Fake Gameplay Ads Are Killing the Format — An Honest Argument

You have seen the ad. A character stuck in a tunnel, water rising, two pins to pull in the right order. A parking lot of cars that need untangling. A stack of blocks collapsing toward a pit. Played once, failed deliberately, followed by an invitation to try it yourself.

Then the install, and the game is a city builder with a timer. The pin puzzle exists as a minigame four hours in, or it does not exist at all.

Everyone in this industry knows this happens. Most people in it will say, without much prompting, that it is bad. And the volume of it has not meaningfully fallen, which should tell us something about the argument we have been making against it.

The standard case goes: the ad misleads a person, the person installs something other than what they were shown, that is lying, and lying is wrong. Every step is correct. It has also been in circulation for years without much effect, and it is worth being honest about why.

It asks a studio to absorb a certain, immediate, measurable cost — fewer installs this quarter — for a diffuse benefit landing mostly on other people. The person who was not misled is not their customer. The trust preserved is not theirs to bank. Framed that way, the ethical argument asks one party to pay for something everyone else consumes, and requests like that lose to a dashboard.

There is a second reason, less comfortable. A fake ad often works — not in producing good customers, which has been written about enough, but in the narrow sense that the number it was built to move goes up within days. An argument depending on the practice failing on its own terms loses to anyone who has actually run the campaign.

So the moral case is true, insufficient, and not the interesting one. The interesting question is why a practice everybody condemns keeps spreading. The answer is not that the industry is full of liars. It is that the incentive structure quietly punishes the people who refuse.

Here is the part that gets missed, and the reason this is a structural problem rather than a conduct problem.

A studio does not evaluate its ad in isolation. It evaluates it against what ads in its category achieve — the install rate, the cost per install, the click-through figure circulating as roughly what a decent creative does. Those benchmarks are handed down by nobody. They accumulate from what everybody's campaigns actually return.

Now introduce a fake ad. It advertises a mechanic more compelling than anything the real games contain, because it was never constrained by having to be true. It performs well, and its numbers join the pool that forms the benchmark.

The honest studio down the road runs a truthful ad for the product it genuinely built. It converts at a rate appropriate to what that product is. And in the dashboard, against a benchmark inflated by creative advertising an imaginary game, the truthful ad reads as underperformance.

Nobody told them to lie. The market simply informed them that honest work is below average, in the only language a performance team is fluent in.

This is the whole argument, and it reframes who the victim is. The person misled by the ad is harmed, certainly, and that harm is real. But they are one person with one uninstall. The studio that refused to lie is harmed continuously, in every campaign review, by a standard it had no part in setting.

That is why the practice is corrosive rather than merely wrong. It does not just deceive players. It systematically disadvantages the studios doing the thing we would want them to do, and it does so through a mechanism nobody has to intend.

It is tempting to picture a meeting where somebody proposes advertising a game they have not made. That meeting mostly does not happen. The path there is gradual, and every individual step is a small, locally sensible optimisation that would be difficult to object to in the room where it was taken.

This matters because it tells you where an intervention has to sit. Appealing to the conscience of whoever approves the final creative is too late — by then the decision is between a principle and a number, and the number has evidence.

A misleading video makes a claim. You watched footage, the footage was not representative, and when the product disappoints there is a gap between what you were told and what you got. That gap is available to you — you can remember being shown something and conclude the showing was false.

A misleading playable does something else. You did not watch the mechanic, you performed it. You made the decision, you pulled the pin in the order you chose, and the outcome followed from your input. What you are left holding is not a claim somebody made. It is a memory of having done the thing.

Memories of your own actions are harder to discount than claims other people made, and they carry no label identifying where they came from. So the deception is not merely more effective, it is more durable.

Which means the interactive format cannot afford the level of fakery video has normalised — not because anyone external holds it to a higher standard, but because the mechanism does more damage per impression, and a format known for it gets discounted wholesale by the people it is shown to.

Given all of the above, exhortation is not a plan. If the practice is sustained by structure, only structural things move it. Four are worth taking seriously, and they are not equally likely.

A piece like this from a company that sells a playable ad editor is suspect by construction, and it would be worth very little if it only condemned other people. So here is the uncomfortable half.

Every tool in this space, ours included, makes some ads easier to build than others. And the ad that is easiest to build in nearly all of them is the invented one — a clean, self-contained mechanic assembled from generic parts, unconstrained by any product. Building an ad out of a real game is harder. It means getting the actual assets, matching the actual behaviour, and accepting the actual mechanic rather than a better one you could have designed instead.

That asymmetry is not a law of nature. It is a consequence of how these tools are built, and it means that when a studio takes the easier path, the tool participated in that. We do not have a clean answer to this. We are not going to pretend that a policy document solves it, and we are not going to claim our tool prevents something it does not prevent.

What we can say is what we think the standard is, and be held to it. An ad should demonstrate a thing the product does. Not a thing adjacent to what it does, not a thing planned for a later update, not a thing that tests well. If the mechanic in the ad is not in the game, the ad is for a different game — whatever the numbers say, and whatever the category benchmark has drifted to.