I have sat in both chairs.

I started in sonar in 1984 and spent the first part of my career on the receiving end of these decisions. I was project manager for the CSU-83 sonar system on the Västergötland class submarines, for the flank array on A14 Näcken, the first Swedish submarine fitted with one, and for the integrated CSU-90 suite on the Gotland class. Later I managed the HYDRA multi-mission sonar system from General Dynamics Canada for the Visby stealth corvette, and contributed to Submarine 2000, Viking, and what eventually became the A26.
On the mine warfare side I worked on the Landsort class MCMV upgrade and on MCM programmes for navies around the world. I was also involved in one of the first commercial autonomous underwater vehicles, the Thales AUV, carrying an integrated payload of sidescan, forward-looking sonar, multibeam echosounder and sub-bottom profiler.
Then I crossed the table and spent two decades on the supply side. That period taught me how sonar actually gets sold. How performance figures are chosen, which conditions get quoted, what a demonstration is designed to show and what it is quietly designed to avoid. None of it is dishonest. A datasheet is simply an argument, and arguments have authors.