Diagnose a KPI Drop: Junior Analyst's 5-Step Fix

Who This Helps

This is for you, Junior Analyst. You just saw a number drop—maybe revenue per user slipped 12% this week. Your job is not just to report it. Your job is to find the "why" and hand your boss a fix they can act on. The Founder Finance Basics Mission Pack teaches you to do exactly that, starting with the Unit Economics Snapshot mission.

Mini Case

Imagine you run a small SaaS. Last month, your monthly recurring revenue (MRR) was $50k. This month, it's $44k—a 12% drop. Your first instinct might be to panic. But you have a process. You grab the data and start asking: Is it fewer customers? Lower spend per customer? Or both?

You check your unit economics. Turns out, your average revenue per user (ARPU) fell from $100 to $88. But your customer count stayed flat at 500. So the problem is pricing or usage, not churn. That's a huge clue. You now know where to look: your pricing page or your feature adoption metrics.

Do This Now (5 Steps)

1. Pull the raw numbers. Get your KPI for the last 30 days and the 30 days before. Write them down. Don't guess.
2. Split the KPI. If it's revenue, break it into customers times average spend. If it's profit, break it into revenue minus costs. Find the piece that changed.
3. Compare the pieces. Which piece dropped the most? In our case, ARPU dropped 12%. That's your target.
4. Ask "why" three times. Why did ARPU drop? Maybe fewer upgrades. Why fewer upgrades? Maybe a pricing change. Why the pricing change? Because you ran a test. Now you have a root cause.
5. Write one recommendation. Example: "Revert the pricing test for segment A and monitor ARPU for 7 days." Keep it short. Your boss will love it.

Avoid These Traps

• Don't report the drop without context. A 12% drop sounds scary until you know it's from a pricing test. Always add the "why."
• Don't blame one thing too fast. Check at least three possible causes before you pick one.
• Don't forget to check the denominator. If your customer count changed, that changes everything.
• Don't write a novel. Your recommendation should fit in one sentence. If it doesn't, you haven't found the root cause yet.
• Don't skip the fix. Analysis without a recommendation is just noise. Ship a decision.
• Don't assume the data is clean. Verify your numbers before you present them. A typo can waste a whole meeting.
• Don't ignore seasonality. Did this drop happen last year too? Check the calendar.
• Don't overcomplicate. Your first pass should take 30 minutes, not 3 hours. Speed matters.

Your Win by Friday

By Friday, you will have diagnosed one real KPI drop from your work. You'll have a one-page analysis with a clear root cause and a one-sentence recommendation. Your boss will see you as the person who doesn't just find problems—you find answers. And honestly, that feels pretty good. Plus, you'll have practiced the exact skill from the CAC Payback Triage mission in the Founder Finance Basics Mission Pack. That's a win you can take to the bank.