In aerospace we have technical reviews for anything worth producing and showing to other people: requirements, technical drawings, test plans, integrated designs, etc. We work with things too complicated for a single person to understand how it all works when it's put together. So we get groups of people together to find the problems,

The problem is that many people who approach the reviews as the producer of the content to be reviewed really hate that role. It is an uncomfortable role. Your literal job is to hold up things you've spent blood, sweat, and tears working on so that other people can hit them with baseball bats. (OK, maybe there's a better word than "literal", but we'll go with it.)

So what you often get is one of two things: (1) People who don't produce a review package that is easy to understand for reviewers; (2) people who host reviews for the sole purpose of having the review being considered done. Both are a problem for everyone.

The first is a problem because it makes the reviewer's job harder. Now a reviewer has to burn more time and effort to come to the minimum level possible to begin contributing to the review—actually having to work to get to level zero to start. The second is a problem because it subverts the purpose of the review. The purpose is not to check some box labeled "have review", the purpose is to find problems so that they can be fixed.

That's my crank theory: make it easy for your reviewers to find problems.

You have to produce a review package in addition to the thing being reviewed. Bringing a technical spec up for review? You should have something else that explains the tough parts, explains how the spec relates to other specs and products—things that aren't being reviewed themselves, but are related to the integrated whole. Make it absolutely, 100% as easy as you can for your reviewers to contribute. It's not about holding their hand (the pejorative usually lodged against helping people understand things), it's about setting them up in the best possible situation to do good work.

You have to point out the things that you think will be problems, and promote those things so that reviewers will see them, consider them, smash them, and ultimately help solve them. Aerospace is no place for letting problems slide. Everyone knows that when they make something that it has problems—it's a complicated field. Spray paint those things orange and get to work collecting issues against them.

It's an uncomfortable place to stand in front of people for the purpose of being told you're wrong. It also results in the best work. You have to choose discomfort to win.