Subscription based pricing: choose the model by how your buyers budget

Subscription based pricing charges customers a recurring fee, monthly or yearly, for continued access to a product. Choosing a subscription based pricing model comes down to one question: what should make one customer's bill bigger than another's? Pick the answer from how your buyers budget and why your customers leave, which means buyer interviews and churn data, not a competitor's pricing page.

There are four common models: one flat price, a price per seat, tiers, and a subscription with a usage charge on top. Then there is a second decision that gets less attention than it deserves: monthly or annual terms.

The four models differ in what makes the bill grow

We read these pricing pages on October 15, 2026. Prices change, so check the live pages.

Tiers sit on top of the other three (you can have tiered flat plans or tiered per-seat plans), so the first decision is the charge unit. Tiered pricing covers how to draw tier lines, and usage-based pricing covers choosing a metered unit. This post is about which base model to start from.

Flat rate is easiest to buy and leaves money with your biggest customers

A flat price is one number. Nobody has to count users or forecast anything, and a single manager can often approve it without asking finance. Basecamp makes this the pitch: its Studio, Pro and Unlimited plans all advertise unlimited users with no per-user fees.

The cost is that a 400-person customer pays the same as a 12-person one. Choose flat pricing when your customers are similar in size, when the people using the product are a small part of the value, and when a quick purchase decision matters more to you than capturing large accounts. If your customer list spans a wide range of company sizes, flat pricing is a gift to the big ones.

Per seat tracks value when every added person benefits, and invites rationing when they don't

Per-seat pricing grows with the team, so it fits collaboration products where each new person makes the product more useful for everyone else. It is also easy to explain, which is one reason it is so common.

It fails when value doesn't follow headcount. If one analyst runs reports that the whole company reads, a per-seat price charges for one person and delivers to 200. The other failure is rationing: customers share a login or remove occasional users to save money, and those people stop forming the habit that keeps the account alive.

One way around rationing is to charge only for people who use the product. Slack's fair billing policy treats a member who hasn't used Slack in more than 28 days as inactive and credits the account for the unused part of the billing period. Customers can invite everyone without paying for the people who never show up.

Add a usage charge only when heavy customers cost you, or get far more value

A seats-plus-usage hybrid keeps a predictable subscription for most of the bill and adds a meter for the part that varies. Intercom's pricing shows the pattern: a per-seat price for support staff, and a separate $0.99 for each outcome its AI agent handles.

Add a meter when two things are true. Usage varies widely between customers who pay the same subscription, and either your costs rise with that usage or the heavy users get much more value. If neither is true, the meter adds a line to the invoice and a reason for finance to ask questions. (Finance always has questions. Give them fewer.)

Annual terms move churn to one date a year

The other decision is the contract length. Monthly plans let customers leave in any month. Annual plans collect a year up front, usually at a discount, and concentrate the leaving into renewal time.

Slack's pricing shows a typical discount. Pro is $8.75 per user a month billed monthly and $7.25 billed annually, which is about 17% off by our calculation. Basecamp sells its Unlimited plan only on yearly billing.

ChartMogul's SaaS Retention Report 2023, based on aggregated data from more than 2,100 SaaS businesses (ChartMogul sells subscription analytics software), shows a large drop in new-customer retention in months 11 and 12 and attributes it to churn on annual plans. Annual terms don't remove the decision to leave. They schedule it.

Two cautions when you compare annual and monthly churn in your own data:

• Customers choose their plan. People who pick annual are often already more committed, so part of the gap you see comes from who signs up for annual, not from the contract length.
• An annual customer who stopped using the product in month three is still counted as retained until month 12. Watch usage, not only billing, or the bad news arrives all at once.

Short version: Offer monthly and annual plans with a discount for paying up front. Make annual-only the rule for large contracts where the buyer expects a yearly procurement cycle anyway.

Let churn data tell you which model is failing

If you already charge customers, your billing history is the cheapest evidence you have. Pull cancellations, downgrades and seat changes for the last year and cut them by account size. Churn analysis covers the calculations.

Then look for these patterns:

Treat each pattern as a lead for interviews, not a verdict. A dozen cancellations in one segment can move a percentage a lot.

Ask buyers how they budget, not which model they like

People asked "would you prefer per seat or flat?" will usually say whichever sounds cheaper. Ask about real purchases instead:

• "How did you pay for the last tool like this, and who approved it?" Approval rules decide whether a growing bill is welcome or a problem.
• "Is your budget for this set per person, per team, or as one line for the year?" Buyers who budget per head accept per-seat pricing more easily.
• "How many people would use this, and how many would you actually pay for?" The gap between the two answers is your rationing risk.
• "Do you pay for software monthly or annually, and what happens at renewal?" This shows whether annual terms help them or block them.
• "What would make you cancel?" Listen for the events that end subscriptions: a budget cut, a new manager, a project finishing.

Talk to people who bought, people who chose a competitor and people who cancelled. Pricing strategy covers the research for setting the price level itself.

A made-up example: inspection software for property managers

Here is a made-up example. Dana runs a company selling inspection checklist software to residential property management firms. Today it is $199 a month, flat, for any company size.

The churn data shows two problems. Firms managing fewer than 200 units cancel at more than twice the rate of larger firms, and the firms managing 3,000 units pay the same $199 as everyone else. (One of them has 41 people logged in, including someone whose job title is simply "Clipboard.")

Interviews with eight property managers show that budgets are set per property, not per person. Inspectors change often, so nobody wants to count them. Larger firms approve software once a year.

That points away from per-seat pricing. Dana moves to a price per managed unit with a monthly minimum, so small firms start cheaper and large firms pay in proportion to the work. (It keeps the logic of per-seat pricing but counts apartments instead of people.) She offers an annual plan at about two months free, and sells it as annual-only above 1,000 units, where buyers already run a yearly approval.

Your next step

Write down what makes one of your customers worth more than another: people, usage, properties, transactions. Then check last year's cancellations and downgrades by account size to see whether your current model charges for that.

To test the model with people who budget for this kind of software, Instant Expert can find people matching a description such as "operations directors at property management firms with more than 1,000 units." You review who it finds, it sends your invitations, and you pay only if someone books a call or answers. The directory pages for subscription pricing experts and pricing strategy consultants are a good place to start.