Different decisions need different scoring systems.
We do not force prop firms, software, books and countries into one generic score. Each review area uses its own criteria and explains what is being measured before a ranking or recommendation is shown.