A surprising number of serious game databases began as exactly what you'd expect: a spreadsheet someone kept updating out of personal interest. The gap between that spreadsheet and a genuinely searchable archive is bigger than it looks.
Flat Rows Versus Relational Structure
A spreadsheet naturally encourages flat rows — one row per game, with genre or platform typed directly into a cell as free text. A searchable database instead treats genres, platforms, publishers and developers as their own linked entities, which is what actually enables consistent filtering rather than fragile text matching that breaks the moment two entries spell something slightly differently.
Consistency Requires Constraints
Free-text spreadsheet fields tolerate "RPG," "Role-Playing," and "role playing game" existing side by side as if they were different things. A structured database enforces a single, canonical value per field, which is unglamorous but is the actual mechanism that makes browsing by genre or platform return complete, reliable results.
Search Needs to Match How People Actually Ask
A visitor searching for "old PC games from 1996" or "DOS adventure games" is asking a structured question even if it looks like a free-text search. Supporting that well requires the underlying data to already be organized by year, platform and genre — the payoff for the structural work described in What Makes a Good Old PC Games Database.
Growing Pains Are Normal
Almost every serious reference archive went through an early, messier version of itself before adopting proper structure. See Searching an Old PC Games Database for what a mature version of that search experience looks like from a visitor's side.
