SaaS Expansion Revenue Is 40% of New ARR. Model It.
Across the SaaS market, expansion revenue now contributes roughly 40% of total new ARR. For companies above $50M in revenue, that number climbs past 60%. The economics are hard to argue with: expanding an existing customer costs a fraction of acquiring a new one, often 5x to 10x less in fully loaded sales and marketing spend.
The gap I see most often is not in the model itself. It is in the org chart. Scaling companies pour energy into new logo acquisition. They build pipeline reviews, hire reps, and measure everything from lead-to-close. Once a customer is onboarded, the intensity drops. There may not be a customer experience team with the same rigor, the same targets, or the same accountability for expanding wallet share and increasing switching costs. The expansion line in the forecast reflects that: it is a single flat growth rate because nobody owns the outcome with the same discipline that sales owns new bookings.
When expansion is modeled, the question is usually "what is the overall growth rate?" not "where is it coming from?" Which cross-sell or upsell products are working? Is growth usage-based, or are customers buying incremental features? That depends entirely on the pricing architecture, and most models do not break it apart.
Most finance teams I work with can tell you their new logo bookings pipeline down to the rep, the stage, and the expected close date. Ask them how they forecast expansion, and you get a flat percentage applied to the existing base. That approach was adequate when expansion was 15% of new ARR. At 40%, it is the single largest gap in most SaaS revenue models.
Why Expansion Deserves Its Own Model
The core problem is that expansion and new logo acquisition follow completely different dynamics. New logo revenue depends on pipeline generation, sales cycle length, and win rates. Expansion revenue depends on product adoption, customer health, and pricing architecture. Treating them as a single growth metric hides the real story.
Consider two SaaS companies, both growing ARR at 30% year over year. Company A gets 70% of new ARR from new logos and 30% from expansion. Company B gets 50% from each. On a growth slide, they look identical. In the model, they are fundamentally different businesses. Company B has structurally lower customer acquisition costs, a stickier base, and (according to current valuation data) a 30% to 50% higher valuation multiple. A company with 120%+ NRR commands 10x to 12x ARR, while a company at 100% NRR trades at 6x to 8x. That gap is almost entirely an expansion story.
The FP&A function is the only team in the company that can see this clearly, because it requires connecting product usage data, customer success outcomes, and financial results in a single view. Sales knows their pipeline. Customer success knows their health scores. Finance is the one that ties it all to revenue and margin.
Three Metrics That Deserve Their Own Forecasts
1. NRR by Cohort
The headline NRR number is useful for a board slide. It is not useful for planning. A company with 108% blended NRR might have 125% enterprise NRR, 105% mid-market, and 88% SMB. That tells a completely different story about where growth is coming from and where churn is eating it.
The 2026 benchmarks make this concrete. Median NRR for enterprise SaaS (ACV above $100K) is 118%. Mid-market ($25K to $100K ACV) sits at 108%. SMB (below $25K) is 97%, which means the average SMB cohort is contracting. If your model assumes a single NRR rate across all segments, your forecast is wrong by definition.
Build a cohort model that tracks NRR by customer segment, sign-up quarter, and product tier. The sign-up quarter matters because expansion behavior changes as customers mature. A cohort that signed six months ago behaves differently than one that signed two years ago. Most companies find that expansion rates accelerate between months 6 and 18, then plateau. Knowing that curve shapes your revenue forecast for the next four quarters.
What I have seen across clients is that breaking cohorts by lead source and by customer age exposes patterns the blended number buries completely. A cohort acquired through a channel partner behaves differently than one from inbound marketing. A two-year-old customer expands at a different rate than a six-month-old one. These are not edge cases. They are the core behaviors that determine whether your expansion forecast is credible or aspirational. The blended NRR number hides all of it.
2. Expansion Rate by Product Line
Not all expansion is created equal. Seat expansion, usage-based growth, and module upsells each have different predictability, margin profiles, and sensitivity to economic conditions.
Seat expansion is the most predictable. It correlates directly with the customer's headcount growth and follows a relatively stable curve.
Usage-based growth is the most volatile. It can spike 30% in a quarter or drop 20% if a customer's business slows down. Module upsells are the hardest to forecast because they depend on sales execution and product readiness, but they carry the highest incremental margin. The pricing architecture matters here: whether expansion shows up as incremental feature purchases, consumption overages, or seat additions shapes how you forecast it and who in the org is accountable for driving it.
Your model should break expansion into these categories and forecast each one separately. The inputs are different for each: headcount data for seats, consumption trends for usage, pipeline data for module upsells. When the board asks "what happens if expansion slows by 5 points?" you need to know which type of expansion is at risk and what drives it.
3. Time-to-First-Expansion
This is the metric most SaaS finance teams do not track at all, and it is one of the most predictive. Time-to-first-expansion measures how many months pass between a customer's initial contract and their first upsell or expansion event.
It matters for two reasons. First, it is a leading indicator of NRR. Customers who expand within the first 12 months retain at dramatically higher rates than those who never expand. If your median time-to-first-expansion is 18 months, a large portion of your base will churn before they ever reach the expansion window. Second, it tells you whether your onboarding and customer success motion is converting adoption into revenue. A long time-to-first-expansion usually means the customer is underusing the product, not that they do not need more of it.
Track this metric by segment and product. Compare it against your average contract length. If your average contract is 12 months and your median time-to-first-expansion is 14 months, you have a structural problem: most customers will hit their first renewal decision before they have experienced expansion value. That is a churn risk hiding in your retention numbers.
Putting It into the Operating Model
These three metrics connect directly to the revenue forecast, the hiring plan, and the capital allocation discussion.
If NRR by cohort shows your enterprise segment expanding at 125% while SMB contracts at 88%, your go-to-market investment should shift upmarket. That is not a sales strategy decision made in isolation. It is a financial planning decision grounded in unit economics. The same logic applies to expansion rate by product line. If usage-based growth is volatile but seat expansion is steady, your forecast confidence interval tightens when you weight toward the predictable component.
The first question I ask is: are you tracking this at all? Not in a board slide. In the operating model. Most companies are not, or they are tracking a blended number that obscures everything useful. If the analysis does not exist, we build it the right way from the start: cohort analysis by customer size, by product usage, and by lead source. The goal is to identify the unique behaviors that let you capitalize on expansion opportunities and spot weaknesses before they show up in churn. If the analysis already exists, we expand it. Make it more granular, more actionable, and tied directly to the forecast. Either way, the starting point is the same: get the foundation in place, then make it more insightful over time.
The time-to-first-expansion metric feeds directly into customer success capacity planning. If you know that customers who expand within 9 months retain at 95%+ and those who don't expand by month 12 churn at 3x the rate, you can calculate exactly how many CSMs you need to drive expansion within the critical window. That turns a qualitative "invest more in customer success" argument into a quantifiable ROI case.
For companies building or rebuilding their FP&A operating model, expansion revenue modeling is one of the highest-impact additions. It changes the quality of the revenue forecast, the precision of the hiring plan, and the defensibility of the growth story in board meetings and fundraising conversations. If your current planning process treats expansion as a flat growth rate, that is the first assumption worth revisiting.
