It’s been eight years since I wrote about the promise of attribution in moving us towards truly omni-channel campaigns. Now that I’m older and, hopefully, wiser, I’ve begun to wonder if I was wrong about one fundamental aspect of the whole thing in the first place: do we really need to be able to assign value to every last touchpoint in the marketing journey?
We’ve overindulged on measurement
As marketers, we have spent years asserting that the virtue of online channels lies in the ability to track and measure everything, so that every penny invested can be attributed back to either a lead or a sale. We’ve cultivated the impression that ‘wastage’ is a sin specific to traditional marketing, then developed increasingly complex and expensive ways to ‘prove’ the value of each channel. And because these models show that customer journeys aren’t linear, we have deluded ourselves into believing that we understand the logic behind how purchase decisions are made, and as a result that we can somehow reverse engineer them to create the perfect marketing plan.
There are just a couple of issues with this. First, we’re assuming that otherwise illogical beings (customers) will make purchases based on some predictable, and therefore logical, manner. Second, and arguably worse, we perform some mental gymnastics to convince ourselves that decisions are somehow incremental and that each touchpoint contributes some fraction of the end result, as if our customer can make 0.94 of a purchase in a binary system. Clearly, this is utter nonsense.
Now that’s not to say that attribution or econometrics models don’t serve a purpose, nor are they even necessarily inaccurate. It’s just that our obsession with measuring the “value” of every input has made us lose sight of broader marketing principles. Maybe I’ve been reading too much Les Binet and Mark Ritson recently, but it seems to me that performance marketing over-prioritises short-term metrics at the expense of long-term growth. Nevertheless, I believe that the upcoming phase-out of third-party cookies* may actually help us to reset and find a happy medium between the “fluffiness” of brand marketers and the hard-nosed “data is king” attitudes of performance marketers.
*I say upcoming because there’s actually some firmer testing dates now, but seeing as this was all originally announced in February 2020 you’d be forgiven for burying your head in the sand once more.
Nourishment over short-term fixes
Plenty of doom-mongering articles will be written in the coming months about the challenges advertisers will face as Google enforces a leaner diet, cookies are phased out and data is lost, but any change presents an opportunity, if you approach it in the right way. This transition is no different: it gives us the chance to realign digital with the wider marketing concepts that underpin attribution in the first place.
I’ll give an example of how this can actually elevate digital marketing. As it’s January and the gyms will be packed with people determined to make changes to their diet and fitness, let’s consider a business which sells protein supplements, and have experimented with blog articles about weightlifting. These articles have driven high numbers of site visits, but no subsequent product sales. Is this traffic valuable? Based on performance marketing metrics, no, this has been a waste of time and resources and the experiment should be stopped.
But this is where the data is potentially misleading. Hidden within that site visits metric, there are hundreds, maybe thousands, of new weightlifters. Chances are, they’re not even all that aware of protein supplements, and it’ll certainly take a while before they even consider buying any, but there will come a time when (assuming the advice in the blog was good) a significant proportion of those visitors start looking into making greater gains. And when it does get to that point, there’s a brand they already know and trust with exactly what they’re looking for.
Now I’m no weightlifter, but I suspect that the time between starting out as a beginner, looking for basic tips and reaching a level where protein supplements become necessary, is longer than most attribution windows, so none of these sales would ever be attributed back to the original blog content. But it’s in interpretation and insight that the real value lies, not in the data itself. True, the eventual purchase decision wasn’t influenced by the blog article any more than it was the result of going to the gym, but the content undoubtedly provided brand awareness and trust which can’t really be measured.
Why this is important
Alternative data collection, such as a greater focus on first-party sources, means that the phasing out of cookies probably won’t fundamentally change how we approach digital marketing, but it should at least trigger some discussion about what pitfalls there are in our addiction to measuring the direct impact of every marketing decision. The New Year’s resolution metaphors are getting particularly strained now but, if you’ll allow me just one more, cutting back on the cookies in 2024 doesn’t have to make digital marketing any less sweet. In fact, I’d say it promises to be a recipe for fuller, longer-term digital marketing which drives both short-term gains and long-term profit.