Ever notice how different a product looks after a few years of regular use, compared to the day you first took it out of the box? Aside from normal wear and tear, some products just seem to fare better than others when it comes to long-term durability. Yet aside from fielding complaints from highly vocal customers, most companies probably never get to see what their typical product looks like after it’s spent time in the wild.

Sure, you can run simulations in the lab to mimic the effects of 10,000 key presses or 50 drops from a coat pocket. But these can’t possibly account for the huge variety of ways that real people use and abuse a device in their day-to-day lives. For instance, I once talked to a customer whose coworker had spilled an entire soft drink into their laser printer. Somehow I doubt that HP or Samsung runs lab tests on their printers for sugary beverage resistance. And there are thousands of other weird things that people do to products because, well, they’re people.

I propose a different approach to long-term durability testing. Manufacturers should randomly select a group of customers who would participate in a sort of trade-back program. After a set period of time, maybe three years or however long the product usually lasts, they would be able to exchange the old product for a brand new one, at no charge. The catch is that they have to return the old product to the company along with some notes on what they liked and disliked, especially regarding durability, along with anecdotes of how they used or abused it during ownership.

By using this process, product designers would learn a lot more about how their devices hold up in real usage. Coupled with the anecdotal evidence of how a particular customer made use of that device, this information should make it much easier to improve product durability and reliability.