Measuring Customer Experience
Published: June 17, 2013
In the current era of business, the customer is more important than ever before. Check out analysts’ recent work on the financial performance of companies that get it versus those that don’t. It’s eye opening. Brands like Amazon who innovate their business to provide the highest-quality, easy, personalized interactions to their customers are taking their markets by storm.
Since customer experience (CX) is so important, shouldn’t we all want to know how our digital products, services, and interactions compare to those of our competitors? Are they sparkling examples of interactive delight that rival those of the CX champions or more like the punch-in-the-face customers get when they deal with health-plan providers?
Comparing Different Customer Experiences
By taking complex, subjective, and sometimes emotional experiences and measuring them using a single yardstick—that is, the same grading scale—we can compare them. While measuring a rich customer experience using a single numeric rating certainly cannot capture the entire rich story of a customer experience, it does allow you to compare the net amount of positivity that your customer experience engenders in customers. It can also provide a way to track the progress of your UX design effort at an executive level. Did a design problem that you fixed last month materially move the CX needle, or did it turn out to be a fairly meaningless detail? Without measuring your customer experience, it is difficult to understand where the truth lies and whether your brand will benefit as a result.
Can You Really Measure Customer Experience?
Customer experiences are complicated, subjective, and often emotional, so there is no doubt that some information gets lost if you try to collect, organize, and report information about them systematically. Certain measurement approaches have attempted to get around this problem by breaking down customer experiences into what they consider to be multiple unique attributes. However, the by-product of using techniques that decompose customer experiences into their attributes is that, as measurement becomes more sophisticated and capable of providing better and more actionable data to a company, measuring those experiences also becomes more disruptive to customers.
Remember my point about the customer being more important than ever? I don’t know of any customers who would want to complete a lengthy multipage survey during an important task—especially if the delivery method was so disruptive that it would interfere with their ability to complete their goals.
To minimize this sort of disruption, many CX professionals have adopted the use of surveys that contain only a single question and randomly get delivered to a small, representative subset of customers. This approach is relatively painless and achieves higher completion rates by real customers because of the extremely short length of the survey. Thus, the responses are often more representative of your true set of customers, so more generalizable, because the customers who receive the survey actually complete it.
Because of the simplicity and fast completion times for these surveys, you can easily administer them when a customer is either in the middle of a task or has just completed a task, or when you’re doing a general relationship check-in that you periodically deliver through email. While you won’t capture everything about your customer experience, you will have an objective yardstick that you can use to measure the amount of positivity that your customer experience engenders. You can then begin experimenting with different ways to improve things.
Can You Measure Customer Experiences Without Using Web Surveys?
Beyond Web surveys, there are several other methods of measuring customer experience, including emotion detection through social media–monitoring tools. Alas, social mentions are not representative of all of your customers’ viewpoints—and even if they were, the current state of computer-based emotion detection is still in its technological infancy, so only rough measurements are possible.
Other approaches include using social forums or idea-management services such as UserVoice or GetSatisfaction, but these tools don’t allow quantitative benchmarking or comparisons. Purists who reject quantitative methods altogether may advocate for proven qualitative methods such as direct observation, usability testing, and ethnography. While these methods are certainly valuable, none of them achieves scalability or allows for truly direct comparisons across very different contexts. For these reasons, I’ve chosen to omit these approaches from further discussion in the remainder of this article.
Measuring a Single Touchpoint Versus an Entire Customer Journey
Using a dipstick to measure your car’s oil level helps you to avoid engine troubles, but what about all of the other systems under the hood? To prevent the need for costly car repairs, you also need to check your car’s brake fluid, coolant, and power-steering fluid levels regularly and so on. Similarly, while it may seem obvious, measuring the customer experience of just your Web site tells you only about that single touchpoint. In an omnichannel, or multichannel, world in which your customers may interact with you through Web chat, phone calls with live agents, Interactive Voice Response (IVR) systems, SMS messages, Twitter, Facebook, mobile applications, and in many cases, in-person or in-product interactions, you need to measure customer experience on all touchpoints if you want a holistic view.
In addition to omnichannel CX measurement, you might choose to measure customer experience by lifecycle touchpoints such as awareness, research, purchase, in-product, and customer care; or by data-driven marketing segments such as high-value customers, low-value customers, demographics, or psychometrics, using data in your CRM (Customer Relationship Management) or other enterprise systems. Each of these approaches helps to tell a larger story about how customers perceive your brand across their entire customer experience. There is no single right way to approach this problem.
To supplement the various touchpoint approaches I’ve just mentioned, you might want to take a general customer experience poll using a relationship-based survey like Amazon does. This approach involves sending a survey through email, SMS, or whitemail, independent of any given customer interaction, and helps you to gather information about your aggregate brand customer experience that transcends individual departments, groups, and touchpoints. While there are many different ways to approach your various touchpoints, always keep in mind that your customers’ time is the most precious thing of all, so don’t overdo it by subjecting customers to too many measurements.
What Flavor of CX Survey Measurement Would Be Best for You?
There are many different methods and measurement tools that you can use to benchmark and compare the customer experiences of your digital properties. Each of these has its own strengths and weaknesses—and in this author’s humble opinion, there are far more similarities than differences between them. Despite this, it seems that there is a new approach in vogue every year or two that is said to be completely different from the old way of doing things.
Religious debates aside, you should choose the measurement approach that would be the least painful for your customers, while still allowing you to meet your business goals. Each tool for measuring customer experience simply tries to report on the degree of subjective positive or negative feeling that resulted either during or after a customer experience. Keep this in mind when reading vendors’ marketing messages that claim their tool is the only tool that provides what you need.
Here is a list of some useful CX survey measurement tools:
- General CSAT (Customer Satisfaction)
- question—How satisfied are you with X?
- scoring—Typically, a 1–5 Likert scale; or whatever scale you want
- usage—Best for general purpose measurement
- ACSI (American Customer Satisfaction Index)
- questions—Three questions, with 1–5 Likert scales
- scoring—A 0–100 scale
- usage—Strong in retail and ecommerce, with benchmarking and delivery through Asci.org and Foresee Results
- example—ACSI benchmark report
- NPS (Net Promoter Score)
- question—How likely would you be to recommend X to a friend or colleague?
- scoring—A 0–10 survey scale, with a resulting promoter score between –100 and +100, based on the percentages of promoters, detractors, and neutral results
- usage—Do on your own or get benchmarking through SatMetrics. Popularized by Fred Reichheld’s book, The Ultimate Question.
- example—NPS benchmark report
- CES (Customer Effort Score)
- question—How much effort did you personally have to put forth to get your request handled?
- scoring—Simple 1–5 scale
- usage—A fairly new entrant to the space. Probably has the greatest correlation to user experience. Popularized by a Harvard Business Review article.
- CxPi (Customer Experience Index)
- questions—Ask three questions—about whether it meets customers’ needs, is easy to work with, and is enjoyable.
- scoring—1–5 scales
- usage—Another fairly new measurement that Forrester Research has popularized and seems to have broad applicability.
- example—CxPi benchmark report
Delivery Methods and Tools
The cost of using the CX measurement methods that I’ve described in this article can range from nearly zero—if you decide to take a roll-your-own approach and custom design a survey delivery tool and reports—up to tens or even hundreds of thousands of dollars a year for an enterprise organization that is looking to deliver a comprehensive Voice of the Customer (VoC) measurement infrastructure and purchase relevant benchmarking reports.
One promising measurement and benchmarking tool that is worth mentioning is CX Benchmark, which provides a free initial benchmarking study to show how you compare to others in your industry.