Sizing the Compounding Engine: Forecasting PLG Expansion Revenue
Objective: Given a starting PLG customer base and two NRR scenarios, forecast 3-year revenue and decide whether the current expansion motion is strong enough to hit a growth target.
You're on the growth team at Wise modeling what happens to the self-serve customer base's revenue if the expansion motion (seat/feature upsells to existing accounts) stays flat vs. improves.
Use the lesson's NRR and expansion-engine framework to forecast revenue under two scenarios and recommend which expansion lever to prioritize.
Before you start
What you'll need
Free path (everything below is enough to finish)
Free, handles both the formula table and the ranking without a paid analytics seat
Free tier supports funnel/cohort queries against real usage-limit events
No access? Use the provided sample data instead of a live Mixpanel account
Paid upgrades (optional, faster/deeper)
Free tiers of Mixpanel or Amplitude cover this exercise completely; upgrade only once you're tracking limit-hits across dozens of features.
Built-in behavioral cohorting scales past what a manual Sheets pull can track
The process
2 steps
Step 01 of 02
The lesson states best-in-class PLG companies hit 120%+ NRR, the compounding engine behind PLG's value even with some churn.
Starting at $2M MRR from existing self-serve accounts, what does the base become after 3 years at 105% NRR vs. 120% NRR?
Procedure
- Set up a 3-year compounding table for $2M MRR at 105% and 120% NRR
- Compute year-end MRR for each scenario using MRR × NRR^year
- Compute the dollar gap between scenarios at year 3
Starting MRR: $2,000,000
Year 1 Year 2 Year 3
105% NRR $2,100,000 $2,205,000 $2,315,250
120% NRR $2,400,000 $2,880,000 $3,456,000
Year-3 gap: $1,140,750/moHealthy
The team sees the 15-point NRR gap compounds into a >$1.1M/mo difference by year 3, purely from the existing base, no new customers assumed.
Unhealthy
Treating NRR as a single-year metric and missing that a 15-point gap barely shows up in year 1 ($300K/mo) but triples by year 3.
What this means
NRR's real cost of being average instead of excellent only shows up when you compound it, which is exactly why the lesson calls it the flywheel's compounding engine.
So what do I do about it?
| Symptom | Action | Effort |
|---|---|---|
| NRR sits near 105%, below the 120% PLG benchmark | Prioritize expansion levers (seat limits, feature gates) over new-logo acquisition this quarter | half day |
Step 02 of 02
The lesson notes 18% of ARR in high-performing PLG companies comes from expansion, driven by usage ceilings (seats, storage, API calls) that feel generous early and limiting at scale.
Which of these three usage ceilings is the strongest expansion lever: transfer volume limits, currency-count limits, or support-ticket limits?
Procedure
- Pull the % of users who upgraded within 30 days of hitting each limit
- Rank the three limits by upgrade conversion rate
- Recommend the top lever for the next expansion campaign
Upgrade rate within 30 days of hitting limit Transfer volume limit 38% <- strongest lever Currency-count limit 14% Support-ticket limit 3%
Healthy
Transfer volume limit gets prioritized because it converts far better, a natural ceiling tied directly to the core value the customer already gets.
Unhealthy
Building a campaign around the support-ticket limit because it's easiest to implement, ignoring that it barely converts.
What this means
A usage ceiling only works as an expansion lever if hitting it means the customer is already deep in the product's core value, not just generating support volume.
So what do I do about it?
| Symptom | Action | Effort |
|---|---|---|
| Expansion campaigns are spread evenly across all three limits | Reallocate next quarter's expansion campaign budget to the transfer-volume-limit trigger only | 30 min |
Final deliverable
A 1-page forecast memo: 3-year MRR projection at current vs. target NRR, the dollar gap it creates, and the single highest-converting usage-limit lever to prioritize next quarter.
See a reference example
Slack, self-serve expansion forecast (excerpt) Current NRR: 108% Target: 122% Year-3 gap at $5M starting MRR: $2.6M/mo Top lever: message-history limit, 44% upgrade rate within 30 days of hitting it Recommendation: shift Q3 expansion campaign budget to message-history-limit triggers
Success criteria
You're done when you can:
- Correctly compounds MRR across 3 years for both NRR scenarios
- Identifies the transfer-volume-limit lever as strongest, not the easiest-to-build one
- Ties the recommendation back to a real usage-based trigger, not a generic upsell email