Last week, The Guardian published an article which challenged how we judge children’s screen experiences. Marc Guldimann, CEO & Co-founder of Adelaide, responds, exploring what the next generation may be telling us about technology, pace and brands.

Natalia Kucirkova studies children's media, and she thinks parents may be judging quality by the wrong things. Screen time tells them how long a child watched. Labels like “educational” tell them something about the content. But neither captures pace, which Kucirkova argues is a critical dimension of the experience.

Take two cartoons with the same runtime and the same lesson. In one, a character takes a toy and the story pauses on the other character’s reaction before the apology, giving the child time to process it. In the other, the apology comes immediately, and the story moves on to the next gag. The message and time spent may be the same, but the experience is not.

In March, the UK's Department for Education told parents of children under five to choose "slow-paced” content without defining it. Kucirkova's group responded with a rubric grounded in developmental psychology and learning science that evaluates features of the show itself, including narrative pace, sensory load, and natural stopping points.

None of those criteria is an outcome. They are properties of the media that research suggests influence what a child takes away from it. Screen time isn’t included because, despite being easy to measure, it only tells you how long a child watched.

Advertising faces a similar problem when duration is treated as a proxy for media quality. Attention duration tells us how long someone looked at an ad, but that measure depends on the person, the moment, and the creative as much as the placement. Using predicted duration as a buying signal can therefore reward longer gaze without necessarily improving media quality or performance.

Kucirkova's rubric offers a better framework. Start with the outcome you care about, work out which media properties contribute to it, and score them before you buy.

That’s the approach we take with AU at Adelaide. We use eye-tracking and attention research to understand which characteristics of a placement influence attention, then feed those signals into a model trained on full-funnel outcomes. The result is a predictive measure of media quality that can be used before an impression is served. Because AU draws on numerous signals working together, it is also very difficult to game.

Kucirkova’s work shows why the easiest thing to measure is not always the best measure of quality. Media metrics deserve the same scrutiny. What do they actually measure, why should they predict the outcomes advertisers care about, and what behavior do they incentivize?