Customer feedback is important to any business’ growth. That’s why many send out customer satisfaction surveys and managers pay attention to that feedback.

However, sometimes the feedback system is faulty or gets misinterpreted to the point of being useless. Take the feedback system Uber used to have. At the end of each trip riders were prompted to leave feedback in the form of a star rating from 1 to 5. As Uber drivers are contractors, the corporation wanted to ensure customers received good service and a great customer experience from any given Uber driver. Any driver that fell below a certain star rating was disciplined by the company with a suspension or, in some cases, contract cancellation.

An article in Wired magazine details how Uber came to realize that the star system was faulty. It failed to truly reflect a customer’s experience and was vague to the point of being meaningless. Some people gave 5-stars to a driver even though they didn’t have the best ride, simply because they felt bad doing otherwise. Other customers gave drivers low star ratings for reasons that were beyond a driver’s control, including traffic, or how long it took to get picked up.

Uber also realized that its star system didn’t provide the drivers themselves with any useful feedback on what their customers liked or didn’t like about a trip. Therefore, drivers had no opportunity to improve or correct something as they had no way to know what was bothering their passengers if it was not stated during the trip. Uber has now revamped its system so that it includes the star rating while also allowing customers to give feedback in the form of virtual stickers and, if the passenger desires, they can leave personalized feedback via a note. This provides Uber with a better assessment of the driver and can also instill a sense of pride or fulfillment in the driver for a job well done.

In the auto industry we face a similar dilemma. Manufacturers survey customers and assign CSI scores to each survey. These scores can mean the difference between a dealer making or losing a substantial amount of money. However the surveys are formatted similar to Uber’s old 5-star rating system. Each question is either scored on a 1-10 scale or a Yes/No answer. This provides little clarification to the manufacturer or the dealer as to how the customer experience actually went. In addition, manufacturers assign different weights to the questions so a dealer can receive a failing grade on a survey even if every question is answered perfectly except one. Many manufacturers even consider anything less than perfect (100%) as a failing grade. There is nothing wrong with demanding perfection, but it is tough when the grading system is not itself perfect.

Venture back to your school days. Imagine getting a 90% on a math test. In the real world, that would be an “A” and if you or your child brought home straight “A’s”, you would be proud. Yet a dealer who receives a 90% on a survey could be punished monetarily and, unless the customer actually bothers to write feedback, does not know or have any way to change a process or hold someone accountable. In addition, the manufacturer, just like in Uber’s case, has no true idea of just how the customer’s experience went at the franchisee’s dealership.

This isn’t anything new. Dealers have long complained about the unfairness of survey grading and losing money as a result of the weighted questions which don’t really reflect how their customers are treated.

Perhaps the industry can take a page out of Uber’s playbook. True feedback should be in-depth and judged on an individual experience basis. Only in this way can a dealership be judged accordingly, change any needed processes, truly improve and be fairly rewarded.