This article originally appeared on HBR.org 

When you are driving a car, you get a steady stream of feedback from the road, from other drivers, from the dashboard gauges, and from the car itself. You know how to interpret the feedback and respond to it in real time, such as slowing down when the road feels bumpy. You may also begin observing patterns—the car begins to shimmy when you hit 35, say — and the patterns may lead you to other actions, such as making an appointment for repairs.

Meanwhile, dealers and automakers are gathering feedback of their own from you and millions of others and are making plans for fixes, improvements in the next model, even recalls. Note the close connection here between feedback and action. Feedback that doesn't lead to action is meaningless. They are inseparable aspects of the same system.

Customer-centric companies learn to connect customer feedback to action in just this way. Consider the experience of a manufacturer of premium kitchen and bathroom fixtures. At one point, the company’s customer surveys were indicating that its distributors, most of whom sold competitors’ products as well, needed more help communicating the competitive advantages of its products, such as innovative design and ease of installation.

Hearing the feedback, the company’s managers and frontline employees took action. Sales reps began sponsoring workshops in distributors’ showrooms to teach contractors how easy the products were to install. The reps created compelling new floor and window displays to showcase the products’ decorative appeal. Analysis of the feedback revealed other patterns, such as the fact that sales reps seemed to be visiting some distributors too frequently and others not frequently enough. The company cut back on the unproductive sales calls, freeing up an estimated 25% of selling capacity.

All such actions are like natural experiments. A company can act, observe the results, and modify the action as necessary. Often, of course, addressing feedback from customers requires an investment of resources—redesigning a product, say, or reengineering a process. A company then needs to compare the required cost with the likely benefit.

Take Carolina Biological Supply, a midsize family-owned company that Fred Reichheld and I featured in our book The Ultimate Question 2.0. The company sells math and science education products, mostly to high school teachers and college professors who need the products to fit into precise points in their curricula. At one point, customer feedback indicated that product availability was a major concern. The fill rate per line item stood at only 92% — and with an average of 2.5 lines per order, CEO Jim Parrish explained, “that means 20% of the time we didn't have everything the customer wanted.” To mitigate the problem, the company increased its investment in inventory, raising the fill rate to 98% on top items and 95% on the rest. Analysis had shown that this was the appropriate balance point between costs and benefits.

Of course, the precise connection between feedback and action will differ, depending on what a particular group of employees or executives is responsible for, how that group receives feedback, and what the cost of proposed actions may be. But the framework is always the same. Whatever their level, people in a company that gets regular customer feedback instantly understand the goal when someone proposes an action designed to improve the customer experience. If the economics justify the move, support for it tends to be widespread.

In a car, the feedback you get and the actions you take help you arrive at your destination safely. Why should it be any different in a company?

Rob Markey is coauthor of the book The Ultimate Question 2.0: How Net Promoter Companies Thrive in a Customer-Driven World (HBR Press). Markey is a partner and director in Bain & Company’s New York office and leads the firm’s Global Customer Strategy and Marketing practice.