Reconstructing a project from evidence
The process behind the case studies on this site. Evidence first, writing second: nothing gets written until the pass across the repository, the strategy documents, the design tokens in the code, the live product and the print collateral is complete.
The rules that make it work are the restrictive ones. Every claim carries a confidence label. Dates come from version control, never from memory. Brand elements are never reconstructed — if the mark is a traced path in the code, that exact path is used. Anything that cannot be captured safely is documented from code and declared as not captured, which is why every study ends with a stated-limits section.
It produced five complete archives, each with its own evidence inventory.
No download yet — this one is written up here rather than handed over as a file. Ask if you want it.