SaaS Pricing Models Compared: Flat, Tiered, Per-Seat and Usage-Based
Pricing is the growth lever most SaaS founders touch least. They set a number at launch, copy a competitor's tiers and never revisit it. This guide compares the four main pricing models, shows where each one wins and gives you a process for choosing and testing your own.
Key takeaways
- There are four core SaaS pricing models: flat rate, tiered, per seat and usage-based. Most mature products combine two of them.
- The best model charges along the same line that value grows for the customer, which is called the value metric.
- Per-seat pricing is simple and predictable but punishes you when your product saves labor instead of adding users.
- Usage-based pricing lowers the barrier to start and grows with the customer, at the cost of less predictable revenue.
- Treat pricing as a product feature. Review it at least once a year and test changes with new customers first.
Pricing matters more than most founders assume. In a widely cited 1992 Harvard Business Review article, McKinsey consultants Michael Marn and Robert Rosiello estimated that a 1% improvement in price, with volume unchanged, raised operating profit by about 11% for the average company in their sample. Few other decisions in a software business have that kind of leverage, and almost none are as cheap to change.
Yet pricing is often set in an afternoon. The founders look at two competitors, pick a number somewhere in between and split it into three tiers because that is what everyone does. The following sections explain the main models so you can make the decision on purpose.
Start with the value metric
Before you choose a model, decide what you are charging for. The value metric is the unit that grows as the customer gets more value from your product. For an email tool it might be contacts or sends. For a video platform it might be minutes streamed. For a sales tool it might be seats, because each extra salesperson is another person getting value.
A good value metric has three qualities:
- It tracks value. When the customer gets more out of the product, the metric goes up, so the bill feels fair.
- It is easy to understand. The customer can predict next month's invoice without a spreadsheet.
- It grows with the customer. As the account succeeds, your revenue from it rises without a renegotiation.
Each pricing model below is really a different way of packaging a value metric.
Model 1: Flat-rate pricing
Flat-rate pricing offers one product, one set of features and one price. Everyone pays the same amount regardless of how much they use it or how many people log in.
Basecamp is the best known example. For years the project management tool charged a single monthly fee with unlimited users and projects, and the company has repeatedly positioned that simplicity as a selling point against per-user competitors.
Flat rate: strengths
- Easiest model to explain and to sell
- No surprise invoices, so churn from billing shocks is low
- One plan to build, support and market
Flat rate: weaknesses
- Large customers get the same price as tiny ones, so you leave money on the table
- Small customers may find the single price too high and never start
- No built-in expansion revenue as accounts grow
Flat rate works best when your customers are similar in size and use the product in similar ways, or when simplicity is a core part of your brand.
Model 2: Tiered pricing
Tiered pricing offers two to four packages, usually labeled something like Starter, Professional and Business, each with a higher price and a bigger bundle of features or limits. This is the most common model in SaaS, partly because it lets you serve several customer segments with one product.
Tiers can be split along different lines:
- Feature tiers: higher plans unlock capabilities such as reporting, integrations or single sign-on.
- Limit tiers: higher plans raise a cap such as contacts, projects or storage.
- Service tiers: higher plans add priority support, onboarding or a dedicated account manager.
Most successful tiered pricing combines all three, so that the features, limits and service a large company needs all sit in the top tier.
Common mistake
Creating a middle tier that nobody wants. If the Starter plan has everything a typical customer needs and the Business plan has the one feature large accounts cannot live without, the Professional plan becomes dead weight. Each tier should map to a real customer segment with a real reason to upgrade.
Model 3: Per-seat pricing
Per-seat (or per-user) pricing charges a fixed amount for each person who has access to the product. Salesforce, Slack, Zoom and most collaboration tools use it, often combined with tiers so that price per seat rises with the plan.
The appeal is predictability for both sides. The customer knows that ten people cost ten times one person, and you know that revenue grows as the customer's team grows. Finance teams like it because the invoice is easy to forecast.
The weakness appears when your product saves labor rather than adding users. A tool that lets one accountant do the work of three is hugely valuable, but per-seat pricing charges for a single seat. Customers also tend to share logins when per-seat prices feel high, which depresses revenue and hides real usage.
Real-world example
Slack addressed one of the main complaints about per-seat pricing with what it calls fair billing. Under this policy, the company only charges for members who are actively using the workspace, and it credits the customer for seats that go inactive. The policy removed a common objection to buying seats for a whole company, since nobody wanted to pay for people who never logged in.
Model 4: Usage-based pricing
Usage-based pricing (also called consumption or pay-as-you-go pricing) charges for what the customer actually consumes: API calls, gigabytes, messages, minutes or transactions. Amazon Web Services, Twilio, Snowflake and Stripe are the best-known examples.
The model removes almost all friction at the start. A developer can begin with a few dollars of usage and grow to a large contract without ever renegotiating. Revenue tracks value closely, which keeps customers feeling the deal is fair.
The cost is predictability. Customers can be nervous about open-ended bills, and your own revenue swings with their activity. Seasonal businesses and companies that cut usage during a downturn will show up directly in your numbers. Many usage-based companies answer this with committed-use discounts, where the customer prepays for a volume of usage at a lower unit price.
The four models side by side
| Model | Charges for | Best when | Main risk |
|---|---|---|---|
| Flat rate | Access to the whole product | Customers are similar and simplicity is your brand | Big accounts underpay, small ones never start |
| Tiered | Bundles of features, limits and service | You serve several distinct segments | Tiers that do not match real customer needs |
| Per seat | Each user with access | Value grows with the number of people using it | Shared logins and undercharging for labor savings |
| Usage-based | Units consumed | Value grows with volume and customers start small | Unpredictable revenue and bill anxiety |
Hybrid models: what most companies actually do
In practice, the four models blend. A few common combinations:
- Tiered plus per seat: three plans, each priced per user per month, with features that unlock as you move up. This is the default for most B2B collaboration software.
- Platform fee plus usage: a fixed monthly subscription for access, plus charges above an included allowance. HubSpot, for example, includes a contact allowance in its marketing plans and charges for additional contacts.
- Tiered usage: plans defined by a usage cap, such as up to 1,000 subscribers, up to 10,000 and so on, with a price step at each threshold. Email marketing tools often use this.
- Free plus any of the above: a free tier that caps usage or features, designed to pull customers into a paid plan once they get value.
Hybrids let you keep the predictability of a subscription while still capturing expansion revenue as customers grow. The trade-off is that your pricing page gets harder to read, so keep the number of moving parts small.
Pro tip
If you are stuck choosing between per seat and usage, look at who in the customer's company gets the value. If every added person benefits, charge per seat. If one person benefits more as volume grows, charge for the volume.
How to choose and test your pricing model
- Interview ten customers about value. Ask what outcome they get from your product and how they would describe "more" of it. Their words usually point straight at the value metric.
- Shortlist two or three candidate metrics. Score each one for how well it tracks value, how easy it is to understand and how naturally it grows with the account.
- Model revenue under each option. Take your current customer list, estimate what each customer would pay under every model and compare total revenue, the spread between large and small accounts and how many customers would see a price increase.
- Check the cost side. Make sure the model does not let a heavy user cost you more to serve than they pay. Usage-based and flat-rate pricing are the usual danger zones.
- Test with new customers first. Show the new pricing to new signups before you touch existing accounts. Watch conversion rate, average contract value and the questions sales receives.
- Grandfather or migrate existing customers deliberately. Decide in advance whether current customers keep their old price, move with a notice period or get a discount for switching. Communicate it early and plainly.
- Set a review date. Put a pricing review on the calendar every six to twelve months. Costs, competitors and your product all change, and pricing should change with them.
Common pricing questions founders get wrong
Should the lowest plan be free?
Only if the free plan has a clear path to paid. A free tier works when usage naturally grows past the limit or when a team feature sits just behind the paywall. If free users get everything they need forever, you have built a charity, not a funnel.
Monthly or annual?
Offer both. Annual plans at a discount of 15% to 20% improve cash flow and reduce churn, because the customer makes one decision a year instead of twelve. Keep the monthly option for customers who want to try before they commit.
How many tiers?
Three is the usual answer because it gives a cheap entry point, a plan most customers should pick and a premium option that makes the middle one look reasonable. Add a fourth "Enterprise" plan with custom pricing only when you have sales staff to handle it.
Should I publish prices?
For self-serve products, yes. Hidden pricing kills conversion for small buyers who do not want to book a call. For large enterprise deals, "contact sales" is normal because the price depends on scope.
Your pricing review checklist
- Write down your current value metric and whether customers would describe value the same way
- List your customers by size and check whether the largest pay meaningfully more than the smallest
- Identify any plan that fewer than 10% of customers choose and decide whether to fix or remove it
- Model one alternative pricing structure against your existing customer base
- Run the new structure for new signups for 60 to 90 days and compare conversion and contract value
- Schedule your next pricing review and put the owner's name on it
Frequently asked questions
Which SaaS pricing model is most common?
Tiered pricing combined with per-seat charges is the most common structure in B2B software. Usage-based pricing has grown quickly, especially for infrastructure, API and data products, and many companies now mix a subscription with usage charges.
How do I raise prices without losing customers?
Give notice well ahead of time, explain what has improved since the customer signed up and consider grandfathering loyal accounts for a period. Testing the new price on new customers first tells you how the market reacts before you risk existing revenue.
Is usage-based pricing good for early-stage startups?
It can be, because it lowers the barrier to try the product. The risk is that revenue is hard to forecast while you are still small. Many early-stage companies pair a modest base subscription with usage charges to get some predictability.
How often should I change pricing?
Review it at least once a year and change it when the evidence says customers are getting far more value than they pay for, or when the mix of customers has shifted. Frequent small adjustments are easier to absorb than one large jump every few years.
The bottom line
Flat rate, tiered, per seat and usage-based pricing are four ways of charging for the same thing: the value your customer gets. Pick the value metric first, choose the model that packages it most clearly, model the revenue before you change anything and test with new customers before you touch existing ones. Then put the next review on the calendar, because pricing is never finished.
Written by
Creator
Writer at BizTips, sharing practical, research-backed advice for founders, investors and small business owners.