Feedback Lost — Satisfaction Surveys
A Concise Case Study on Processes with Timing Bias
Every human interaction creates some level of feedback but sometimes our products don’t. Even when we augment our process with opportunities for feedback, we often fail to recognize important biases in our timing.
In a society where technology makes it easy to collect feedback, we often overlook the incentive structure, or lack there of, that bias how we collect it. Feedback collected at the start of a process may not apply or inform over the lifetime of that interaction.
Fortunately, at least for our purposes today, we have the typical large purchase sales process as an example… of under performance. The bias in the timing of their customer satisfaction surveys is typically not our biggest complaint, but perhaps it should be.
As you read this article, think about your own products and services. Have you designed feedback systems prone to timing bias? Is it something you have ever considered? Well here is a great example to learn from:
Five Star Surveys
Large ticket purchases like cars, cellphones, and appliances often end the sames way. We are told to expect a phone call. A manager or customer satisfaction specialist is going to ask a few questions and if we don’t give our salesperson 5 out of 5, they may not get paid.
The first problem here is that this is an example of binary feedback system — more on those here. But perhaps it is more important to ask this question-”How do I know if I am satisfied a few hours or even days after my purchase?”. Unfortunately, this call has nothing to do with your product satisfaction. At best it is a weak attempt at collecting feedback about the sales process.
Online Review Sorting
Most online retailers, app stores, and other e-commerce platforms collect feedback. A smaller subset call-out the time between the reviewers purchase and review. Worse still, those that do default to sorting by the recency of the review.
Even when the best tools are available and easy, we are subject to response bias. Few customers return to the site they purchased a product on six months ago to express their continued satisfaction. The irate owner of thousand dollar espresso machine that broke a day after the warranty expired might… but even they are more likely to vent through other channels.
Recency Bias
The businesses that rely on processes like customer satisfaction surveys are either the victims or exploiters of recency bias. This is the believe that recent trends will reflect future performance. The victims believe that today’s satisfaction levels predict the health of future sales. The exploiters suspect that displaying recent satisfaction will convince people they can expect future satisfaction from the product.
Buyer vs Owner
This confuses the customer. I likely mean that a little differently than you might expect. Every business has two types of customers. There are those customers who are actively or have recently bought your product and those that currently own it.
While short term thinking may cause you to favor the former over the latter, great companies are built on satisfying the needs of the latter. Customer satisfaction surveys are not a bad tool, assuming they are employed with the proper perspective and timing.
Sadly, Medium does not collect the amount of time that has past between when you read the article and when you recommend it… but don’t let that stop you from clicking the heart icon anyway. If you have the time, go back and recommend an article you read a while ago — especially if it still has you thinking!
Feedback Lost is a ongoing series provided by Corsair’s Publishing. We seek to provide engaging content that is both thought provoking and entertaining. Other articles on related topics can be found within our other Medium publications.
