people walking under arches of a building
By bringing together four different forms of analytics, brands can gain a clearer picture of their customer's behavior and needs PHOTO: Jānis Skribāns

I don't believe only one type of analytics method or platform can speak "The Truth" about the multi-channel customer experience of your brand.

Yet this belief can be found at every level of the analytics world: Software vendors that promote their solution as having all the answers. Digital marketers who only look at metrics reporting from their agencies. Analytics program managers who believe that enormous data sets are the only way to understand the customer lifecycle. Channel managers who only see customer experience through the lens of measurement of their tactics.

Instead, I propose a method for combining analytics methods and tools to help improve digital customer experience I like to call Big Tent Analytics, or more formally, an Integrated Analytics Approach.

Different analytics tools perform different functions — and that's OK. But each adds to our understanding of how customers respond to the digital experiences we present, and it's important to appreciate how.

Know the Problem Before Looking for the Answer

You probably know this, but it bears repeating: figure out your questions before you start to measure. What business problem are you trying to solve with data, for example, customers leave your site before they complete whatever task you wanted them to complete (e.g., a purchase, a lead gen form, an account transaction).

Let’s say you’re a product manager for an outdoor recreation company that sells inflatable kayaks. You have a budget to perform content and design changes. How do you use analytics to inform your plans?

Start by asking these questions:

  • What are customers doing?
  • Why are they doing it?
  • How are they doing it?
  • What are the brand risks for not improving their experience in using the site?

We can answer these questions using the Integrated Analytics Approach to match the right technique to the right question.

Breaking Down the Integrated Analytics Approach

All of the questions we asked above can be answered through four data sets: clickstream, voice of customer, user experience and social media.

an integrated analytics approach

You may think you’re ready to start diving into all of this data. Many organizations are so excited about having the data to explore that they don’t develop a strategy to plan and prioritize all of the steps it takes to go from data to insights. Don’t be one of those organizations.

Create an analytics plan that describes how data will be used, who owns the data, how data can be shared and used for answering questions and addressing specific issues that are critical for your business objectives. This plan should address data collection, processing, analysis, visualization and also integration with other data sets that are sourced by other systems: email, sales, products, customer, fulfillment, subscription, partners and so forth.

If you’d like to learn more about creating an analytics plan, check out a few of my previous articles: "Bringing Big Data Analytics into Focus for Marketers: 3 Principles to Simplify Your Life" and "Blue Collar Big Data: The Work Behind Big Data Analysis." And if you want to learn how to think of adoption of insights as part of your strategy, "Understanding and Meeting the Challenges of Digital Transformation" will give you a good start.

How to Use Clickstream Data

As the product manager of inflatable kayaks, you'll want to know what actions customers and prospects on your website are taking.

Clickstream analysis from platforms like Google Analytics and Adobe Analytics tells you “what” is happening on your site. This data provides a view into behaviors you want to encourage and points out challenges visitors may be encountering on your site.

Negative behavior, such as exits from pages, fields, internal search, signals possible areas for improvement of content, functionality, navigation and overall user experience.

Positive behavior, such as content shares, successful form submission, completion of transactions, multiple cart adds and check out, indicate what’s working. You can go a step further and determine which segments exhibit this behavior more frequently through segmentation analysis that shows attributes such as geography and marketing channel, campaigns and tactics, and first time versus repeat visits.

As a result of your review of clickstream reports, you, as project manager, find there’s a high number of exits after new prospects use the interactive tool that allows them to compare different kayaks. Which is good information, but it doesn’t tell you why people are leaving. Let’s ask them.

How to Use Voice of Customer Data

Targeted surveys from tools like Qualaroo, Informizely, Knowclick, Hotjar and other platforms help you get to the “why” behind the “what” as long as you use these tools to specifically troubleshoot — not as general customer satisfaction surveys. Ask specific questions about issues that cause visitors to drop or not continue. Ask visitors what they were looking for and whether or not they found it, and if it was as quickly as they expected. Use targeted and short surveys on the page with the problem. If the issues are site wide, ask at exit.

In your product manager role, as a result of doing short targeted surveys on the product comparison pages, you find many visitors left because they didn’t find comments from other buyers.

Now we’re getting closer to figuring out how to address this customer experience issue in the site design. Let’s see how this has affected the user experience on the page itself.

How to Use User Experience Data

For the purposes of this analytics discussion, I’m defining user experience data as the type that’s derived from heat mapping tools like ClickTale, Decibel Insights, Crazy Egg and Hotjar, among others. There are other ways to capture user experience data, such as TryMyUI, a platform that signs up “testers” that simulates traditional user experience scenario testing. In my opinion, whatever technique you want to use, it’s important to understand how visitors are actually interacting with the pages and functionality to understand the human element.

With this data, you'll find out where visitors are following or not following content, navigation or getting confused. You’ll see what elements your visitors aren't using that you want them to use and you'll also understand scroll behavior and link click intent. Maybe the user experience data will show your visitors scrolling to the bottom of the comparison pages, and clicking the mouse. This might imply they're expecting more information, such as comments about the kayak they're viewing.  

Clickstream, voice of customer and user experience data all contribute to seeing the what, why and how of digital experience. Combined, these are enough for you to create options for how to improve the part of the customer experience that is driving visitors to leave the site.

But another analytics method can help you make the case for excellent customer experience: social media analysis.

How to Use Sentiment Analysis

Finding influencers, understanding customer sentiment and getting a sense for how customers perceive your brand is the way you should be analyzing both owned and earned social media. If you need a business case for improving customer experience, receiving negative feedback that impugns your brand is a powerful, though unfortunate, motivator.

Platforms like Sysomos, Crimson Hexagon, Salesforce Social Studio, Hootsuite and others are powerful tools. They do require configuration to provide accurate sentiment analysis. Like clickstream analytics, it helps to know what you’re looking for and the questions you want to answer before you configure the data collection.   What often wins the day are the quotes and content that come from the posts themselves and ultimately humanize the need to optimize the digital customer experience.

In your product manager role, if you saw comments on a kayak enthusiast blog that said it’s difficult to get the feedback they need from your site to buy online, you’d want to know this, right?

Plan for Analytics Success

Clickstream, voice of customer, user experience and sentiment analysis all have their roles in understanding how to improve digital customer experience and ultimately help you achieve your multi-channel goals. Don’t get hung up on one method over the other. Learn how to use them for what each does best and work with your partners and teams to develop processes and governance so that all perspectives are included and leveraged to their maximum value.