Diagnose a KPI Drop with a Runway Trigger Tree

Who This Helps

This is for junior analysts who need to move from spotting a problem to explaining it clearly. It pulls directly from the Board Finance & Runway Narrative course, where you learn to build a disciplined finance story. Think of it as your first step toward creating that one-page board memo.

Mini Case

Your weekly report shows a 15% drop in new user activation. The team is pointing fingers at the new signup page, a recent marketing campaign, and even server performance from last Tuesday. You have 45 minutes before the stand-up to get a real answer. No pressure, right?

Do This Now (5 Steps)

1. Isolate the Signal: Don't look at everything. Pick the one KPI that matters most right now. Is it activation rate, purchase conversion, or daily active users? Write it down.
2. Check Your Time Frame: Is this a one-day blip or a week-long trend? Compare the last 7 days to the previous 7. A single bad day is a different story than a steady decline.
3. Segment the Drop: Slice your data. Did the 15% drop hit all user types, or just those coming from a specific source like paid social? This tells you where to look next.
4. Build a Simple Trigger Tree: This is a key tool from the course's 'Runway Trigger Tree' mission. Draw a line for your KPI. Branch out with possible causes: 'Marketing Source,' 'Product Feature,' 'Technical Issue.' Under each, list 1-2 specific, testable hypotheses.
5. Find the First Proof: For your top hypothesis, find one piece of data that supports or kills it. For example, if you think it's the new signup page, check if the drop started the day it launched. Got it? Great. No? Move to the next branch.

Avoid These Traps

• Chasing Every Number: You see a 2% dip in another metric and get distracted. Stick to your main KPI until you have a cause.
• Starting with a Solution: 'We need to redesign the onboarding' is not a diagnosis. That's a prescription before the check-up.
• Blaming Without Data: 'The engineering deployment broke it' is a story, not a root cause. Find the evidence first.
• Getting Lost in Details: Your goal is a clear, probable cause for a meeting, not a PhD thesis. Limit your deep dive to 3 potential root causes.

Your Win by Friday

By Friday, you'll have a one-sentence explanation for that KPI movement. Something like: 'The 15% activation drop is concentrated in users from our recent Partner X campaign, likely due to mismatched expectations set in their ad copy.' You'll present it with one clear chart. That's how you stop the meeting chatter and start the solution talk. You've got this.