For scrappy hardware startups and billion-dollar tech giants alike, product reviews matter. They’re the first indicator of success or failure. As timelines from announcement to launch tighten, product reviewers get shorter and shorter deadlines. So companies came up with a solution: to offer beta hardware, also known as preproduction units.

It is incredibly difficult for journalists to serve the reader with a preproduction unit. Am I supposed to guess which issues will be ironed out in the final product? Am I supposed to ignore all the software bugs because they can be fixed more easily and instead focus on the hardware issues?

The logical thing to do is to include every issue and simply tell the reader this is not the product they will end up buying. But that is also a fruitless endeavor — it doesn’t help the reader or the company.

As a result, I avoid writing reviews solely based on unfinished products.

Do you want me to beta-test your new device? Great, send me a preproduction unit and I’ll tell you, not the public, what I think. Do you want me to review your device? Great, make sure I get the final version in time.

Let me spell that out again for companies and PR firms reading — there are two good ways to use preproduction hardware. The first is to send the unit over and ask for feedback, not coverage. The second is to send the preproduction hardware and send the final hardware later, but make sure to set the embargo well after the latter is received.

For readers, please hold journalists accountable. If a review does not specify whether the hardware is final, ask in the comments. If “beta” or “preproduction” gets a mention, ask how much time the author had with the final version.

The reader should not have to figure out which kinks were fixed and which were not. And journalists should not agree to write about products that will never see the light of day.

If you send me your unfinished product, I won’t finish my review.

ProBeat is a column in which Emil rants about whatever crosses him that week.