01
Name grain and population
Most fights are grain fights.
Made for Business Analysts
Define a Metric So Product and Finance Stop Using Two Numbers
Also useful for: people in Data, people in Technology
35 min
Intermediate
Data
Business Analyst
Career outcome:
Avoid Professional Risk
Write the definition of one contested number before you build another chart.
We'll use them to write your prompt.
Completed ✓
Make Your Play
Give us the situation. We’ll build the prompt.
When weekly active and "engaged" mean three things, every review becomes a data fight.
Your answers save as you type.
What you'll do
This is the plan inside your prompt. Add your details first. Then follow these steps with ChatGPT or Claude.
02
Write the formula
SQL later. Logic first.
03
List the two numbers that will still disagree
Honesty beats a fake single source.
04
Assign an owner
Orphans drift.
Make Your Play
Define a Metric So Product and Finance Stop Using Two Numbers
Tell us what the prompt needs to know
Start with the questions below. Each answer personalizes the playbook and builds the prompt you’ll use at the end.
Your situation
Paste how they describe it.
0/10000
Warehouses, events, known late data.
0/10000
What someone will do with it.
Bots, duplicates, timezone, deleted users.
0/10000
This playbook was updated. Review your inputs before copying.
What you'll do
Check a step when you finish it.
01
Name grain and population
Most fights are grain fights.
02
Write the formula
SQL later. Logic first.
03
List the two numbers that will still disagree
Honesty beats a fake single source.
04
Assign an owner
Orphans drift.
Fill in the fields above to generate your prompt preview.
Assembled prompt
Beginner instructions
Fill the fields with this week’s real details. Incomplete notes beat invented polish.
Run the assembled prompt, answer follow-ups with specifics, then copy the deliverable into your own doc and verify names and numbers.
Advanced instructions
Add constraints the model can’t see: politics, licensing, cash timing, union rules, or family bandwidth.
After the first pass, ask it to attack the weakest assumption in the draft.
Use this when
Two teams report different "conversion" numbers, or a KPI is about to go on an exec slide.
You'll need
- The contested metric in each team's words
- Where the data lives
- Decision the number will drive
- Known caveats
You'll get
- A a metric spec with grain, filters, owner, and known caveats you can use without rewriting it from scratch.
By the end of this playbook
You will have:
- A a metric spec with grain, filters, owner, and known caveats you can use without rewriting it from scratch.
Examples
Progress 1 of 4