Designing data requests people want to answer
Templates from the Cross-Department Data Requests program, adapted for remote teams.
Bad requests read like riddles. Good requests read like mini briefs with a decision, a timeframe, and a definition reference. We publish a shared intake template inside the Cross-Department Data Requests offering because the social cost of vague asks is higher than most leaders estimate.
Start with the decision, not the chart. Charts are outputs. Decisions force prioritization when time is constrained. Next, specify the population and the exclusion rules you already believe are true. Analysts can correct them quickly when they are explicit.
Close with how you will use the answer: board readout, sprint planning, or vendor negotiation. That context helps analysts propose a cheaper validation path when full rebuilds are unnecessary.
If your organization forbids certain dimensions in shared channels, say so up front. Surprises late in the process erode trust faster than imperfect SQL ever could.