Chapter 12 · Data Analyst
Appendix A · Ten principles to carry into the job
~3 min read
If you forget everything else, keep these. They are the distilled judgment that took years to internalize, and they apply on your first day and your thousandth.
- Know the decision before you query. Data serves a choice; find the choice first.
- Always name the denominator. A rate without its base is a trap.
- State the grain. Know what one row means before you aggregate.
- Watch for fan-out. A join that multiplies rows will inflate your totals.
- Segment before you trust an average. The aggregate often hides the story.
- Reconcile to a known total. Catch errors before the meeting, not during it.
- Correlation is a hypothesis, not a finding. Prove cause with an experiment where you can.
- Pull the minimum personal data. The safest field is the one you never selected.
- Lead with the answer. Takeaway first, evidence on request.
- State your caveats. Volunteered uncertainty is what makes your confidence credible.
Get the next chapter and weekly interview tips by email
One short email per week. Skim in a minute. Unsubscribe anytime.
