Subscription vs. Usage-Based Billing: Which Revenue Model Is Right for Your B2B SaaS Company
Subscription billing and usage-based billing are not just different pricing structures — they produce fundamentally different businesses. Subscription gives you predictable revenue, upfront cash, and a simple story. Usage-based gives you a revenue model that grows with your customers automatically, but introduces forecasting complexity, cash flow timing issues, and operational overhead that most founders don't see coming. For B2B SaaS companies that process payments, enable transactions, and orchestrate workflows for their clients, the choice between these two models — or how to combine them — is one of the most consequential decisions you'll make.
How subscription billing works
Customers commit to a fixed price — monthly or annually — regardless of how much they use your product. Everyone who signed the contract pays the same amount until renewal.
The advantages are real:
- Predictable revenue: you know exactly what's coming in next month
- Cash collected upfront on annual plans, which funds growth before you earn it
- Simple to model, simple to sell, and straightforward to recognize under ASC 606
- Renewal conversations are calendar-driven, not consumption-driven
The structural problem: subscription billing has no expansion motion tied to product usage. Your revenue grows only when you add new customers or push through a price increase at renewal. A customer who triples their usage of your platform pays you exactly the same as a customer who barely logs in. That's the ceiling.
Kyle Poyar's research across 240 software companies found that flat-fee subscriptions have dropped from 29% to 22% of the market in just 12 months — not because subscriptions are dying, but because founders are realizing the ceiling gets lower as products scale.
How usage-based billing works
Customers pay based on what they consume — transactions processed, API calls, tokens used, records enriched, emails sent. The bill varies month to month based on actual product usage. Schematic's 2026 guide to usage-based billing describes it well: when implemented correctly, usage-based billing becomes a revenue engine; when poorly executed, it creates customer confusion and unpredictable results.
The advantages compound over time:
- NRR above 100% without a dedicated upsell motion — revenue grows automatically as customers grow
- Pricing aligned with value delivered, which makes renewals easier and churn less binary
- A natural land-and-expand motion: customers start small, prove value, then scale
The problems are real too, and they hit you operationally before they show up in your financials:
- Revenue is unpredictable month-to-month, especially before you have 12+ months of consumption data by cohort
- Cash is collected in arrears — you invoice after usage occurs, sometimes 30–60 days after the period ends
- Sales comp becomes contested — do you pay AEs on minimum commit or on actual consumption? Most comp plans aren't built for this
- Customer success changes — instead of managing renewals, you're managing consumption patterns, overage conversations, and usage coaching
For AI-native products specifically, the problem compounds: Paid.ai's analysis of usage-based pricing for AI agents shows that AI workloads can create 100x cost variance between customers, making standard usage-based pricing models negative-margin for heavy users without careful floor pricing and commit structures.
Scott Woody, co-founder of Metronome (acquired by Stripe), identified billing as the single biggest bottleneck to growth at Dropbox — the place where new pricing ideas went to die because the billing system couldn't keep up. He built Metronome specifically because every company on earth wants to grow faster, and the billing system was slowing them down.
Subscription vs. usage-based billing: side by side
| Subscription | Usage-based | |
|---|---|---|
| Revenue predictability | High | Low–Medium |
| Cash collection timing | Upfront (annual) / monthly | In arrears |
| NRR potential | Capped at ~100% | 110–150%+ |
| Expansion motion | Renewal / upsell | Automatic with usage |
| Forecasting complexity | Low | High |
| Sales comp complexity | Low | High |
| Customer success motion | Renewal-focused | Consumption-focused |
| Best for | Predictable workflows | Transaction / AI / API products |
What actually changes when you switch
This is the part founders underestimate. Switching from subscription to usage-based billing is not a pricing page update — it's a systems change that runs through your entire operation.
Pricing and packaging: you need to define a value metric, set a consumption floor (if any), and decide how to handle overages. None of this is trivial without usage data.
Sales: the pitch changes from "here's what you pay" to "here's what you pay based on how much you use." Enterprise buyers resist this because their finance teams need fixed budgets. Minimum commits and annual caps are what make usage-based pricing palatable to procurement.
Cash flow: if you've been running on annual upfront subscriptions and you switch to monthly usage billing in arrears, your cash position changes immediately. A company doing $1M ARR on annual subscriptions collecting cash in January looks very different in March than the same company billing monthly for December usage. The Operators Guild workshop on usage-based pricing identified consumption and cash collection as two separate problems that need to be designed independently — conflating them is a common and expensive mistake. RightRev's breakdown of usage-based vs. subscription revenue recognition covers how this cash flow difference also flows into your ASC 606 treatment — usage revenue recognized at consumption, not at invoice date, which changes your close process materially.
Customer retention dynamics: subscription churn is binary — a customer either renews or they don't. Usage-based churn is gradual — you can see it coming in the consumption data months before a customer churns. That's actually an advantage, but only if you're watching the right signals. James Wood, former head of pricing at Segment, notes that traditional SaaS metrics like logo churn become misleading under consumption models — you need cohort-level usage data to understand what's actually happening.
Growth and profitability: usage-based companies typically see lower gross margins than pure subscription businesses in the early years — the cost of serving high-usage customers can be significant, especially in AI or transaction-heavy products. But the long-term economics are better: NRR above 100% compounds in a way that flat subscription growth cannot.
Which model is right for your company
Choose subscription if:
- Your product's value is consistent regardless of usage volume
- Your buyers are budget-constrained and need fixed costs
- You're early-stage and don't have enough usage data to price dynamically
- Your COGS doesn't scale materially with customer usage
Choose usage-based if:
- Your product processes transactions, generates outputs, or performs work whose volume varies
- Your best customers use dramatically more than your average customers
- You want NRR above 100% without building a dedicated upsell team
- You can instrument usage accurately in real time
In practice, choose hybrid: the Operators Guild workshop on usage-based pricing found that pure usage models "often evolve into hybrids to balance flexibility with revenue and cash flow stability." A minimum annual commit with usage on top gives you the cash predictability of a subscription and the expansion upside of usage billing. Shopify is the most cited example: fixed subscription plans plus transaction revenue that scales with merchant GMV. Less than 30% of Shopify's revenue is subscription; the rest grows automatically.
The sequencing question: when to make the switch
Don't switch cold. The companies that get this right follow a three-phase approach outlined by Steve Love and James Wood at the Operators Guild:
- Assess: instrument your usage data, identify your value metric, and model what usage-based billing would have generated over the last 12 months on your existing customer base
- Pilot: launch usage-based billing with new customers only, while keeping existing customers on their current plans
- Migrate: once you have 6–12 months of usage data and a clear expansion motion, offer existing customers a migration path — never a forced switch
Grandfather existing customers. The Forbes / Metronome playbook on shifting to usage-based pricing recommends a 90-day grace period minimum before applying new billing structures to existing accounts.
Sources: Steve Love & James Wood, Operators Guild — "Making Usage-Based Revenue Work" (March 2026); Operators Guild — "Usage-Based Pricing Workshop" (January 2026); Operators Guild — "Mastering Usage-Based Revenue in an ARR Construct" (November 2023); Scott Woody, Metronome CEO, Grace Gong interview (January 2026); Scott Woody, The Room Podcast (April 2026); Riya Grover, Sequence CEO, Turpentine Finance with Sasha Orloff (July 2024); Kyle Poyar, Growth Unhinged — "The State of B2B Monetization in 2025" (June 2025)