Making the Case for Improving the Customer Experience
By Dave Ensing, Vice President of Automotive Solutions Strategy, MaritzCX
Our clients are often asked by their executives, “How do we know improving the customer experience is important? How do we know it is worth it to our business?” To answer those questions, we commonly conduct “linkage studies.” These studies examine the relationship between measures of customer experience and downstream business metrics such as customer retention, increased sales and higher profits. Most people think it would be easy to establish a positive relationship between good customer experiences and good business outcomes. The general assumption goes something like this: happier customers churn less, buy more, are willing to pay more, and recommend products and services to friends. Therefore, happier customers lead to increased revenues, increased profits, and higher stock prices.
Unfortunately, it isn’t as simple as that. In their paper, “The High Price of Customer Satisfaction,” Keiningham, Gupta Aksoy and Buoye note that the overall academic literature “consistently finds that there is a positive, statistically significant relationship between satisfaction and a host of business outcomes…” However, their work also indicates that customer satisfaction explains only about 1% of the variation in companies’ stock prices. They go on to cite studies where there appears to be a reverse relationship between customer satisfaction and stock price (i.e., most hated companies perform better on stock price than their peers) and reference calls from consultants to abandon customer satisfaction initiatives because they are a waste of money.
So, why the mixed bag of findings? Why can it be difficult to show a positive and meaningful relationship between good customer experiences and good business outcomes? There are several reasons why demonstrating this connection can be problematic. Many of these problems stem from the fact that these analyses are correlational in nature. Generally, all we can do is look at how two or more variables (e.g., customers’ levels of satisfaction and company profits) change together in real-world situations. We cannot conduct an experiment actually manipulating the customer experience to determine the effects of that manipulation. To do that, we would have to convince our clients to randomly assign their customers to groups — intentionally treating some customers poorly while giving others the normal level of customer service — and compare the different outcomes of those groups. Obviously, that will never happen, nor should it.
The fact that these studies are correlational in nature leads to two classic problems in interpreting their results.
• THE THIRD-VARIABLE PROBLEM:
College students in research courses often learn that a third variable can sometimes affect both the predictor and outcome variable in correlational research, and thereby distort the relationship between the two. In the case of linking the customer experience and business outcomes, the third variable is often high product demand. A product that suddenly is in very high demand can lead to increases in sales and profits, but decreases in customers’ satisfaction with the sales process. This is because customers may have trouble finding the product, encounter long lines and wait times, have to deal with sales personnel who are overwhelmed with unusually high volumes of customers, and pay a premium for the hot product. Therefore, high demand is driving sales and profits up, while at the same time driving customer satisfaction with the sales experience down. Thus, there appears to be a negative relationship between sales and profits and the customer experience. When conducting linkage studies, there is not a lot we can do about this problem other than to try to control for potential third variables that might be influencing both the customer experience variable and the outcome variable.
• BI-DIRECTIONAL RELATIONSHIPS:
Not only can increased sales and profitability seem to affect customer satisfaction through the indirect effects of a third variable like a hot product, but sales may affect customer satisfaction directly. Specifically, if sales and number of customers are down, it is much easier to satisfy those few customers who are left. I love to shop at stores that are in the early phases of going out of business (before they lay off staff). There are fewer shoppers to deal with, the staff is readily available to help me, and the shelves are usually well stocked. Because of this, stores with fewer sales often have higher levels of customer satisfaction. To address this issue when looking at the relationship between customer experience and sales, it is usually better to conduct the analysis across at least two points in time and relate change in customer satisfaction to change in sales. There are also a number of other pitfalls we need to be aware of when trying to relate the customer experience to business outcomes. These are related to what we actually measure and how we link them together to form our model of how customer experience links to business outcomes.
There are also a number of other pitfalls we need to be aware of when trying to relate the customer experience to business outcomes. These are related to what we actually measure and how we link them together to form our model of how customer experience links to business outcomes.
• INDIVIDUAL EXPERIENCES OFTEN AREN’T ADDITIVE:
Rawson, Duncan and Jones point out in their Harvard Business Review article “The Truth about Customer Experience” that many companies track transactional customer satisfaction and relate that to measures such as customer loyalty. However, more important is a customer’s entire relationship with the brand or product. A customer’s overall satisfaction often will not be the sum or average of his or her individual transactional experiences. For instance, I might be quite happy with the way my cable TV installation went or with how my automotive service event went — but I would have been even happier if I had not had to go through those events in the first place. In another common scenario, the effect of several great customer service events can easily be undone by one poor service event. Therefore, it is much better to use overall relationship measures of customer satisfaction rather than transactional measures when linking the customer experience to business outcomes.
• LAGGING INDICATORS:
Businesses typically respond quickly to low customer satisfaction ratings and slumping sales with improved customer service (e.g., better training, more employees, greater follow-up, etc.) and anticipate similarly quick increases in satisfaction ratings and business outcome metrics. However, while changes in customer experience metrics occur quickly, many changes in business outcome metrics take weeks, months or years to materialize. For instance, if a dealership improves its sales experience, it may produce a small, short-term bump in sales due to more positive word of mouth recommendations. However, the real bump in sales will come three to five years down the road, when customers who experienced those improved sales experiences are ready to purchase a new vehicle. This lagging indicator effect can cause it to appear as if changing the customer experience does not change business outcomes. Therefore, when conducting linkage studies, it is best to choose business outcome metrics that will show change in a relatively short amount of time.
• NONLINEAR RELATIONSHIPS:
Humans tend to think linearly. In this case, we think that there should be a straight-line relationship between the customer experience and business outcomes — the worse the experience, the worse the outcomes, and the better the experience, the better the outcomes. While this is generally true, the relationship often isn’t a straight line. The relationship tends to “flatten out” at the extremes of the customer experience scale. We’ll discuss why below. What is important to note is that linear measures (e.g., correlation coefficients and regression coefficients) of the link between customer experience and outcome variables may underrepresent their true relationship.
• UNDERSPECIFIED MODELS:
When people ask, “Why doesn’t the customer experience relate to sales?” they are often thinking of a very specific measure of customer experience and a very broad measure of sales. This is called an underspecified model because it is only considering one factor of many that influence the outcome measure. For instance, in the automotive industry people often wonder why dealership sales satisfaction doesn’t relate more strongly to a vehicle brand’s sales. The reason is that the dealership sales experience is important but only a small part of what affects a brand’s sales. Other factors include vehicle quality and reliability, vehicle desirability, vehicle availability, the brand’s overall reputation, customers’ interactions with the brand, overall economic factors, and a whole host of other things. To address this issue in linkage studies, it is best to either use a simple linkage study in which the predictor variable and the outcome variable are “close together” in terms of causality, or perform a complex linkage study that takes into account all of the variables likely to affect the outcome variable.
What do these examples illustrate? We need to be thoughtful when investigating the relationship between improving the customer experience and producing positive business outcomes. However, when we are thoughtful, the positive relationships between customer experience and business outcome variables are almost always demonstrated.
Conclusion
Sometimes linkage analyses can produce counterintuitive results if researchers or businesses are not careful to account for all of the nuances in the data. Factors such as underspecified models, direction of causality, third variable problems, and putting too much distance between the predictor variable and the outcome variable all have deleterious effects. However, when done thoughtfully, a positive relationship between customer experience metrics and business outcome variables is almost always found. 
About MaritzCX
MaritzCX combines technology, data, expertise, and services to help you instantly understand and improve every customer experience. This powerful combination brings together deep research expertise, award-winning technology, and best practice services unlike any other CX company. MaritzCX provides everything companies need to transform all their static customer feedback into meaningful, actionable insights that ensure better business outcomes.
If you’re interested in CX, implementing a CX program, or need research for a specific line of business (LOB), let us help you.
To demo a product or to contact MaritzCX sales, call 385.695.2800
To download the Customer Experience Is Your Business eBook visit maritzcx.com/mcxbook
