Blog post

Customer Journey Analytics: A Manifesto

By Martin Kihn | March 04, 2016 | 3 Comments


It’s time for a new kind of analytics. It’s time for a fourth dimension. Time to recognize the reality of cause and effect, of unpredictable decisions and uncertainty and motion in marketing. It’s time for a new approach to measuring and improving the way we talk to our people.

It’s time for real customer journey analytics.

Before we go there, reflect where we’ve been. Statistics are always more or less an approximation of reality — sometimes more, often less. Marketing statistics can drive this approximation right up to the border of wishful thinking. The problem is not a lack of analytical ability but rather a lack of information. An inherently unstable sample. A partial picture of an irrational target.

People are hard to influence, confounding to predict, impossible to know. They stop opening your emails and don’t say why. Their lifetime value is rock solid until it’s all wrong. They clear their browser cache and you lose sight of them. And it’s difficult to know just why they don’t drop by more often, or ever, what ad they saw or remembered, what they really think of you, your products, your brand.

Most likely, they don’t think anything. In addition to being conceptually difficult, marketing analytics has the indignity of getting applied to a field that isn’t all that important, really. We’re not finding quarks here or fighting crime. But our task can be just as imposing.

If you’re bored as a marketing analyst then you’re just not awake.

Why a Manifesto?

It’s time to assemble an analytical system that recognizes the reality of digital life. One that combines the best of customer identity resolution, channel and media measurement, static and time-series methods, and text analytics — time to create a single version of the truth at the individual level, across marketing and media, to estimate the impact of our efforts on the things we are trying to do.

It’s time to apply the math and scale of media attribution to things beyond advertising: to site and app engagements, email, social and distributed content, call centers and search.

Today’s marketing analyst uses too many tools. We have to: it’s just the way that it is. These can be divided into five basic types:

  1. Web and Mobile Analytics — measuring engagement with sites and mobile apps (which themselves often need a different tool)
  2. Social Analytics — applying text analytics, counting and graph methods to posts
  3. Media Analytics — attribution and marketing mix modeling (either separately or combined) for measuring the impact of paid media
  4. Customer Journey Analytics — using CRM and marketing system data to track known individuals over time
  5. Voice of the Customer Analytics — mining customers’ perceptions and opinions using surveys, text analytics and feedback cues

These systems are all looking at bits and pieces of what to a person — whether called a target or audience or prospect or customer, it’s still a person, people — is a single relationship.

It Gets Worse

Very often, web and mobile analytics is performed by a digital marketing department or agency, social analytics by an intern, media analytics by a high-end platform or consultancy, voice of the customer by the support team, and customer journey analytics by no one or an intrepid CRM or I.T. numerati.

Most likely, these people don’t even recognize one another at the same holiday party.

It’s time to stop the madness, combine the tools, starting with the customer record. We’re getting there: already, web and mobile analytics are coming together, social and voice of the customer. There is a heroic effort under way to build a master customer record using onboarding platforms, cross-device math, APIs and DMPs and integration systems.

And then there is the rise of customer journey analytics itself, which promises to do for marketing what attribution does for media.

Customer Journey Analytics: What is this? It’s actually an interesting and relatively early-stage set of methods and tools that grew out of CRM. Leaders in the space include Thunderhead (“It’s Not Your Journey, It’s Theirs”) and Clickfox, names little known in digital marketing departments. And there’s some promising start-up activity (see Pointillist). These tools build a customer profile based on known attributes — value, loyalty, product preferences, locations — and layer on how they behave over time in digital channels or stores. Any data scientist will see the value of such information, gathered from enough individuals, to marketings’ efforts to improve offers, messages, timing and sales.

It strikes me that customer journey analytics of this type is the future of a combined marketing and media analytics solution. It’s our money shot. Why? Because it incorporates the element of time, the fourth dimension, and it can be adapted to respond rapidly based on triggers. The freshest information is the best, no matter what anyone tells you.

McKinsey has talked about the customer journey. So have we (clients see here). McKinsey has a formal partnership with Clickfox. Our ambition as analysts is not to say that “the funnel is broken” or “customers jump around” — everybody knows that — no, our aim is to encourage dreamers to believe there is a market for coherent analysis. Systems that capture customer-level detail and use time-series analysis. Systems that take the incredible advances made by ad tech players, in infrastructure and stream processing and time-series analysis — that take these advances and apply them directly to the enterprises’ own marketing technologies.

I’m not saying we’re not trying. Just that we’re not trying hard enough. We need a Moon Shot, an Omega Project, a Zero Moment of Truth Machine for Marketing. We need to start thinking about time.

The Gartner Blog Network provides an opportunity for Gartner analysts to test ideas and move research forward. Because the content posted by Gartner analysts on this site does not undergo our standard editorial review, all comments or opinions expressed hereunder are those of the individual contributors and do not represent the views of Gartner, Inc. or its management.

Comments are closed


  • Leroy Marvin says:

    Interesting point of view – really well said. As a marketing analyst I am frustrated by the swivel chair reality if my life. I agree media analytics seems more advanced but is also more complex and expensive. The systems are not there yet in marketing – but I agree we are trying. Part of the problem is internal resources and even politics. But thanks for the shot in the arm.

  • At Thunderhead we applaud this fantastic article, thank you Martin. As leaders in the customer journey analytics space it probably comes as no surprise that we’d take this “Moon Shot” a step further to explore the true value marketers can derive from Customer Journey Analytics. We recommend a fifth dimension as marketers need to be able to see the end-to-end multiple journey view, all of the roads not taken as well as the ones which are, the holistic view as well as that of individual journeys, and you need to see these in real-time to be able to take the appropriate, relevant action. That’s when customer journey analytics gets really, really interesting. Read our blog to learn more about a 5th dimension to customer journey analytics and why marketers need to be Time Lords of customer journeys.

  • Tim Friebel says:

    Great piece, Martin. And not just because you mention us in it. 🙂 This new thinking is long overdue.
    I love your comment, that the customer journey “incorporates the element of time, the fourth dimension, and it can be adapted to respond rapidly based on triggers.” We couldn’t agree more and think the time component has been missing in analytics for years. Journeys can often take weeks or months, consisting of many events that occur along the way. Journeys should also be classified by the business to establish a common language across the enterprise. (e.g. How is the purchasing journey defined by one organization vs. another – it’s like you say, “these people don’t even recognize one another at the same holiday party”)

    There are many analytics tools out there, but you need the right data to get the most out of them. We are now making our connected, contextualized journey dataset available for downstream consumption by any tool or process. Clients and partners have said, “we like your analysis capabilities, but can we just get our hands on that connected data?” We’ve listened.

    In our humble opinion, many are viewing customer journey analytics is a new label for old techniques. We disagree with this. A recent blog post of mine goes into more detail on our unique approach to journeys. I’d love to hear your thoughts on it: