Which Door Is Stuck? Diagnosing a Real Funnel with the AARRR Decision Tree
Objective: Given a real four-number funnel snapshot, apply the lesson's diagnostic decision tree in the correct order to find the one stage worth fixing this quarter.
You're the growth analyst at Lenskart. Leadership wants to know which AARRR stage to fix before the next budget cycle, and they've asked for a one-page answer, not a 20-tab dashboard.
Run the lesson's four-question decision tree, top to bottom, against Lenskart's real Q3 numbers. Stop at the first 'yes' and don't skip ahead to a stage that looks more exciting to fix.
Before you start
What you'll need
Free path (everything below is enough to finish)
No account friction, handles four numbers and four if/then checks with zero setup
Paid upgrades (optional, faster/deeper)
Source of truth for cohort-based retention once you move past a single provided snapshot
The process
1 step
Step 01 of 01
The lesson's decision tree checks stages in a fixed order: Retention first, then Activation, then Referral, then Revenue, then Acquisition last. You stop at the first stage that fails its threshold.
Lenskart's Q3 snapshot: Day 30 retention 34%, activation rate 41%, K-factor 0.6, LTV:CAC 2.4. Which stage does the decision tree tell you to fix first?
Procedure
- Import funnel-snapshot.csv, which has one row of Q3 numbers: retention, activation, K-factor, LTV:CAC
- Check question 1: is Day 30 retention below 20%? Lenskart's is 34%, so answer No and move on
- Check question 2: is activation rate below 37%? Lenskart's is 41%, so answer No and move on
- Check question 3: is K-factor below 0.5? Lenskart's is 0.6, so answer No and move on
- Check question 4: is LTV under 3x CAC? Lenskart's is 2.4x, below the 3:1 target, so answer Yes and stop here
Lenskart Q3 funnel snapshot Day 30 retention: 34% (threshold 20%) -> pass Activation rate: 41% (threshold 37%) -> pass K-factor: 0.6 (threshold 0.5) -> pass LTV:CAC: 2.4x (threshold 3.0x) -> FAIL, stop here Diagnosis: fix the Revenue stage (pricing, packaging, or upsell path) before scaling Acquisition further.
Healthy
The team stops at the first failed threshold and builds a pricing/packaging brief, leaving Acquisition spend untouched until LTV:CAC clears 3:1.
Unhealthy
The team sees a decent-looking activation number and decides to run a splashy referral campaign instead, skipping past the actual failing gate.
What this means
The decision tree is sequential on purpose. A stage that passes its threshold is not 'done', it's just not this quarter's bottleneck.
So what do I do about it?
| Symptom | Action | Effort |
|---|---|---|
| LTV:CAC sits at 2.4x against a 3:1 target | Pull the top 20% of customers by LTV and check what pricing tier or bundle they're on before touching acquisition spend | 30 min |
Final deliverable
A one-page diagnosis memo: which stage failed the decision tree, the exact threshold it missed, and the one recommended fix for next quarter.
See a reference example
Warby Parker Q2 funnel snapshot Day 30 retention: 16% (threshold 20%) -> FAIL, stop here Diagnosis: fix Retention before anything else. Referral and Revenue numbers are irrelevant until the leaky bucket is patched; scaling acquisition now would only accelerate churn.
Success criteria
You're done when you can:
- Checks all four thresholds in the lesson's specified order, not by eyeballing which number looks worst
- Stops at the first failed threshold instead of continuing down the tree
- States a specific fix tied to the failing stage, not a generic 'improve marketing' recommendation