Memorable, not moving: study finds image stickiness outranks emotion in digital ads

A digital banner that a viewer can still picture an hour later is more likely to move them to the checkout than a campaign engineered to make them feel something. That, at least, is the conclusion of a review published in the International Journal of Business and Systems Research, which argues that memorability, rather than emotional intensity, is the dominant driver of purchase intent in online advertising.

The finding lands at a moment when advertisers are being asked to do more with shrinking attention budgets, and when the platforms themselves have quietly shifted the optimisation target from clicks to completed view-throughs. If distinctive imagery is what sticks, the economics of the creative brief change accordingly: brand teams that win are the ones paying for distinctiveness, not sentiment.

The study, in plain terms

The review, summarised on 21 July 2026 by Phys.org, synthesises existing experimental work on digital advertising and concludes that consumers are more inclined to follow through on a purchase when they retain a clear mental image of the ad than when the ad evokes a strong emotional response. Emotion, in this reading, is neither necessary nor sufficient: an ad can be stirring and still fail to register, or flat-feeling and still convert, provided the imagery is sufficiently distinctive to lodge in memory.

That is a narrower claim than the industry conventional wisdom, which has spent two decades treating emotional resonance as the gold standard. The paper does not deny that emotion helps. It argues that memorability is the prerequisite, and that the lift attributed to "emotional" campaigns in older studies is partly an artefact of the fact that memorable ads tend, on average, also to be the more emotional ones.

What the counter-narrative looks like

The dissenting view inside the marketing research community runs as follows. Memorability, critics say, is downstream of emotion, not independent of it. Arousal, in particular, is one of the best-established amplifiers of memory encoding in cognitive psychology, and a creative campaign that evokes fear, humour, or tenderness is, almost by construction, easier to recall than one that does not. On this account, what the review is identifying is not a separate variable but a correlated one.

There is also a methodological objection. Online purchase behaviour is mediated by recommendation systems, retargeting pixels, and price comparison tools that no experiment on a static banner fully captures. A study that holds platform context constant risks mistaking a feature of the test environment for a feature of the consumer. Marketers who ignore that risk will over-fit creative to lab conditions and under-fit it to the auction in which the ad actually runs.

The structural frame

What sits underneath both readings is the same platform reality. The major ad networks have spent the last five years moving budgets away from interruptive display and toward short-form video and shoppable formats, where the creative is judged not on whether it interrupts but on whether it converts within a measured window. In that environment, the metric of merit has migrated from recall to action, and from action to attributed revenue.

A finding that memorability outranks emotion is, in that sense, an argument about which inputs to that chain are doing the work. Distinctiveness is what lets an ad survive the auction. Emotion is what lets it survive the encoding step. The study reviewed here suggests the auction is the more binding constraint, at least for purchase intent measured at the point of exposure.

What to watch next

The practical question for brand teams is whether to reweight creative review criteria toward distinctiveness tests (forced-exposure recall, image-sort tasks) and away from emotional-arousal panels. The research base for either move is thinner than vendor decks suggest. Replication outside the lab, on live platform inventory, is the test that matters.

It is also worth flagging what the published summary does not address. The paper is a review of existing work, not a new primary study, and the underlying experiments vary in stimulus type, population, and outcome measure. The headline finding is robust at the level of direction; the magnitude, and the boundary conditions under which it holds, are less settled than the press cycle around the review implies.

For now, the safer inference is the narrower one: in digital display, being remembered is a precondition for being acted on, and creative teams that treat distinctiveness as a side constraint are quietly subsidising the rest of the funnel.

This publication treats the finding as a research-input, not a verdict: brand decisions made on a single review should be tested against live platform data before being committed to a planning cycle.