Beyond Reach and CTR: What the Attention Pivot Actually Means
Reach and click-through rate are still on almost every campaign report, and they are not going anywhere. What has changed is how much they can tell you on their own. As more budget moves into video, streaming and formats where nobody is expected to click, the industry has been building a different set of measures around the question these metrics never answered: did anyone actually notice the ad?
That is what attention measurement is for. Here is what it measures, what the evidence shows, and where it still falls short.
Why the old metrics run out of road
The limits are easiest to see in the numbers themselves.
Average display click-through rate on the Google Display Network sits around 0.046%. At that scale, ordinary week-to-week variation can easily look like a meaningful performance difference when it is not. Optimising hard against a number that small tends to chase noise.
Viewability was meant to fix part of this, and it helped. Cross-network viewability under the MRC standard — 50% of pixels in view for at least one second on display, two seconds on video — reached a 72% average in 2026, up from 67% in 2024. But roughly 28% of impressions still fail that bar, and more importantly, viewability only confirms that an ad rendered on screen. It says nothing about whether a person looked at it.
An ad in a fast-scrolling feed and an ad on a focused editorial page can be identical by viewability standards and completely different in reality. Closing that gap is the whole point of attention measurement.
What attention metrics actually measure
The IAB and MRC published Attention Measurement Guidelines developed by a cross-industry task force of more than 200 people from brands, agencies, publishers and measurement companies. The framework sets out three tiers:
Exposure-based — did the ad appear on screen in a viewable position? This is the baseline, essentially viewability.
Engagement-based — did the viewer interact with it?
Outcome-based — did it drive a measurable business result?
Vendors have built their own measures on top of this. Lumen's Attention Per Mille (APM) counts attentive seconds per thousand impressions. Others score environments on a relative scale: one 2025 comparison rated CTV at 58.9 attention units against 52.5 for linear TV, 39.7 for online video and 23.2 for display. Those numbers are useful for ranking environments against each other, not as absolute scores.
Does it hold up?
The outcome evidence is reasonably encouraging.
Adelaide reports that campaigns optimised toward attention signals delivered around 41% higher brand lift and 55% better lower-funnel outcomes than campaigns optimised for viewability alone. IAS found that campaigns using attention-based strategies recorded a 56% increase in attention scores alongside a 76% drop in cost per click, with household-level sales lift running significantly higher for high-attention video placements.
Independent validation is catching up too. The Advertising Research Foundation released Phase 3 of its Attention Measurement Validation Initiative in June 2026, testing attention across four national campaigns spanning television, online video, social and digital display — one of the more thorough real-world assessments so far.
The caveats worth knowing
This is where a lot of attention marketing gets ahead of the evidence, so it is worth being direct about the limits.
Scores are not comparable across vendors. As of mid-2026, an attention score from Adelaide, IAS and DoubleVerify cannot be reliably compared like for like. Picking one vendor and staying with it matters more than picking the "best" one.
Attention is a complement, not a replacement. Both the IAB and MRC state this explicitly in their guidelines. Attention sits alongside CTR, viewability and conversion data rather than replacing any of them.
It adds least where you already have a clear signal. For direct-response campaigns where clicks and conversions are the real measure of success, attention data adds limited predictive value. Its strength is in campaigns where no direct behavioural signal exists — brand video, CTV, upper-funnel display.
How to start without over-committing
A sensible first step is narrow rather than wholesale:
Pick one channel where attention is most likely to matter, usually video or CTV.
Choose one vendor and stay with it long enough to build a baseline. Switching vendors mid-test destroys comparability.
Compare attention data against business outcomes you already track, not against another vendor's scores.
Keep existing delivery metrics running in parallel. The goal is a fuller picture, not a replacement report.
The takeaway
The attention pivot is less dramatic than the phrase suggests. Nobody is throwing out reach and CTR. What is happening is that a measurement layer is being added where the old metrics were always weakest: confirming that an ad was delivered is not the same as confirming it was seen, and confirming it was seen is not the same as confirming it worked.
For advertisers, the practical question is not whether to adopt attention metrics wholesale. It is whether any current campaign is being judged mainly on a number too small or too technical to carry that weight. Where the answer is yes, attention data is worth testing. Where clicks and conversions already tell a clear story, the existing metrics are doing their job.
