Some advertisers are buying too much attention. Surprising take-out from new Kantar analysis on how attention builds brands. The study based on over $3bn of media spend across 873 campaigns contrasted brand impact with the latest attention data from Lumen. Biggest finding: Attention doesn't correlate with cost effective brand building. How could this be? The analysis highlights 5 reasons: 1. Some brands need a lot of attention for ads to be effective, but others really don't. 2. Not every second of attention costs the same, no point having 20% more attention if it costs 50% more. 3. A second of attention doesn't always deliver the same return, some formats are more attention efficient than others. 4. Even in a single campaign a mix of attention works better. Just because you have a few longer hero assets doesn't mean your entire campaign needs to be high attention. 5. Attention isn't created equally, there is a difference between passive (eyes on screen) and active attention (feeling something) and that difference isn't just about time. Kantar point out that neuroscience supports the findings. Daniel Kahneman found how long you experience something doesn't necessarily mean stronger memories. Obvious when you think about it. Seeing someone else kiss your partner for a millisecond would create a very strong memory, it certainly wouldn't take 2.5 seconds to have an impact. Kudos to Duncan Southgate and the team at Kantar for moving the attention discussion on and bringing some much-needed nuance. Aligns with what some in the industry have been saying for a while. At the Meta Brand Summit last year Professor Byron Sharp said “Be humble. We’ve always had fleeting exposures. We’re always going to have fleeting exposures.” “I would not optimize only for long view duration, it's not a good objective and it's terribly expensive.”
Thanks for sharing Pete. Although everyone loves things in 3s and 5s, I think a key 6th point is missing. Highest levels of attention come from those already familiar with the brand/ existing customers, meaning cost effectiveness is diminished as budget is optimised towards those non incremental groups. Its the same issue as the myopic objective of lowest CPA through digital performance channels, whereby the algorithms optimise to the path of least resistance (existing customers and those already likely to convert without ad spend) and then over claim the resultant value generated. Cost efficiency looks great but effectiveness and incrementality die.
Just a working example was the work that Heineken did here: https://www.exchangewire.com/blog/2024/05/30/heineken-playground-xyz-partner-to-find-optimal-attention-drive-real-world-outcomes/ They found that for their Cruz Campo ad, they required relatively low levels of attention to drive awareness lift. Their campaign was achieving a much higher average attention than their optimal threshold. This gave them the opportunity to look at where they were getting surplus attention, what the cost of that was and work out what adjustments could be made.
We also find that ads mostly do a good job at getting attention. Creatives most know how to get it. There are other bigger challenges.
I believe the most important point is missing, or maybe it was not part of the research? That point is "what is the objective of the campaign?" There are different studies in several countries (not just Lumen in the UK) that look at the difference between, for example, reminding/reinforcing a brand vs introducing a new product, and the type of attention (=fixation) needed for different obectives.
This is by far the biggest misconception about using an attention measurement methodology that is able to understand duration. Just because advertisers can identify highly attentive inventory, doesn’t mean that you should always buy that inventory that garners the MOST attention possible. Highly attentive inventory costs quite a bit, and there’s always a point of diminishing returns. Alternatively there’s incredibly cheap CPM’s that offer quite low levels of attention. Ultimately it’s about finding the point in the middle that is the right duration for your brand to achieve outcomes. That’s what we call optimal attention and the entire goal of leveraging biometric powered attention methodologies in the first place. Ads need to be seen in order to be effective, especially interruptive units. Marketers HAVE to be able to understand how much duration their formats garner in order to give their CREATIVE the best shot at breaking through.
Duration is a very 1.0 way of thinking about attention, especially when it comes to media quality. I might be oversimplifying this but I think it's media's job to create an opportunity for attention, ward off distractions and eventually contribute to an outcome - so better to look at things like probability of attention, distraction and outcomes.
"You have most to gain by scaling attention insight focused on the intersection of creative and media." ✅ Whenever I see studies like this it does seem obvious that media, audience and creative need to work together to create a bespoke-to-the-brand/product amount of attention at a certain cost-threshold to drive value (i.e the cost of attention is less than the margin created by the measured output). Many agencies and brands can tie themselves up in knots (and sink insane amounts into research related costs) to understand this. For most, the best way is to get started with a minimum viable amount of attention from media, and test audience and creative on top of that. That's what we're speaking to smart buyers about at Impact Media.
Attention and reach are blunt metrics - all ‘what’ and ‘when’, no ‘how’ or ‘why’. Maybe… just maybe… all attention isn’t equal. Maybe it’s less effective if your brand isn’t meaningful or capable of building memory structures But I guess when you flog ‘attention’ as a commodity that matters more than understanding how you turn it into something influential and effective.
Connection Planning Director, Meta
1moNew Kantar analysis here: https://tinyurl.com/4h67st4d