Why Interactive Ads Feel Less Like Ads (and Why That Matters Ethically)

Ask people about playable ads and a particular phrase comes up repeatedly, in almost the same words: it did not feel like an ad.

This is usually quoted by people in the industry as a compliment, and taken as evidence that the format is working. It is evidence that the format is working. It is also, if you sit with it for a moment, a fairly strange thing to be pleased about.

The sentence describes a person who was advertised to and did not experience it as being advertised to. That is a gap between what happened and what was perceived to happen, and gaps of that kind are worth examining rather than celebrating, particularly by the people who benefit from them.

So: why does the format feel different, and does the feeling matter?

The intuitive answer is that ads are recognised by how they look — the product shot, the logo, the particular gloss of commercial photography. That is part of it, but it is not the main part, because plenty of things with all those features do not feel like advertising and plenty of things without them do.

The better answer is that an advertisement is recognised by what it asks of you. It is a demand wearing content's clothing.

Consider what an ordinary interruptive ad actually requires. It takes time you had allocated to something else. It requires you to wait before resuming what you were doing. It asks you to receive a claim and hold it. And it ends by asking for an action — go here, install this, remember us.

Every one of those is a small imposition, and the accumulation of them is the texture people recognise. The feeling of being advertised to is the feeling of being asked for things by something you did not invite.

Strip four of the five demands and the thing stops matching the pattern people recognise. That is the whole of the perceptual answer, and there is nothing sinister in it — the ad feels less like an ad because it is, in several concrete respects, asking for less.

The ethical question is not raised by any of that. It is raised by what the fifth row keeps quietly doing while the other four are absent.

Here is the part of this that we think is genuinely underexamined, and it is not about deception.

People are not defenceless against advertising. Over a lifetime everyone develops a working theory of how persuasion operates — that claims are selective, that the flattering comparison was chosen, that the person smiling was paid to. Researchers call this persuasion knowledge, and it is a real competence that develops through childhood and keeps developing.

The important property is that it is triggered by recognition. The defences are not always running; they engage when something is identified as an attempt to persuade. That identification is what moves a person into the mode where they discount, question and hold at arm's length.

Which means a message that is not recognised as persuasion is received without the discount. Not because the person is careless, and not because anyone lied to them, but because the machinery they would ordinarily apply was never activated.

This is the ethical problem with the format, stated as precisely as we can manage. It is not that interactive ads deceive people. It is that they can be entirely truthful and still be evaluated less critically than the same claim would be in a format the person recognised.

It would be easy to write the alarming version of this argument, and it would be overstated. Some proportion is worth keeping.

The stakes are usually low. Most of these ads are for free games that cost a few minutes to try and nothing to abandon. A person who installs something they turn out not to want has lost some storage and a little time. That is not nothing, but it is a long way from the harms that usually justify this kind of language.

There is also a real counterargument, and it is stronger than the industry usually bothers to make. A demonstration is more informative than a claim. Someone who played a game for fifteen seconds knows more about it — accurately, from direct experience — than someone who watched thirty seconds of edited footage. If the worry is people making poorly informed decisions, the format is arguably an improvement on what it replaced.

But the counterargument only holds where the demonstration is representative. It is the same condition every other argument in this category has come down to, and it is doing a great deal of work here: the format is more informative than video when it shows the real thing, and worse than video when it does not, because the false impression now arrives through the person's own experience rather than through a claim they could have discounted.

The asymmetry in that table is the thing worth taking away. Honest interactive advertising is better than the alternative. Dishonest interactive advertising is worse than the alternative — not equally bad, worse — because a belief someone arrived at themselves is held more firmly than one they were told.

The format amplifies in both directions. That is a reason for the people using it to be more careful than the people selling video, not less.

The unhelpful conclusion is that ads should be made worse on purpose — more intrusive, more obviously commercial, so that defences fire. Nobody wants that, it would not survive a single campaign review, and it would make advertising less informative rather than more honest.

The useful version is narrower. If the risk is that someone does not realise they are being advertised to, the remedy is that they can tell — not by being interrupted, but by the ad being legible as an ad to anyone who looks.

We build a tool for making these ads. That is a direct interest in the conclusion, and it would be reasonable to discount what we say here accordingly.

What we would argue is that the format is defensible, on one condition, and that the condition is not a formality. Interactive advertising is more informative than the thing it replaces when it shows the product honestly. A person who tried a game for fifteen seconds genuinely knows more than one who watched a trailer, and knowing more is the outcome any defensible advertising should be aiming at.

But the condition is doing all the work. Without it, every advantage this category has described — the ownership, the memory, the trust, the conclusions people reach themselves — becomes a mechanism for installing a false belief more firmly than a video ever could. The same properties, pointed the other way.

So we would rather the standard for this format were higher than for the ones it competes with, not lower. That is an awkward position for a company selling the tools, and we think it is the correct one. A format that people stop trusting is worth nothing to us, and a format that deserves to be trusted is worth a great deal.

Some limits, because this is philosophy rather than measurement and should be labelled as such.

Persuasion knowledge is well established in consumer research. The specific claim here — that interactive formats trigger it less reliably than interruptive ones — follows from how the model works rather than from a study designed to test it. The reasoning is sound and the conclusion is an inference, which is a distinction worth keeping in view when anyone starts quoting it as a finding.

Nor is the line between demonstration and manipulation as crisp as any of the tests in this piece imply. Every advertisement selects what to show. Every game designer tunes an experience. There is a continuum between showing your product at its best and arranging an experience so that only one conclusion is reachable, and reasonable people will place the line at different points along it.

What we would insist on is that the question is a real one and deserves asking during production rather than afterwards. The industry's habitual answer — it converts, so it works — settles a commercial question and leaves this one entirely untouched.