Back in July I wrote a piece that highlighted this theme and reflected on the broader view of customer experience that has opened up to our practice since joining EY. I followed that with one of my favorite posts of the year – describing some beautiful work we’ve recently completed that used re-survey and statistical analysis techniques to arrive at real answers to the key questions that often dog non-ecommerce sites. Questions like “What’s the real value of a Website visit by visitor and visit type?”, “What’s the value of increasing satisfaction with the Website or Mobile Application?”, and “What are the correct measures of engagement and how strongly correlated are those measures to actual downstream behaviors?”. For non-ecommerce sites, these are questions that mean everything in terms of understanding the role and value of digital touchpoints, but they are rarely addressed and almost never solved. I expanded on this theme with two subsequent posts focused on the more technical aspects of using Voice of Customer (VoC) research to understand the drivers of customer choice and decision-making. New at EY, Jim Hazen’s guest post was a more personal take on digital analytics but ultimately drove at the same fundamental question – how do you use analytics to drive impact and why don’t more organizations do that effectively. I followed that up with a post on “nudges” – a topic that’s public sector focused – but that has broader implications for all of us. A Website or Mobile Application necessarily embodies a “choice architecture” and while private sector and public sector have different imperatives and constraints when constructing an ideal architecture, in both cases the key is to understand the customer’s needs and decision-making and construct the best possible process around that understanding. Finally, Kelly Wortham’s post (also new at EY) helped build the bridge between analytics and experience with a walk-through of the key steps in creating a robust test design. It should be no surprise that analytics to understand the customer was fundamental to everything she describes.
In this post, I wanted to take a step back and look at the broader topic – how should you integrate measurement and analytics into the process of building a better customer experience – to put all of these separate posts and threads into a consistent framework.
When you deeply integrate analytics into this process, you not only make every step better, you actually re-shape the process. Measurement and analysis re-creates experience design as experience engineering - an iterative process driven by ever deeper cycles of customer research, testing design, and learning.
Here, in a nutshell, is how I think analytics impacts each of these steps and how, ultimately, the process ends up looking.
Assessing the Current Experience
Customer Journey mapping is the traditional starting point of customer experience projects. But this ought to be a data-driven exercise. One of the virtues of techniques like our Use-Case Analysis is that it discovers the real behaviors of customers – including journeys that the design teams never imagined. Use-Case analysis combines sophisticated multi-channel VoC and behavioral data to identify the unique cases that drive usage of key touchpoints and then analyze their actual success. It provides a deep, data-driven look at what customers are trying to accomplish and how successful they are. It’s the right way to begin re-engineering customer experience. But customer journey’s aren’t all there is to assessing the current experience. The type of Customer Intelligence System I’ve been writing about and the use of VoC to delve deep into customer decision-making go beyond journey identification and success to help illuminate real drivers of customer choice and behavior. It’s just not possible to accurately assess the current state unless you know what customers are happy/unhappy about AND why they feel that way.
Predicting the Future
In a time of rapid transformation, every customer experience project has elements of predicting the future. After all, you need to plan for the future state, not just what customers are doing right now. How do you do that? It’s not easy and the right answer isn’t to pay “futurists” or “visionaries” for their fanciful, anecdotal versions of what’s to come. I don’t think that’s business, it’s fantasy. Using data, there is a far more sensible approach to predicting the future. It’s called segmentation. By isolating early adopter populations, you can understand how your broad customer-base is going to look 1 to 3 years into the future. It’s analytics not anecdote - and it works. Obviously, this type of approach isn’t always going to be right. It’s going to miss potentially radical changes in technology – and we all know those do happen. But for realistic planning timeframes, using segmentation as your prediction tool will likely work better than any other method you can find.
Envisioning the Ideal State
This is the step that remains more art than science. But with proper analytics, your art will be a lot more grounded in reality. It will be segmented by audience and use-case. It will be informed by deep knowledge of customer decision-making and attitudes. It will integrate a viable prediction of the likely future state for technology adoption. So while I doubt that analytics can actually describe a true ideal state, I do believe that it is a necessary precursor to getting this function right. In particular, analytics is a cure for one of the great blunders you can make in designing a customer experience – ignoring the importance of the individual. Digital has created massive individuation – everyone wants and demands a highly personalized experience. The analytics methods I’ve described will keep experience engineers from imagining that there’s one “right” experience. There never is. Finding the ideal state is about finding the right places to deliver that level of personalization.
Architecting the Experience
I think the fundamental truth of digital transformation is individuation. The closer you can come to treating each person uniquely, the closer you are to achieving an optimum customer experience. What does it take to achieve optimum customer experience at scale? Data and analytics.
That’s why it’s essential to build measurement deeply into a customer experience. This is also the realm of the nudge. Because it’s not enough to understand what customer’s want. You have to be able to understand how constraints on their knowledge and time may impact their ability to navigate your offerings and your content. Choice architecture is a fundamentally customer driven and the integration of segmentation into choice architecture makes for robust, customer-centric view of how to architect an experience that truly works for the customer.
Mobilize and Launch
The days of boiling the ocean are over. Highly individuated experiences are never created whole hog. They need to grow organically from a process that builds increasingly levels of personalization into an experience. So it’s best to focus on creating an experience factory. What’s an experience factory? It’s a process driven view of customer engineering whose navigation system is your CIS (Customer Intelligence System) and whose engine is the type of experiment design and testing process Kelly described. By combining these elements, it creates a process that can drive continuous incremental improvement in the customer experience; a process that can deliver rapid returns via incremental change, and process that, over time, can fundamentally transform the entire customer experience.
So it’s no surprise that when you build measurement and analytics deeply into every step of customer experience design, you end up with a process that looks more like a spiral than a sequence.
Yes, I know. It’s just the old measurement virtuous cycle in slightly new clothing. But in fact, the new clothing is important. Because it’s the clothing that provides the deeper engine for knowing what to test and how to design and architect those tests. The classic measurement cycle of improvement was largely quiet on how you might actually decide what to test. The underlying assumption was that this decision might emerge somehow from looking at data. The process I’ve outlined here ends up in an experience engineering factory that embodies that cycle of continuous, data-driven, incremental improvement. But it includes an integration of traditional customer experience design and analytics that creates a process for directing the whole cycle.
I think this blending of analytics and experience design creates true experience engineering. Not just measuring what you change to make sure it’s better, but using your measurement to drive what to change. Based on powerful methodologies, data-driven decisioning, and advanced digital analytics, it’s the right approach to creating world-class customer experiences that never stop getting better.
[We're pretty much all set to go for X Change. I'm looking forward to seeing lots of old friends and, hey, if we don't know each other and you're a reader of blog and are coming out, seek me out. Yes, it's an incredibly busy time for me but getting