SaaS Pricing Models: Tiered, Usage-based, Flat

• The Three Dominant Models
• Side-by-Side Comparison
• When Tiered Pricing Wins
  • Designing tier breakpoints
  • The annual vs monthly discount
• When Usage-Based Pricing Wins
  • Hybrid: Platform Fee + Usage
  • Implementing usage pricing
• When Flat-Rate Pricing Wins
• Pricing and LTV Math
• Price Discrimination and the Good-Better-Best Pattern
• When to Change Pricing
• How to Choose
• Next Steps

The pricing model you choose shapes everything: gross margin, expansion ceiling, sales cycle length, and the kind of customer you can serve profitably. There is no universally best model — Slack thrives on tiered, Twilio on usage-based, and Basecamp on flat — but each fits a specific product economics shape. This guide walks through the three dominant SaaS pricing models, when each one wins, and how to model the financial impact before you commit to a change.

Pricing is also the highest-leverage decision a SaaS company makes. A 10% lift on a $20M ARR base is $2M in pure margin. The same percentage lift in growth rate or churn takes a year or more of operational effort. Yet most SaaS companies revisit pricing every 24-36 months at best, leaving substantial value on the table between reviews.

The Three Dominant Models

Every SaaS pricing scheme is a variation on three primitives:

• Tiered (per-seat or feature-gated): customers pay a fixed amount per user or per package. Examples: Slack, HubSpot, Asana.
• Usage-based (consumption): customers pay for what they consume — API calls, GB stored, transactions processed. Examples: AWS, Twilio, Snowflake, Stripe.
• Flat-rate: one price for everyone, unlimited usage. Examples: Basecamp, ConvertKit (originally), Plausible Analytics.

Many modern companies blend two: a "platform fee + usage overage" hybrid is now the dominant model among infrastructure SaaS over $100M ARR. The hybrid captures the predictability of tiered pricing with the upside of usage pricing — the best of both worlds when implemented carefully.

Side-by-Side Comparison

When Tiered Pricing Wins

Tiered, per-seat pricing dominates B2B SaaS for good reason: buyers understand it, finance teams can budget for it, and expansion mechanically tracks customer growth. It is the default for any product where the unit of value is "a person using it" — CRMs, project management, design tools, sales engagement.

The risk with tiered pricing is the seat-cap problem. Once every relevant person at the customer has a seat, expansion stops. This is why mature tiered-pricing companies layer on a second axis: usage-based add-ons, premium features per-seat, or an enterprise tier with platform fees. Slack, for example, started with pure per-seat pricing and now offers a hybrid that includes message history limits, API call limits, and storage limits — all of which become expansion triggers as customers grow.

Designing tier breakpoints

Most pricing pages have three tiers. The middle tier should be the anchor — what you actually want most customers to buy. A common shape:

1. Starter ($10-25/seat): limited integrations, basic features, used to lower the trial barrier.
2. Pro ($30-75/seat): the workhorse. 60-70% of revenue should come from here.
3. Enterprise (custom): SSO, audit logs, dedicated support, annual contracts. Used to expand ACV.

Use the SaaS Pricing Calculator to model how a 20% price lift on the middle tier flows through to ARR. Most pricing changes deliver more revenue than expected because the anchor effect — the middle tier becoming relatively more attractive compared to a more expensive Pro tier — drives mix shift, not just average price increase.

The annual vs monthly discount

Most tiered SaaS offers a 15-20% discount for annual billing. This is more strategic than it looks: annual contracts dramatically reduce involuntary churn (fewer credit card failures, fewer monthly cancellation windows) and front-load cash collection. The right discount is the one that produces 70%+ of new customers on annual contracts. Below 50% annual adoption, the discount is too small; above 90%, you may be giving away margin unnecessarily.

When Usage-Based Pricing Wins

Usage-based pricing aligns revenue with the value customers actually get. It removes the buyer's risk ("I'll only pay for what I use") and lets the vendor capture upside from power users automatically. Snowflake, Twilio, and Datadog all hit 130%+ NRR partly because of this model. The big shift since 2020 has been adoption beyond pure infrastructure — modern AI products, observability tools, and data platforms increasingly default to usage pricing.

The trade-offs are real:

• Predictability suffers. Both you and your customer have less forecastable bills.
• Sales-led motion is harder. Pure usage pricing is awkward for enterprise procurement.
• Bad months hurt twice. When customer usage drops (seasonality, recession), revenue drops with it.

Pure usage-based pricing only works if your costs scale with usage too. Otherwise you pass volatility through to your P&L while customers get the smooth experience.

Hybrid: Platform Fee + Usage

The modern compromise is a minimum platform fee (locks in baseline revenue, makes ACV more predictable) plus usage overage above a generous included quota. This shape:

• Gives sales a predictable contract value to anchor on
• Gives customers a "free up to X" psychological win
• Captures expansion automatically as usage grows
• Reduces churn risk because customers paid the platform fee upfront

Datadog's pricing is a textbook example: hosts and containers have per-unit pricing, but minimum commits and bulk discounts give the contract enterprise predictability while still capturing usage growth automatically.

Implementing usage pricing

Usage pricing is harder to implement than to design. You need:

• Accurate, real-time metering infrastructure (custom-built or via services like Lago or Orb)
• A clear billing presentation customers can understand
• Detailed usage analytics customers can self-serve
• A predictive billing email or notification before bills arrive
• Spending limits and alerts that customers control

Plan for 6-9 months from decision to first dollar of usage revenue in a non-trivial product. Many SaaS companies underestimate this and ship half-built metering, then face customer complaints about unexpected bills.

When Flat-Rate Pricing Wins

Flat-rate pricing is the simplest model and arguably the most underrated. It works exceptionally well when:

• Your target customer is solo or very small team (1-5 people)
• Your COGS per customer is low and roughly fixed
• You compete on simplicity rather than feature depth
• Your buyer is the user (no procurement layer)

The classic examples — Basecamp, ConvertKit, Plausible — all serve markets where pricing complexity is itself a competitive disadvantage. Flat-rate is also the easiest model to A/B test, since there is only one price to move.

The downside is the expansion ceiling. Without seats or usage to grow, your NRR is capped near 100%. Most flat-rate SaaS companies eventually add a "team" or "agency" tier to break this ceiling, at which point they have effectively converted to tiered pricing. Even Basecamp, the most committed flat-rate company, eventually introduced a "Basecamp for Teams" tier.

If you choose flat-rate, accept the trade-off explicitly: you are choosing simplicity and conversion velocity over expansion revenue. The math only works if your customer acquisition cost is correspondingly low.

Pricing and LTV Math

The pricing model directly shapes Lifetime Value. A back-of-envelope comparison for a $100/month customer:

• Tiered (modest expansion): LTV ≈ $100 × 36 months × 1.1 NRR multiplier = $3,960
• Usage-based (strong expansion): LTV ≈ $100 × 36 × 1.3 NRR multiplier = $4,680
• Flat-rate (no expansion): LTV ≈ $100 × 30 months = $3,000

Run your own LTV estimates with the LTV Calculator before locking in a model — the difference between 1.1x and 1.3x expansion compounds into the largest line item in your three-year plan.

Price Discrimination and the Good-Better-Best Pattern

The most reliable pricing pattern in SaaS is good-better-best (sometimes called fence-pricing). Three tiers, each with progressively more features, designed so that different customer segments self-select into the tier that fits them. This is price discrimination done legally and elegantly — you charge more to customers who derive more value, without forcing a different price across the same product.

Research from the pricing world consistently shows that adding a third tier (especially an artificially-expensive top tier) lifts average revenue per customer by 10-25% via the anchor effect, even if very few customers actually buy the top tier. The top tier exists primarily to make the middle tier look like an obvious choice.

When to Change Pricing

Three triggers should prompt a serious pricing review:

1. Win rates above 50%. If you win most deals you enter, your price is too low. Strong companies have win rates in the 25-40% range with healthy margins.
2. NRR below stage benchmark. Expansion problems often have pricing-structural causes.
3. Competitive entry shifts. A new entrant pricing 40% below you forces a positioning response — sometimes lower price, more often clearer differentiation.

Don't change pricing more than once every 12 months on the same axis. Customers tolerate occasional pricing changes; they don't tolerate constant ones. When you do change, grandfather existing customers for at least 12 months — the goodwill cost of not grandfathering is almost always higher than the revenue gained.

How to Choose

Walk through this short checklist:

1. Does value scale with people or with consumption? People → tiered. Consumption → usage. Neither → flat.
2. Is your buyer technical or non-technical? Technical buyers tolerate usage pricing; CFOs and procurement prefer predictable contracts.
3. What is your gross margin per unit of consumption? Usage pricing only works if margin is positive at the meter, not just on aggregate.
4. What is your competitor doing? If the category standard is tiered, going usage-based is a positioning bet, not a default choice.

Next Steps

Pricing changes are among the highest-leverage decisions a SaaS company makes. They are also the hardest to reverse, so model before you ship.

• Quantify each model's impact on year-1 ARR with the SaaS Pricing Calculator.
• Pair every pricing change with a fresh LTV Calculator run to confirm unit economics still work.
• Plan to revisit pricing every 12-18 months — the most valuable lever in SaaS is also the most neglected.