In response to recent tragedies, many commentators have suggested requiring every police officer to wear a video camera and microphone at all times. Some activists even suggest that these videos be publicly live-streamed for maximum accountability. That’s not necessarily a bad idea, despite some technical, budgetary, and legal hurdles.

Other commentators have pointed out that videos aren’t as objective as people think. They are open to interpretation, and risk inflaming opinions on both sides without solving anything. But that’s not the biggest problem with videos. Their biggest weakness is this:

Videos can’t provide the systematic, standardized, quantifiable feedback that we need to understand people’s own perspectives on their everyday experiences with the police.

That’s why police departments should start acting more like the best companies, and measure their customers’ experiences. There are several key tools for doing this, and one of the most important is a customer experience survey. We all get these surveys – they ask about how well the company met our needs, how easy it was to work with, how we felt during the process, how likely we are to say positive things about it, etc.

Retailers use these surveys, as do investment firms, hotels, nonprofits, and every other organization that wants to understand what its customers think and feel, and uncover the drivers of great experiences. If we can easily rate the people who deliver food, sell cars, and hook up TVs, why can’t we do the same for the people who carry guns, write tickets, and make arrests?

The process would be simple and secure: At the end of an interaction with the public, an officer would provide each interlocutor a card with her name and badge number, website and phone number options for completing the survey, and a unique passcode. Each person would then use the anonymous, one-time passcode to answer some basic questions that Forrester Research has identified as essential to measuring customer experiences of all kinds.

Because survey respondents would identify the officers they interacted with, senior public safety officials could compare the public’s experiences with different individual officers – just like my ISP can compare my experiences with the two call center representatives I spoke with to get my Internet service fixed last week. That’s far better than simply measuring people’s attitudes toward the police in general because it would show senior officials which officers need additional training or reassignment away from the public. It would also help leaders identify which officers are worthy of recognition or advancement. As more police departments start requiring officers to wear video cameras, this survey would also point leaders to which video clips they should watch, either for possible problems or as examples of great public interactions.

Some critics might charge that this survey would distract officers from doing their jobs and refocus them inappropriately – even dangerously – on their interlocutors’ perceptions. This objection is a red herring. Nobody expects an officer to stop returning fire at violent drug smugglers to offer survey cards to bystanders, or to wrestle a carjacker with one hand while handing him a survey card with the other. The vast majority of police interactions with the public are calm and routine, so there’s no danger to an officer simply handing out survey cards at the end. But officers would give out cards even following confrontations and arrests – after the situation is under control and everyone’s safety is ensured.

Detractors might also worry that people who are ticketed or arrested will score officers poorly out of spite. Yes, that is going to happen sometimes, but the survey would ask the nature of the respondents’ interaction with an officer to help officials understand the context of their responses. The survey would also allow officials to identify which officers receive the best scores from those they’ve ticketed or arrested – exactly the officers to help train others in creating the best experiences for every member of the public. (The last time I got a traffic ticket my experience with the officer was so good that I emailed the police department to tell them about it. Not kidding.) By the way, it’s also easy to create an algorithm that levels the field between officers with more tickets or arrests and officers who have less, so that comparisons among officers are fair.

There are two possible objections focused on officers’ opportunities to abuse the survey. Simply put, officers could either intimidate respondents into scoring them higher than they deserve, or just take the survey themselves. The best way to address these problems is the same way that law enforcement works for everything else. First, make the sanctions so high that few people will risk breaking the rules. Second, provide enough oversight – in this case, advanced algorithms to spot aberrations in officers’ data and video spot checks to see if an interaction seems to match respondents’ reviews – to be a credible deterrent.

Another potential concern is survey fatigue. People who feel bombarded by too many surveys either don’t finish them or don’t take them at all. However, there are effective strategies for combating survey fatigue, as my colleague Maxie Schmidt-Subramanian has shown. One of the most important is showing people that their responses really matter. If police departments show the public that they are using survey responses to improve, people will keep responding. Brevity is a virtue, as well. To help keep response rates up, these surveys would need to stay focused on measuring the customer experience, and not become vehicles for interesting but extraneous questions.

Skeptics could come up with all kinds of excuses for not implementing police customer experience surveys. But the advantage far outweighs all of them. Public safety officials would gain detailed knowledge of the experiences the public has with individual officers, in all kinds of interactions, every single day. It would be a treasure trove of data that police departments could use to understand what creates great public perceptions, then design training programs, procedures, and incentives to help officers produce those experiences for the public with every interaction.