How New Products Can Help With 1st-Party Data Collection
Ever since Apple announced that it would start erasing 3rd-party cookies with ITP, there has been a slow, but very steady countdown to the inevitable end of cookie-based programmatic advertising.
I certainly won’t cry for it. Why publishers thought giving up control of their revenue was a good idea, I’ll never know. But the outcome is that many media companies, large and small, are starting to accept the reality that they need to do a far better job collecting information about their audience.
It helps for two reasons.
First, if you know who your audience is, you can do a better job serving them the content and products they need. A good example of this is when a b2b media company expands into a complementary vertical; it’s likely they found they had a decent number of people in that second vertical coming to their site.
Second, if you know who your audience is, you can do a better job selling advertising to prospective partners. Going to a partner and saying, “we have the exact demographic or psychographic you’re looking for and here’s the proof” is a really great position to be in.
Naturally, the most important and fundamental way of capturing this data is by acquiring users email addresses. It should come as no surprise to anyone here that I firmly believe newsletters and free registrations are critical to any media company actually owning its relationship with the reader.
The New York Times is really quite good at this. If you go visit the site as an unregistered user, you can view a single article. After that, you are prompted to create an account. As a quick aside, this plays right into their whole subscription business. Piano, the registration system, has found that any sort of change from anonymous to known user results in a 10x boost to paid conversions.
But back to capturing data…
I find many publishers understand how to build a registration system. There is plenty of software out there to support it. They even capture a little bit of data. I would argue that this is just the beginning and there are creative ways in which publishers can collect additional data without the user realizing that’s what you’re after.
I suggest accomplishing this by thinking about your website as more than just content. What other features and products could exist that the user might engage with that would give you additional information?
This also results in a fundamental change to how your site is perceived by people as well. With most people visiting our sites through the side door, creating features that get users hooked is critical to our sites becoming destinations. Therefore, think about how you can become part of the user’s daily or weekly habit.
A basic idea is to require additional data if a user wants to leave comments. The Information takes this even farther by requiring users apply to comment. We need a completed profile, which basically tells The Information everything about our professional lives.
I think there are more advanced implementations that could be very interesting and very niche-specific. Here are some ideas that I had across a few different niches. You know your audience best, though, so what can you identify?
I only own a single pair of sneakers—and I’d be a liar if I said that they were fashionable—but there is no denying that sneaker culture is a very real thing. People wait in lines to spend hundreds of dollars and there is even a secondary market where you can spend thousands to buy a single pair.
One potential idea could be creating a virtual shoe rack. You would ask each user to tell you what sneakers they own. On the backend, you would then create a taxonomy of sneaker types.
As time goes on, you would begin to understand what types of sneakers people are fans of. Perhaps no one likes Pumas, but everyone is dying for the next Air Jordan (that’s about all I know).
To take it one step farther, you could use this same product as a sort of wish list. Have users select their fantasy shoe rack.
In this case, you now understand the type of shoe collector the person is. If their shoe rack only has a couple pair that are all in the price point of <$100, you know they’re not going to chase after limited edition, expensive sneakers. On the other hand, if their shoe rack has thousands of dollars worth of sneakers, that’s an audience advertisers likely want to target.
I’ve spent the past few years of my career working in new financial media—crypto—and it can be an incredibly lucrative industry. Because we’re talking about people investing money, advertisers are willing to spend large amounts of money to either get you interested in their investment product or to get you onto their platform.
I remember as a kid, my grandparents would login to AOL to check their stock quotes—it was the tech bubble, after all. As I started working in media, I realized how brain dead simple the concept of a portfolio tracker was.
Users are literally telling us the exact companies they want information on. You can then start delivering them a personalized experience around that content. Looking back on the sneaker idea, if you find that there are a lot of people looking at a specific category of asset, it might inform additional investment in content creation.
The final idea I had was inspired by my kitchen cabinet. My girlfriend has amassed an incredible collection of Le Creuset cookware. Why we would need multiple cast iron pans and, I think, three braisers is beyond me. But hey! That’s her collection.
Like the previous two ideas, imagine if a recipe or cooking site gave users the ability to create a kitchen cabinet. With one click, users could share this with their friends. It might also inspire other people to add similar pieces to their wish list.
Additionally, these food sites could take this data and with a few additional questions—specifically around frequency of use—start to understand what types of cooks/bakers they have. This could inform content creation or advertising partners.
I’ll be honest… As someone who doesn’t find joy in cooking, this might not fit. But you get the idea.
Alright, so what?
None of these ideas are terribly innovative, but the purpose is to think about how you can start collecting data about your users without using something basic like a form. People are more likely to give you data if it’s part of the product experience.
Back in June, I wrote that the ad business is going to become far more contextual:
The Hustle and Morning Brew are not selling ads based on consumption and personal data. Both sell ads to brands that are looking to target a specific psychographic audience—it’s not age, gender and location, but rather, psychological factors. What about Industry Dive? It has no programmatic on its network of sites and, before COVID, my guess is it was budgeting over $30 million in revenue this year. It’s all contextual advertising. Advertisers know that if they place ads on specific Industry Dive websites, they’ll target the right audience.
A big reason that Facebook and Google ultimately won the battle for much of the digital advertising business is because they had better data on their users. According to a Washington Post story from 2016, there were 96 different data points that Facebook collected on its users to help with ad targeting. In what way can a publisher expect to compete with that?
I don’t expect us to be able to compete with that level of granularity. But by mixing contextual targeting—what is your site about—with more 1st-party data—what does your user care about—you can start to envision a future where you are able to start offering more targeted campaigns to advertisers.
Why? Because results matter and better targeting can give better results.
As Covid-19 was really hitting the economy back in April, I said:
For B2B media, direct marketing has always been the process. Brand marketing isn’t as big for these companies, so any campaign typically results in leads that the advertiser’s sales people can target.
For consumer brands that don’t deal in “leads” per say, the conversation has to change from impression-based advertising to results-based advertising. If you’ve got a DTC brand that is willing to advertise on a podcast, offer them a no-risk digital campaign on site. It’s not a guarantee like with an insertion order or programmatic (can we even say these are guarantees anymore?), but it’s also tied directly to a response from your audience.
It’s clear that there is ROI with Facebook and Google, which is why no one will ever really stop advertising with them. Can most publishers confidently say there is ROI with their sites?
If we can then take this product-level user data and relate it to the revenue server I wrote about a few months ago, you can really start to get smarter with the types of revenue opportunities on your site.
It’s all part of a balanced breakfast
Everything starts with the email address. By having that data, you can communicate with the user. But over time, you need to start augmenting that with additional data about the user.
For business media companies, that has always been done through longer lead capture forms, webinars, white papers and things of that nature.
But for consumer companies, there has been a very lackadaisical approach to understanding their audience. There has been an inherent laziness tied to relying on programmatic advertising for all revenue. Suffice it to say, going forward, the successful consumer media companies will find opportunities to collect data about their users and understand them better.
To sum up, these sorts of creative data collection products should be treated as part of the process. But also, don’t be afraid to deploy progressive data capture that clearly asks users to provide more about themselves.