Attention planning dashboard

The headline attention metric. The number of attentive seconds a placement delivers per 1,000 impressions served. Combines how many people look (% eyes-on) with how long they look (attentive view time). High APM = lots of attention per impression bought. In the calculator, Cinema 60s sits near 46,700 and Twitch Sponsored Streams near 29,000; a mobile Facebook feed video is around 315.

The attention-adjusted version of CPM. Calculated as CPM × 1,000 ÷ APM. It tells you what you actually pay for a thousand seconds of human attention, rather than for a thousand impressions that may or may not be seen. A "cheap" CPM with low attention can have a worse aCPM than a "premium" CPM with high attention — which is the whole argument for attention-led buying.

The traditional buying metric: cost per 1,000 impressions served. Says nothing about whether anyone looked. Attention planning treats CPM as the input and aCPM as the truer measure of value.

The share of a placement's impressions where a person's eyes were actually on the ad at some point. A measure of whether attention happened. OOH digital screens score 100% (you're physically in front of them); a desktop static banner is as low as 16%, and Twitter pre-roll just 4%.

The average number of seconds of attention per view, among impressions that got any attention. A measure of how long. Drives whether a format can clear the deeper objective thresholds (3s, 5s).

The share of a placement's impressions that reach at least 1, 3 or 5 seconds of attention. These map directly onto the objective thresholds: 1s ≈ awareness/familiarity, 3s ≈ favourability, 5s ≈ consideration/purchase intent. A placement can have decent average view time but collapse at the 5s mark if attention is front-loaded.

The act of selectively concentrating on something — however briefly — while ignoring other perceivable things. In media terms, it's whether and how much a person actually engages with an ad, as opposed to merely being served it. It's selective, finite and not binary.

The currency of attention measurement: seconds during which a person is actually paying attention to an ad. Aggregated up, these become APM. The closest thing the industry has to a cross-channel attention currency.

"Lean-in" attention — eyes directly on the ad, conscious engagement. It has the strongest relationship with brand outcomes and memory: more active attention seconds means more recall and slower memory decay. YouTube and TikTok tend to skew toward active attention. Amplified Intelligence / Nelson-Field

The ad is seen or heard, but the person isn't actively engaged — peripheral or "lean-back" exposure. It still contributes to outcomes, but less strongly than active attention, and its effect varies by platform. Facebook and Instagram skew more toward passive attention. Amplified Intelligence / Nelson-Field

The "off switch" — the ad is served but receives no attention at all. Research consistently shows no attention drives little to no brand uplift, and can even coincide with decline. This is the waste attention planning is designed to strip out.

Attention isn't constant within a single exposure. A viewer might attend to the first second or two, then drift; or flick in and out across one ad. This is why average view time and the 1s/3s/5s rates can tell different stories about the same placement.

Six recurring patterns of how attention plays out across an exposure (Long Look, Like Glue, High Flyer, Flicker, Scroller, Super Scroller). Two placements can share the same average attention but have very different shapes — and therefore drive different outcomes. Nelson-Field & Jung, WARC 2022

A critical distinction. Viewability (e.g. MRC standard: 50% of pixels in view for 1+ second) only measures whether an ad had the opportunity to be seen. Attention measures whether it was seen, and for how long. An ad can be 100% viewable and receive zero attention.

The traditional reach measure: the number (or %) of unique people exposed to the campaign at least once, deduplicated across channels. Counts an exposure regardless of whether attention was paid.

Net reach filtered for attention — the unique people who not only saw the ad but actually paid attention to it. Always lower than net reach, but a better predictor of effect. Planning for attentive reach typically doesn't reduce total reach significantly while improving brand outcomes. Dentsu / CCS Planner

A further step beyond attentive reach: people reached with a level of attention deep enough to drive deeper engagement, not just registration. Used to gauge the quality end of the reach curve.

The propensity of a brand to be noticed or thought of in buying situations — being "easy to mind." Active attention drives positive MA uplift; no attention drives none. Higher-attention platforms tend to drive higher MA. Amplified Intelligence; Ehrenberg-Bass

A measure of an ad's immediate sales effect: the uplift in brand choice among those recently exposed versus not. Used alongside MA as an outcome that attention is shown to predict.

Minimum attentive-time levels at which different brand metrics start to move (per Lumen norms): ~100ms for ad recall, ~1s for awareness/familiarity, ~3s for favourability, ~5s for consideration and purchase intent. The basis for the Objective thresholds tab.