What makes an analysis decision-ready

A chart is not a decision. Decision-ready work states the question, the unit of analysis, what is missing, and what would change the recommendation. It also says who the analysis is for.

In housing, that might be whether AMI-labeled units are affordable to the neighborhood that has to live with them. In enforcement data, it might be whether two files can be compared at all. In markets, it might be whether a tweet-sentiment correlation is large enough to matter.

I aim for artifacts a hiring manager can open: a report, a documented pipeline, a dashboard with evidence. Confidence is earned by showing the join, the missingness, and the limit—not by hiding them.