Contraction: the revenue lost from customers who stay and pay less

Contraction is the recurring revenue you lose when a customer stays on your base but reduces how much they pay. They did not cancel the subscription: they moved from an expensive plan to a cheaper one, cut seats or removed an add-on. The result is a smaller MRR from the same active customer.

That is why contraction often goes unnoticed. The customer count does not change, the logo is still on the list, but recurring revenue shrank. Anyone watching only the number of active accounts misses that loss; anyone tracking MRR per account catches it.

The most common mix-up is treating contraction as churn. It is not. In churn, the customer leaves and you lose 100% of their revenue. In contraction, the customer stays and you lose only part of it. One becomes zero; the other becomes less.

The distinction matters because the responses differ. Against churn you fight not to lose the whole account; against contraction you fight to preserve value inside an account that wants to stay. Lumping both into a single bucket of lost revenue hides which problem you actually have.

• Churn: the customer cancels, their revenue goes to zero.
• Contraction: the customer stays, but pays less than before.

Contraction almost always comes from one of these. Each tells a different story about the value the customer sees in the product.

• Plan downgrade: the customer swaps a pricier plan for a cheaper one, usually over cost or because they use less than expected.
• Fewer seats or licenses: the customer team shrank, or they realized they were paying for users who did not log in.
• Removed add-on or module: an optional feature stopped being worth the price for that customer.
• Lower usage on usage-based pricing: when part of the bill is volume-based, using less means paying less.

Reading the cause is the first step. A downgrade over price calls for a commercial conversation; a downgrade over low usage calls for adoption and customer success work.

Every month, MRR moves in blocks. MRR movements add what came in (new customers and expansion) and subtract what went out (churn and contraction). Contraction is one of the two negative forces in that math, alongside churn.

Separating contraction inside that waterfall is what lets you diagnose the month. A flat MRR can hide strong expansion cancelled out by equally strong contraction. Without the contraction block isolated, you cheer or panic over the net number without understanding what made it up.

Net Revenue Retention (NRR) measures how much of a cohort revenue you keep after adding expansion and subtracting churn and contraction. Contraction enters as a direct subtraction: every dollar of downgrade lowers NRR even if no customer left.

That is why two companies with the same churn can have very different NRR: the one with more contraction retains less revenue. The private SaaS survey by KeyBanc Capital Markets shows median net revenue retention near 100%, and keeping contraction in check is part of what separates those above that mark from those below it.

To reduce contraction, you first have to see it. Isolate the contraction block in your MRR, tag every event with a reason (price, usage, team size) and look at where it concentrates, by plan, by segment and by customer tier.

• Act before the downgrade: usage drops often precede a request for a smaller plan; a usage alert buys time to act.
• Package value in the right plan: contraction over price is sometimes a sign of a plan poorly designed for that profile.
• Treat idle seats with adoption: cutting unused licenses becomes contraction; helping the customer use them prevents the loss.
• Offer alternatives to a full cancel: a guided downgrade beats a churn, but the goal is to preserve as much revenue as possible, not to push customers down.

Contraction never reaches zero, and forcing it to can make churn worse. The goal is to keep it small and explained, so that expansion outpaces it with room to spare.