Adtech is in love with the wrong type of data - here's where its heart should truly lie

Hannah O’Neill (VP sales, Nano Interactive), looks at adtech's questionable love affair with profiling based on historic data. With a divorce imminent in the shape of cookie phase-out, where should advertising be looking next?

Adtech is facing once in a lifetime change in the shape of signal loss. But it’s not as if we haven’t known this for a while. So why is it we’re only acting now, when it was fully four years ago – back in January 2020 – that Google first announced its intention to phase out the cookie?

I have a theory why. We’ve been focusing on the wrong type of data for years. We often hear about advertising as a fast-paced industry. But have we been in such a rush that we failed to notice a tactic, adopted far and wide, was deeply flawed from the start? 

In a vacuum, data dies 

Profiling has become overwhelmingly the accepted approach in online advertising, facilitated by third-party cookies, which allows platforms to effectively track people across their digital lives. But which also ignores a very real fact. That a signal of intent in the shape of a site visit, search or abandoned purchase takes on a completely different meaning if it happened 30 days ago rather than in the last five minutes. And let’s be honest: until now, for whatever reason, few questioned such an oversight. 

Add to the mix the fact that around 70% of consumers are taking steps to limit tracking off their own backs - on a weekly basis or more. Not forgetting that 40% of the browser market has already been off-limits for quite some time, with Safari shutting the door to cookies years ago already. Has an already flawed approach, been further undermined by the increasingly patchy data behind it? 

Next, the rise of retargeting seemed to go hand in hand with profiling. Shorn of context and timing, it has become a source of frustration for consumers. And according to Nano’s latest consumer research, 49% of consumers who mask their data from trackers do so to avoid just this practice. Cookie deprecation has arguably made it even more scattergun – where IP addresses are used as a backup, the chance of a creative revealing a secret gift to a family member on a separate device only increases.  

29% said they mask to avoid just this eventuality. And meanwhile, several reports emerged suggesting that IP is likely next in the crosshairs of tech gatekeepers: in December, Digiday went so far as to state that all companies “really should reconsider their use of the IP address for tracking audiences.” 

Why advertising should look to intent data to reach consumers 

Instead of basing campaign choices around past actions, intent marketing – including live signals such as content, context and attention - allows advertisers to avoid the limitations of already-limited or dwindling people-based signals. They also allow for buyers to reach that scale in a live, and still relevant way – to groups, beyond any assumption or former event.  

People-based data has always been connected and applied to profiles, without considering that we fulfil all types of different activities and roles, depending on the time of day, device and most importantly, what task or pastime we are pursuing. 

Media and advertising has gone full circle, from proxies for audience - magazine vertical, TV rating points - to directly targeting profile groups via cookies – even if some doubted their accuracy. Back to the present, and signal loss, from link decoration, to IP to SDKs is in the news each week. And with it, a shift back to proxies for intent also in full flow. Whether it is attention, sentiment, content or context, those signals are now live in a way neither print nor cookie or other IDs ever were. 

Most of all, profiling places no value on environment, which has arguably also lead to the devaluation of quality, original journalism, and the rise of the ‘made for advertising’ phenomenon. 

'Absolutes’ vs. ‘today’s context’ 

Several years ago, at least one commentator was already on top of the distinction we’re now seeing play out in front of us – for profiling: “You understand today or try to predict tomorrow by pulling from things you assume to be absolutes, all of which were learned yesterday,” says Jay Acunzo.

Whereas an approach more similar to intent, Acunzo says, you “isolate a variable and test it specifically. This recognizes that generalities are dangerous and that context is everything. So you better test and learn in today’s context.”

This couldn’t apply more perfectly to where we find ourselves right now.

“We are driven forward by that data, but that data comes from the past. And more than ever before, today looks way different than yesterday, and tomorrow will be nothing like the present,” says Acunzo.

With signal loss, the fact that today will no longer be the same in yesterday is an immovable fact. And we will unavoidably have to change the way we work.  

We often hear of advertising as a fast-paced industry. One so proudly high speed, it’s almost impossible to keep up. But in the case of cookies – and broader signal loss – we've known the need for change for years, without changing our approach.  

At this crossroads, a flaw in our approach right from the start can still be corrected.  

Time to shift from past to present, from void back into context. 

By Hannah O’Neill, VP sales, Nano Interactive