There's a type of meeting that happens in every organization. Someone projects a screen full of numbers. Someone else asks "so what?" A silence follows. Then someone says "we should dig into that." Nobody does. The meeting ends. The numbers sit.
That's not a data problem. That's an interpretation problem.
What Interpretation Looks Like
- A slide with three charts and one sentence per chart: what it means and what we're doing about it
- A weekly update where each number is followed by a clause: "which means..." or "so we're..."
- A dashboard where one number is highlighted in red with a callout: "this is the one that matters this week"
- An email that says "we're down 12% on X. Here's why. Here's what we're doing. Here's what we need."
The person who sends interpretation is doing work. The person who sends a dump is offshoring that work to the reader.
How to Convert a Dump into a Decision
For any metric you receive — in a meeting, an email, or a dashboard — apply a three-step filter:
Step 1: What is this number? Name it clearly. Not "Q3 performance" — "new customer conversion rate."
Step 2: What does it mean for us? This is the interpretation step. Is it good? Bad? Better or worse than expected? Why? Be willing to state a hypothesis even if it's incomplete.
Step 3: What are we doing about it? A metric with no implied action is a historical record. A metric with an implied action is a decision tool.
Before:
> Activation dropped from 64% to 52%. Trial starts increased 9%. Support tickets increased 18%.
After:
> Activation dropped from 64% to 52%, likely because the new setup flow added two required fields. Trial starts are up, but support tickets are up 18%, mostly from setup confusion. Recommendation: roll back the required fields for new accounts this week and rerun the activation cohort next Friday. Owner: Product Ops.
The second version is not longer because it has more data. It is longer because it contains judgment.
The Underlying Principle
Data without interpretation is a product nobody bought. The person who produces the data is the one who should shape its meaning. If you can't explain what a number means, you're not ready to share it.
