Brochure specifications tell you what a platform can theoretically do. They tell you almost nothing about what it will cost to integrate, operate and eventually replace.
Technical fit against the actual requirement is the obvious starting criterion, but it should be scored against the requirement document, not against the vendor’s own feature list — a platform with more features than required is not necessarily the better fit if those features add integration complexity.
Integration effort deserves its own line item: does the platform expose an open API, does it have a track record of integrating with the specific adjacent systems on site, and how much custom development has that integration historically required elsewhere.
Vendor track record and financial stability matter more for security technology than for most IT purchases, because a platform with a five-to-ten-year expected service life needs a vendor likely to still be supporting it in year eight.