On this page
"Net expansion above 120% is good" is the standard advice, and it is wrong for most companies. Expansion rates vary dramatically by vertical — infrastructure tools expand 2–3x faster than horizontal SaaS, and vertical SaaS often caps below 110% regardless of execution quality. The company with 112% NRR in legal tech is outperforming its vertical by the same margin as the DevTools company posting 135%. Applying a single expansion benchmark across verticals produces the same error as applying a single churn benchmark across ACV tiers: it systematically overvalues companies in expansion-friendly categories and undervalues those in expansion-constrained ones.
40+ pts
NRR spread across verticals
125-145%
Median NRR for infrastructure/DevOps
105-115%
Median NRR for vertical SaaS
Why expansion varies by vertical
Expansion revenue is driven by three structural factors: the natural growth of the customer's usage, the width of the product's pricing surface area, and the depth of the customer's organization that the product can penetrate. These factors are not evenly distributed across verticals. They are determined by the vertical's characteristics — and they set hard ceilings on how much expansion is structurally possible.
Usage elasticity:Infrastructure and DevTools products sit on top of workloads that scale with the customer's business. When a customer grows, their API calls, data volume, and compute usage grow with them — often faster than revenue. This creates natural expansion without any deliberate purchase decision. Vertical SaaS products (legal practice management, dental office software) serve workloads that are bounded by the customer's physical operations. A dental practice does not double its patient management software usage when revenue grows 20%.
Pricing surface area:Multi-product platforms (data infrastructure, security suites) can expand through cross-sell — the customer buys additional products from the same vendor. A company selling a single-purpose vertical tool has limited cross-sell opportunity. The pricing model determines whether there is even a mechanism for expansion beyond adding seats.
Organizational depth:Horizontal tools (collaboration, analytics, communication) spread across departments. One team adopts the tool, then another, then another. Each department is a new expansion event. Vertical SaaS typically serves a single function within the organization — all users are in the same department from day one, and the seat count is bounded by the team size.
Infrastructure and DevTools benchmarks
Infrastructure and developer tools consistently produce the highest expansion rates in SaaS. The combination of usage-based pricing and workload-driven growth creates a compounding expansion engine that runs automatically as the customer's business scales. The public-company exemplars — Snowflake, Datadog, Twilio, Cloudflare — have posted NRR above 125% for years.
| Subcategory | Median NRR | Top Quartile NRR | Median Gross Expansion | Primary Expansion Driver |
|---|---|---|---|---|
| Cloud infrastructure | 130–145% | 150%+ | 40–55% | Usage growth (compute, storage) |
| Observability / monitoring | 125–140% | 145%+ | 35–50% | Data volume + product cross-sell |
| Developer platforms / APIs | 125–135% | 140%+ | 30–45% | API call volume |
| Data warehousing / ETL | 120–135% | 140%+ | 30–45% | Query volume + data ingestion |
| CI/CD / DevOps tooling | 115–130% | 135%+ | 25–40% | Seat growth + pipeline volume |
The variance within infrastructure is driven by pricing model purity. Pure consumption-based products (pay per query, pay per GB) produce the highest NRR because expansion is automatic — no sales touch required. Seat-based DevTools products (IDEs, project management) see expansion rates 10–15 points lower because expansion requires a deliberate purchase event: someone has to add a seat.
The risk with infrastructure-level NRR is volatility. A customer that doubled their data volume last year can optimize their queries and cut usage by 30% next year. NRR above 140% at the company level often includes large accounts whose consumption could swing dramatically. Investors evaluating infrastructure companies should look at NRR durability — not just the peak number, but how many consecutive quarters it has held above a threshold.
Net MRR Retention
Revenue retained from existing customers including expansion, contraction, and churn.
Horizontal B2B SaaS benchmarks
Horizontal B2B SaaS — products serving a broad function across industries (CRM, marketing automation, analytics, collaboration) — sits in the middle of the expansion spectrum. Expansion comes primarily from seat growth and tier upgrades rather than consumption, which makes it more predictable but structurally capped.
| Subcategory | Median NRR | Top Quartile NRR | Median Gross Expansion | Primary Expansion Driver |
|---|---|---|---|---|
| Security / compliance | 120–140% | 145%+ | 25–45% | Module cross-sell + seat growth |
| Data / analytics platforms | 115–135% | 140%+ | 20–40% | Data volume + user seats |
| Sales / CRM | 110–120% | 125%+ | 15–25% | Seat growth + tier upgrade |
| Marketing automation | 108–118% | 125%+ | 15–25% | Contact volume + add-ons |
| Collaboration / productivity | 105–115% | 120%+ | 10–20% | Seat growth |
| HR / people ops | 110–125% | 130%+ | 15–30% | Headcount growth + modules |
| Finance / billing tools | 108–120% | 125%+ | 12–25% | Transaction volume + features |
Security stands out as the horizontal category with infrastructure-like expansion dynamics. The reason is structural: security products expand with the attack surface, which grows with the organization's infrastructure, headcount, and data volume. A company that adds cloud workloads needs more security monitoring. One that hires employees needs more identity management. The expansion is driven by the customer's growth, not by a sales team pushing upgrades — which is the same dynamic that drives infrastructure expansion.
Collaboration and productivity tools represent the floor for horizontal B2B. Expansion is almost entirely seat-driven, and seat growth tracks the customer's hiring rate. In a hiring slowdown, expansion stalls. In a downturn, contraction can exceed expansion as customers reduce headcount and remove seats. The post-2022 NRR compression hit collaboration tools hardest for exactly this reason.
Vertical and industry SaaS benchmarks
Vertical SaaS products serve a specific industry — healthcare, legal, real estate, construction, restaurants. Their expansion dynamics are fundamentally different from horizontal and infrastructure products because the customer's operations create a natural ceiling on usage growth. A law firm with 12 attorneys does not need 12x more case management capacity when it grows to 15 attorneys. The relationship between customer growth and software usage is sublinear.
| Vertical | Median NRR | Top Quartile NRR | Median Gross Expansion | Primary Expansion Driver |
|---|---|---|---|---|
| Healthcare SaaS | 108–118% | 122%+ | 12–22% | Location expansion + modules |
| Legal tech | 105–115% | 120%+ | 10–18% | Seat growth + practice area modules |
| Real estate tech | 105–112% | 118%+ | 8–16% | Property count + add-on services |
| Construction tech | 106–114% | 120%+ | 10–18% | Project volume + user seats |
| Restaurant / hospitality | 102–110% | 115%+ | 5–14% | Location expansion |
| Fitness / wellness | 100–108% | 112%+ | 4–12% | Location + member volume |
The striking feature of vertical SaaS expansion is how close to 100% the bottom quartile sits. A vertical SaaS company with 102% NRR is not failing — it's operating at the structural floor of its category. The product retains well (otherwise NRR would be below 100%), but the expansion surface area is limited by the customer's operational scale.
The vertical SaaS companies that achieve top-quartile NRR share a common strategy: multi-location expansion. A dental practice management tool that serves a single-location practice has limited expansion potential. The same tool serving a dental service organization (DSO) with 30 locations has a fundamentally different expansion curve. The product is the same. The customer profile changes the expansion physics entirely.
What drives expansion rate differences
The vertical benchmarks above are the output. Understanding the input — what structural factors create different expansion ceilings — helps investors and operators diagnose whether a company's expansion rate reflects good execution or favorable category dynamics.
| Factor | High Expansion Impact | Low Expansion Impact |
|---|---|---|
| Pricing model | Usage-based / consumption | Flat-rate / per-seat |
| Usage elasticity | Usage scales with customer growth | Usage capped by operations |
| Product breadth | Multi-product platform | Single-purpose tool |
| Organizational spread | Cross-department adoption | Single-department use |
| Customer type | Enterprise / multi-location | SMB / single-location |
| Switching costs | Deep integration / high | Light integration / low |
A company can influence some of these factors and not others. Pricing model is a strategic choice — shifting from flat-rate to usage-based pricing can unlock 10–20 points of NRR by aligning revenue with customer growth. Product breadth is a roadmap decision — adding modules or adjacent products creates cross-sell surface area. But usage elasticity and organizational spread are determined by the market, not the company. A vertical SaaS product serving single- location small businesses cannot create the expansion dynamics of a cloud infrastructure platform through better execution.
This distinction matters for setting targets. A vertical SaaS founder aiming for 130% NRR is chasing a structurally unavailable number and may distort their pricing or sales practices trying to reach it. A realistic target for that company is 110–118%, achieved through multi-location penetration and module expansion — the two levers actually available in that market.
Expansion vs new logo acquisition efficiency
Expansion revenue is not just cheaper than new logo revenue — the cost differential varies by vertical in ways that change the optimal growth mix. Understanding the expansion-to-acquisition cost ratio by vertical tells you how much growth should come from each source.
| Vertical Category | Expansion CAC as % of New Logo CAC | Optimal Expansion Mix (% of net new ARR) | Typical Payback Difference |
|---|---|---|---|
| Infrastructure / DevOps | 5–15% | 50–70% | Expansion: < 1 mo vs New: 12–18 mo |
| Security | 10–20% | 40–60% | Expansion: 1–2 mo vs New: 14–20 mo |
| Horizontal B2B SaaS | 15–30% | 30–50% | Expansion: 2–4 mo vs New: 12–18 mo |
| Vertical SaaS | 20–40% | 20–35% | Expansion: 3–6 mo vs New: 10–14 mo |
| SMB horizontal | 25–50% | 15–25% | Expansion: 4–8 mo vs New: 8–12 mo |
Infrastructure companies derive 50–70% of net new ARR from expansion because the economics are overwhelming: expansion CAC is 5–15% of new logo CAC, and payback is under a month. For these companies, the most capital-efficient growth strategy is to land accounts and let usage grow. The new logo function exists to seed the expansion engine.
Vertical SaaS companies operate under different economics. Expansion CAC is still cheaper than new logo acquisition, but the gap is smaller (20–40% vs 5–15%), and the expansion opportunity per account is structurally limited. The optimal growth mix tilts more toward new logo acquisition — 65–80% of net new ARR from new logos, 20–35% from expansion. This is not a failure of the expansion motion. It reflects the market structure.
Expansion MRR
Additional MRR from existing customers through upsells, cross-sells, and usage growth.
What these benchmarks mean for portfolio managers
Vertical-specific expansion benchmarks change three portfolio management decisions: target setting, performance evaluation, and capital allocation.
Target setting:A portfolio company in legal tech should not carry a 125% NRR target. The structural ceiling for the vertical is 115–120% at top quartile. Setting a target above that ceiling guarantees a miss and forces the company into counterproductive behavior — aggressive price increases that accelerate churn, or usage-based pricing in a context where usage does not scale with customer value. The right target is 110–115%, which represents genuine outperformance for the vertical.
Performance evaluation:Comparing a portfolio company's NRR against the wrong vertical benchmark produces misleading conclusions. An HR tech company at 118% NRR and a DevTools company at 118% NRR are performing at very different percentiles relative to their verticals. The HR company is top quartile. The DevTools company is below median. Evaluating both as "118% is good" misses the diagnostic entirely.
Capital allocation: Expansion-friendly verticals reward investment in customer success and product-led growth because each dollar of retention investment produces more expansion revenue. Expansion-constrained verticals reward investment in new logo acquisition because that is where the growth comes from. Allocating the same growth budget split across verticals ignores the structural differences in where returns come from.
Tracking vertical-adjusted expansion with North Metric
Expansion revenue measurement from billing data requires classifying every subscription change correctly: upgrades, downgrades, add-ons, usage overages, seat additions, and seat removals. Each of these is an expansion or contraction event, and misclassifying even one category distorts the NRR calculation. Most spreadsheet implementations mishandle at least two of these categories — typically mid-cycle upgrades and annual-to-monthly switches — producing NRR numbers that are internally inconsistent.
North Metric derives expansion and contraction MRR directly from Stripe subscription events, classifying each change at the invoice level. The expansion rate, contraction rate, and net expansion are decomposed in the default dashboard view — not hidden behind an advanced settings panel — because the decomposition is the diagnostic. A company seeing 20% gross expansion offset by 8% contraction needs a different intervention than one seeing 12% expansion with 0% contraction, even though both produce 112% NRR.
For portfolio investors, the multi-company view applies the same classification logic across every connected account. Expansion rates are comparable because they are computed identically. The vertical context — whether 112% is top quartile or bottom quartile for the company's market — is the layer that turns a consistent metric into an actionable signal.