From Engagement to Conversion: The Role of Attention in Advertising

Irina Dzyubinsky, CPO at Mediaprobe, discussed the expanding landscape of attention providers, why measuring emotion is so important, what needs to happen to create standardization, and why publishers should embrace attention measurement.

Why is attention important, and, why specifically in 2023, is it something that all advertisers should be thinking about?

That's a great question that we, as an industry, have been trying to answer for a while now. Prior to Mediaprobe, I was with Oracle MOAT. Attention and engagement are topics that have been talked about for at least a decade.

Let's talk about the buy side for advertisers and agencies. Because attention is synonymized with engagement, the advertiser believes that attention is a gateway to engagement. If the user isn't paying attention, they're not going to engage with the content. They're not going to remember the brand. They're not going to remember the product, and therefore it's not going to lead to an action.

Also, there's a lot of advertising waste today. If you look through the lens of attention and you measure everything to [determine] what garnered attention, there is a potential to prevent, to prevent wastage of those impressions.

There are a lot of vendors in the space, both from deterministic as well as probabilistic panel-based or impression-based engagement measurement. Advertisers that have honed in on those opportunities and have been able to move the needle and really focus on not most engaged audiences and content can create ads that drive greater attention.

On the sell side, it's also an advantage if a publisher or broadcaster can prove that their media is more engaging for a certain type of audience, and therefore you, as an advertiser of this particular brand, are better off with that publisher. And yes, they want you to potentially pay a higher premium. But they say you will get better outcomes because our content is just better from an attention engagement perspective.

One thing that is still kind of evolving is attention and emotion. When I pay attention to something when it really resonates, there is an emotional engagement and certain solution providers in this space can measure that.

Right now, attention is being approached from an eye-tracking perspective. But unless you're doing facial coding or facial recognition, where you can gain some insights into the emotionality of the face, it doesn't give you the really meaningful engagement signal of the emotional impact of what the user is watching.

That is one of the things that's still kind of missing in the attention debate. Right? Everybody seems to have resoundingly latched onto the eye-tracking mechanism because it's easier and established. But the cognitive measures, I think, are very, very important as well.

How else do advertisers make sense of attention? What else do they need to know to start incorporating it into their ads and campaigns?  

Well, I think we're a way off standardization. We can't, as an industry, establish a definition. And everything is driven from the lens of viewability and eye tracking, but we have audio publishers like Spotify, Pandora, and Amazon Music, who are advertisers and publishers in their own sense.

So how do we serve the marketplace fully in a meaningful way? We can't have a meaningful conversation about what the standard looks like until we, as an industry, agree that there are other components that are not being considered right now as we talk about attention.

As we expand our knowledge, I think a lot of it is education to understand all the solution providers in the space and what they bring.

The marketers and advertisers want to know: how do we leverage it? And so education is important. But also understanding your use cases. What platform? What's the creative? What's the content? What's the contextual relevance? What is the target audience? Who are you trying to reach the most? And where? And then, ultimately, what goal are you trying to drive? Is it brand awareness? Is it ad recall, or is it the upper funnel? Or are you truly trying to drive conversion sales? How are you going to tie those conversions to attention? All of those matter.

You also want to make sure that the methodology is sound, and you want to make sure that the hypothesis is proven.

And I can tell you that I have predictive modelling to say; I know when channel switching is going to happen based on the emotional curve of the individual based on what we're measuring. But I have to be able to put data behind that and say that that's my hypothesis here that I proved out. So if you want to understand that people are switching in the middle of your content, switching out of your commercials, or whatever when the channel switching happens, this is why you want to use me as a provider.

If you're more interested in understanding the value of a PMP that a DSP has put together and an SSP has put together, and you really need impression-level data on a mobile device, then it's a different vendor for that, right? Like, I'm not going to be able to play in that space. So that's what you, as a marketer. You need to understand how to navigate this landscape. And there's a lot of confusion.   

Who has this sounder methodology? Who has the better use case?

But striking a balance is important. And ultimately, figuring out targeting, testing, and optimization. So that's figuring out again from the use case of what you're trying to solve as a marketer, who is the best provider or providers, you know, a combination of metrics and measures that are going to give you what you need for your toolbox.

It's highly competitive right now, e.g., Adelaide's attention unit is really good for this, here's TVision's framework, and here is what Realeyes is good at. There are too many options. If I were a marketer, I wouldn't know what the hell to do, right? 

Then the bigger question for the marketers is, who pays for attention?

Obviously, the advertisers benefit from both the effectiveness and efficiency attention brings, so they can spend less, create an arbitrage of sorts, capture a larger audience, and achieve better results. So with the same amount of spend, they can possibly drive a lot more impact if they know what's being paid attention to.

Is there a way we create a body, whether it's the IAB with this attention task force or the ARF, who's been in this space and kind of putting methodologies and frameworks together for a while? Or is it another player that comes in and says, "Let me bring it all together." Let us aggregate all of these metrics, all of this information that's out there in a meaningful, predictive, consistent, reliable, scientifically sound way.

We need to give you, advertisers or marketers, a toolkit that says here's how you use and leverage and pull the levers of what you need. Here is what works on these platforms. Here's what works for your creatives. Here's what works on short form and long form, audio, video, and digital, right?

You know, here's your menu. If you need XYZ/ABC, here are the metrics and the vendors and the solutions that provide these answers that you are going to be looking for. That's what will simplify the equation of attention.

What other advice do you have for brand marketers that need to get their feet wet with attention?

If we're talking about a brand, one of the things could be to sort of task their agencies to come up with a recommendation, you know, thinking more of the CMOs or a senior marketer that needs to propel the organization forward and pick provider(s) to start really participating in attention and incorporating it into their marketing strategy. The number one thing for CMO is to educate themselves.

It sounds cliche, but they need to embrace AI and, ultimately, data analytics. And I'm not saying that most advertisers haven't established their analytics and data science units, and agencies have made a lot of investments in this space because you can't improve something that you can't measure, right?

You really need to understand what kind of content is going to grab and retain your audience's attention both from an emotional standpoint and not just from an eye-tracking standpoint. If you're an audio, you know, if you're looking to advertise an audio like, you need to understand it from an emotional impact. You need to understand emotional engagement, which is tangential, which is necessary for attention. Right? What drives engagement is that emotional connection.

Just kind of immerse yourself and say, okay, what's going to be meaningful to us?

If 80% of your ad spend is on TV, then it's very important to understand channel switching to understand distraction, to understand inattention. If most of your advertising is digital or mobile and things like that, then you need to define attention for yourself. There is no one definition of attention. The definition of attention is going to be use-case-dependent.

A lot of these things are going to be panel-based, like, how representative is it? How do you take that and parlay it at scale to build predictive models using impression-level data or higher-scale data?

Attention in one medium is not necessarily going to gain the same attention in another medium. If you're spreading your inventory thin and sprinkling spending everywhere, the cross-platform challenge is going to be big.

There are also privacy concerns and increasing regulation; you have to be very careful about how user information is being leveraged and how you're creating your audiences around it.

There's also contextual relevance in the scope of brand safety. A lot of advertisers don't want to appear around certain content that could damage their reputation and brand.

Additionally, sometimes publishers are really hesitant to transact or to guarantee on attention. They'll say attention is a hundred percent tied to the creative. If the creative is bad or makes the user angry, they think, why should I have to pay for that, right? Why should I be on the hook to guarantee some bad creative?

But it may be that the "bad" creative among that content on that channel on that network, within that program, might be fantastic creative somewhere else with another audience within the on another platform over another program.

And speaking of guarantees: at some point, the publishers were really hesitant to guarantee off of viewability. But they control the screen and what's above the fold. But eventually, we will get there as an industry; we will have to guarantee on something more meaningful than 50% of the pixels for one second.

If the content [preceding an ad] is strong enough and emotionally engaging enough? What we've shown is that the creative [that follows] matters less. You can gain emotional engagement with a potentially crappy ad or crappy creative if the content preceding it was highly engaging. We have data to show that there is a spillover effect.