There’s Never Enough 1st-Party Data, Especially With Contextual Targeting

In Austin Powers: International Man of Mystery, we watch as the titular character braces to run over a guard with a steamroller. But what is quickly revealed is that this is the slowest moving steamroller in history. And that’s the joke. The guard could easily step out of the way and avoid the impending squish. And yet, he doesn’t. This feels like how many publishers have reacted to the impending death of the cookie.

We know it’s coming. Google says it’s going to get rid of the cookie at some point, though that date has changed a number of times. There are proposed solutions to allow for the same sort of targeting without relying on cookies, but nothing that has been seriously adopted. And so, the steamroller keeps coming with publishers hoping and waiting.

I’ve long said that the only true solution to this death of the cookie is for premium publishers to finally prioritize the capture and growth of their own 1st-party data. In other words, information derived from a 1st-party cookie that you drop on your user when they hit your site. And the good news is, every day more publishers step outside of the reach of that impending steamroller. It’s incremental, but it’s happening.

But it presents an interesting question. If everyone is collecting 1st-party data, what differentiates you? At the Digiday Digital Publishing Summit, Blavity’s CRO, Mike Hadgis, effectively asks this question:

I think everyone is racing toward their 1st-party data… I think it’s going to get to a point where there’s just too much 1st-party data. You see in the retail space, like, I forget who just announced it, Dollar General or somebody, “oh yeah, we have our 1st-party data, shopper data now.” I’m like, cool, if I’m an advertiser, which one do I go to? I think that’s going to get to a point where there’s so much 1st-party data.”

I disagree with Hadgis’ notion that there is ever too much 1st-party data. And while he talks about it through the lens of retail media, for publishers, we should never stop striving to get as much as possible. The more that we can know about our readers, the better the position we’ll be in and the more we will be able to target according to our advertiser’s needs.

It does present an interesting challenge when looking at retail media. For the unaware, retail media is the recent phenomena where every platform—Amazon, Uber, Instacart, etc.—have moved into the advertising business, generating unbelievably high margin revenue as part of their broader platform. And since each of these requires a registration, addresses, etc., it gives advertisers the ability to target people at massive scale. If anything is a risk to large scale consumer publishers, it’s retail media.

Back to the idea that there is too much 1st-party data, I don’t think any publisher should be concerned with that. We should be doing everything we can to collect as much of this data as possible. But where we need to really focus is on what differentiates us from these platforms. We are never going to have scale at the same level that they do. However, what we lack in scale, we can more than make up for in context.

Let’s use Uber for example. It can show targeted ads to people based on their location. For example, a movie theater chain could show me the times for an upcoming science fiction movie local to me. That’s a great way to activate that 1st-party data. But what Uber can’t do is understand whether I care about that movie. It might use additional demographic data—I’m a 35 year old man—and assume I might want to see that movie. But is that enough?

Compare that to a science fiction publication. Anyone reading that likely has an interest in movies about it. Therefore, promoting an advertisement about that type of movie to me around contextually-related information is a great way to increase the likelihood that I am going to positively react to that advertisement. And that’s where publishers can really shine.

It’s even more prevalent on the b2b side. With an “own the topic” targeted sponsorship coupled with 1st-party data, we can ensure that the advertiser’s target audience is consuming content that is more aligned with that partner. For example, I would suspect that my biggest advertiser, Omeda, would be happy to sponsor a “1st-party data” category on AMO and then build a marketing campaign targeting audience marketers at publishers.

Now we can sit down with an advertiser and tell them that we can get their message in front of the right person at a time when they are most likely to be engaged with it. If I am looking up movie times, I’m probably not thinking about which CDP to purchase. However, if I am reading content about 1st-party cookies, I am likely more interested in a CDP. I’m the same person in both of those cases, but my frame of mind is different depending on what I’m doing.

To be clear, this is what Dotdash Meredith has done with its D/Cipher product. It believes that it can prove intent to brands based on how users are engaging with content. As I wrote when it first launched:

But if we look at DDM’s site, there are a couple of examples that demonstrate a deeper understood intent. Here’s one. The client is looking to target “Retirement Savers.” The basic contextual targeting are the first two article suggestions:

1. What’s the best retirement savings account for me?
2. Should I invest in mutual funds?

Those make sense. But the third one demonstrates a deeper intent that might not make sense at first glance:

3. How much does Alzheimer care cost?

Of course this would relate to retirement savers. Perhaps someone who is actively saving for retirement has a parent who was diagnosed with Alzheimer’s disease and they are now worried about themselves. Or maybe they are already preparing to retire and just want to cover their bases.

I remain convinced that adding declarative information about the individual would make this even stronger.

Nevertheless, this is an example where the more of this data that a publisher has, the stronger the offering is. And so, while it’s true that every retail media platform is going to offer a similar offering, what we can do to differentiate is connect that to contextual targeting. That blended person to content targeting is an advantage even some of the best retail media platforms can’t offer.

Thanks for reading today’s AMO. If you have thoughts, hit reply or join the AMO Slack to discuss. Have you been rolling out contextual targeting on your sites? Have you introduced an offering that blends both declarative and contextual? I’d love to know.

After this week’s price increase for the AMO Summit, we are almost sold out of tickets with fewer than ten remaining. If you’d like to attend the event on October 26th, you need to register now. This is going to be a sold out event.

Have a great weekend and see you next week!