The John Wayne view on marketing attribution
'You’ll never know where you’re going until you know where you’ve been.'
– John Wayne
Within the marketing technology community, we find ourselves speaking on the topic of marketing attribution a lot. It’s constantly on the front burner of every modern CMO – how do I do more with less? How do I demonstrate ROI from my various channels, tactics, campaigns – and then boost it? How do I not only defend my budget, but make a data-driven case for more?
Make my day, pilgrim
It’s not just market competitiveness that’s driving this pressure. It’s becoming a cliché to point out the ever-increasing crush of customer behavioral data today – from digital properties, mobile devices and apps, social media, third-party sources, offline transactions, and so on.
Marketers have more data than ever, and technology vendors are scrambling to put together the right combinations of tools to help them wade through it. But when you strip away the hype, the challenge for all of us is actually pretty obvious: how do we collect and analyze this data effectively, and then use it to produce insights that improve the business?
Marketing attribution – the science of determining what marketing is driving what results – is, in my opinion, at the heart of this very broad goal.
The Art of the Possible
The problem has long been that attribution technologies just weren’t that robust, so using them was frustrating and rife with limitations. For example, many technologies in wide use today still can’t link basic cross-session click and cookie data to offline customer information. This is a big problem if your customers don’t just live on the internet.
Indeed, what we’re seeing in the mobile commerce space is that customers do a significant amount of product research, search for stores in their area, and look for promotions on their mobile devices before converting offline. They do these things across many sessions, and through multiple channels.
Yet the interest in this topic is red-hot, because most marketers still lack the tools to actually measure that, and figure out how they can optimize their marketing posture to fit the contour of their customers’ conversion lifecycle.
There is emerging consensus today that this is as much a people issue as a technological one. Industry observers and practitioners like Tim Wilson, Michele Kiss and Eric Peterson have written extensively on how marketing organizations themselves must realign around metric-driven decisioning, rather than relying on the old HIPPO model (“HIghest-Paid Person’s Opinion”).
The tools to measure, test and decide exist today – and insofar as we don’t take advantage of them to overcome organizational barriers and base marketing investment decisions on data, rather than anecdotes or hunches, we do ourselves and customers a disservice.
But the technological side is an important one, because it defines the horizon of what’s possible to measure digitally. Several interesting pure-play vendors have developed tools to address full-funnel measurement, a concept in which we automatically measure the contribution of every marketing touch at each stage of the customer conversion lifecycle. This remains a challenge for most marketers with many existing tools.
Another challenge is interaction histories – collecting and using cross-channel repositories of all campaign contacts and responses from each and every prospect and customer.
This history of interactions can provide the basis for algorithmic modeling to output the most reliable, data-driven attribution results possible to guide future marketing spend.
With robust statistical models automatically applied to the crush of customer interaction data in real-time, we can quickly identify what’s driving awareness, persuasion or conversions, and make changes as necessary. In many ways, this is the holy grail of attribution – and, indeed, of digital measurement itself.
I did a little bit of research, and it turns out that the John Wayne quote above is fictional. It’s a widely misattributed quote that has come to achieve the status of conventional wisdom that few bother to question with data. How much of your marketing is the same?