improve it from purchase data.
Made for the tap
Measuring and waiting
In his words, 90% of the results come from feeding data into the system. The last four lessons are about measuring honestly while it learns.
If more than 30% of your revenue is coming from one source, you're in a really precarious situation.One suspension can stop everything overnight. His goal is five or six channels, some on a maintenance budget.The fixAdd channels one at a time as each one plateaus, and keep the weaker ones ticking over.
Book a call and name three rivals. We'll read your live ads against theirs: what you all say, who's on screen, and what's lasting.