MRR went up. Great. But why did it go up? Was it because the company signed 40 new logos, or because three enterprise accounts expanded while 35 SMB customers quietly churned? MRR movement analysis decomposes the top-line number into four components — new, expansion, contraction, and churn — and the ratio between those components tells you more about revenue health than the headline figure ever will. For anyone managing a portfolio, this decomposition is the difference between knowing a company's MRR and understanding it.
What MRR movement categories are and why the top-line number hides them
A single MRR figure is a snapshot. It tells you where the company stands on the last day of the month. It does not tell you how it got there, and it actively obscures the forces pulling in opposite directions underneath. A company reporting $500K MRR with zero net change looks stable. A company reporting $500K MRR that added $80K and lost $80K is on fire — the gross movement signals a churn problem that acquisition is temporarily masking.
The four movement categories capture every dollar that entered or left the recurring revenue base during a period. Together they form a complete accounting of MRR change — no dollar is unclassified, and none is double-counted.
New MRR
Revenue from customers who were not paying in the prior period and began a paid subscription. This includes first-time buyers and reactivations, depending on how the company defines customer identity. A trial conversion counts here. A customer who canceled six months ago and comes back counts here unless the company treats reactivation as its own bucket.
Expansion MRR
Additional revenue from existing customers whose MRR increased. Plan upgrades, seat additions, usage overages on metered components, and cross-sells into new product lines all land here. Expansion MRR is the clearest signal that a product is delivering enough value for customers to buy more of it.
Contraction MRR
Revenue lost from existing customers whose MRR decreased without canceling entirely. Downgrades, seat removals, and renegotiated contracts at lower rates all produce contraction. This category is often underreported because many billing systems net it against expansion, hiding a useful signal inside a blended number.
Churn MRR
Revenue lost from customers who canceled entirely. The subscription ended, the account closed, and the MRR it contributed dropped to zero. Churn MRR is the most visible outflow, but it is not always the largest — at companies with tiered pricing, contraction MRR frequently exceeds churn MRR by a meaningful margin.
| Category | Definition | Stripe Signal | Example |
|---|---|---|---|
| New MRR | First payment from a new customer | customer.subscription.created | Acme signs a $2K/mo plan |
| Expansion MRR | Increase from an existing customer | customer.subscription.updated (amount up) | Acme adds 10 seats at $50/seat |
| Contraction MRR | Decrease without full cancellation | customer.subscription.updated (amount down) | Acme drops from Growth to Starter plan |
| Churn MRR | Complete cancellation | customer.subscription.deleted | Acme cancels entirely — $0 MRR |
The movement ratios that reveal revenue health
Raw movement numbers are a start. Ratios built from those numbers are the diagnostic. Three ratios matter most, and each answers a different question about whether the revenue base is healthy, growing efficiently, or dependent on acquisition to survive.
Quick ratio
The quick ratio divides total MRR inflows (new + expansion) by total MRR outflows (contraction + churn). A ratio of 4.0 means the company adds $4 for every $1 it loses. The traditional benchmark is 4.0 or above, though that threshold has compressed at scale — companies above $2M MRR typically land between 1.5 and 2.5 even when healthy. A quick ratio above 4 means the business grows even if acquisition stalls, because expansion alone covers losses with room to spare.
The ratio's power is in its composition, not just its level. A quick ratio of 3.0 driven by new MRR with minimal expansion tells a different story than a 3.0 driven by expansion with moderate acquisition. The first company is growth-dependent. The second has a compounding revenue engine. Same ratio, opposite prognoses.
Monthly Recurring Revenue
The predictable revenue a SaaS company expects each month from active subscriptions.
Expansion efficiency
Expansion efficiency measures how much of the existing base expands each period, expressed as expansion MRR divided by beginning-of-period MRR. A company at $1M MRR that generates $30K in expansion has 3% expansion efficiency. At scale, even small differences in this ratio compound dramatically — a company with 4% expansion efficiency doubles its base revenue contribution in 18 months; a company at 1% takes six years.
Expansion efficiency also functions as a leading indicator for net revenue retention. If expansion efficiency consistently exceeds gross churn rate, NRR will be above 100% — the installed base grows without any new customers. This is the definition of a compounding SaaS business.
Net Revenue Retention
The percentage of recurring revenue retained from existing customers, including expansion and contraction.
New-to-churn ratio
The new-to-churn ratio isolates acquisition against loss by dividing new MRR by churn MRR, excluding expansion and contraction. This tells you whether the company's go-to-market is outpacing customer attrition at the unit level. A ratio below 1.0 means the company is losing more revenue from cancellations than it is replacing with new logos — and unless expansion is significant, MRR is shrinking.
This ratio is particularly useful in portfolio comparisons because it separates the acquisition engine from the product engine. Expansion and contraction reflect product value. New and churn reflect market fit and sales execution. Analyzing them separately tells you which engine needs work.
Deriving MRR movement from Stripe subscription events
The four movement categories map cleanly to Stripe's subscription lifecycle events, but the mapping requires more care than it appears. Stripe tracks subscriptions, not MRR categories — the classification logic sits on top.
Mapping Stripe events to categories
A customer.subscription.createdevent is new MRR, provided the customer had no active subscription in the prior period. If the customer had a different active subscription, the new one may be expansion (add-on product) or a plan swap that should decompose into contraction on the old plan and expansion on the new one. The event alone does not disambiguate; you need the customer's subscription history.
A customer.subscription.updated event with a higher plan.amount is expansion. The same event with a lower amount is contraction. But updates can also reflect billing interval changes (monthly to annual), coupon applications, or proration adjustments that are not true movement. Filtering these requires comparing the normalized monthly rate before and after the change, not the raw invoice amount.
Edge cases
Trial-to-paid conversions are the most common edge case. Stripe fires a customer.subscription.updated when the trial ends and billing begins, not a new createdevent. Treating this as expansion (because the subscription existed during the trial) understates new MRR and overstates expansion. The correct classification is new MRR — the customer was not paying before.
Mid-cycle upgrades with proration create another ambiguity. Stripe generates an invoice for the prorated difference immediately, but the MRR impact is the change in the monthly rate, not the prorated amount. A customer upgrading from $100/mo to $200/mo mid-month generates $50 in prorated charges on the current invoice, but $100/mo in expansion MRR going forward. Confusing the two inflates expansion in the upgrade month and deflates it afterward.
Pause-and-resume subscriptions also require consistent handling. When a customer pauses, Stripe can either cancel the subscription (creating churn MRR) or set it to a $0 amount (creating contraction MRR). When they resume, the reverse classification should match. If the pause was churn, the resume is new MRR. If the pause was contraction, the resume is expansion. Mixing the two directions inflates gross movement and distorts the quick ratio.
Portfolio movement analysis — comparing patterns across companies
Individual movement analysis is useful. Comparative movement analysis across a portfolio is transformative. When you normalize each company's movement categories as percentages of beginning MRR, you get a movement fingerprint — a four-number profile that characterizes how the company grows and loses revenue. That fingerprint is the fastest diagnostic for portfolio health because it compresses months of operational data into a shape you can compare at a glance.
A company with a fingerprint of 8% new, 4% expansion, 2% contraction, 1% churn looks fundamentally different from one at 3% new, 1% expansion, 4% contraction, 3% churn. Both might report the same net MRR growth — 9% versus –3% tells a clearer story than any blended metric. The first company is growing efficiently with a healthy base. The second is shrinking despite ongoing acquisition because the installed base is eroding through contraction and churn.
At the portfolio level, movement fingerprints let you sort companies into operational archetypes. Acquisition-led growers have high new MRR and low expansion. Product-led compounders have moderate new MRR and high expansion. Retention-challenged companies have high gross movement in both directions — they add and lose revenue in roughly equal measure, running hard to stay flat. Each archetype implies a different intervention. The acquisition-led grower needs pricing or product investment to unlock expansion. The retention-challenged company needs churn diagnosis before any growth initiative makes sense.
Comparing fingerprints across companies also reveals outliers that blended metrics miss. A company reporting 5% net MRR growth looks healthy until you see that its contraction rate is 6% — the highest in the portfolio — masked by 8% new MRR that the sales team is generating at unsustainable cost. The fingerprint surfaces this. A single MRR growth number does not.
Trend analysis adds another dimension. A company whose expansion efficiency declines for three consecutive months while its churn rate holds steady is losing pricing power or product engagement, even though the headline numbers may still look acceptable. Conversely, a company whose contraction rate drops while new MRR holds flat is actually improving — the installed base is becoming stickier, which compounds over time even without acquisition growth.
The portfolio operator who tracks movement fingerprints across 10 or 20 companies can answer questions that aggregate metrics cannot. Which company has the most efficient growth engine? Where is expansion MRR concentrating — in a few large accounts or broadly across the base? Which companies have improving retention trends that will compound into NRR gains over the next two quarters? These are the questions that drive allocation decisions, board conversations, and operational interventions. And they all start with decomposing a single number into its four components.