There are four common ways to charge for software: per user (per-seat), per unit consumed (usage-based), in packaged plans (tiered), or free with paid upgrades (freemium). The right one is the model where your price rises in step with the value a customer gets. Most successful products end up combining two of them.
Why does the pricing model matter so much?
Your pricing model decides who buys, how fast revenue grows inside each account, and what your developers must build. Two products with identical features can have very different businesses depending on whether they charge per user or per transaction.
It also shapes the sales conversation. A buyer can predict a per-seat bill to the cent. A usage bill needs trust and good reporting. A free plan creates thousands of sign-ups but very few conversations. Choosing deliberately, before launch, saves painful migrations later, because changing what existing customers pay is one of the hardest things a SaaS company does.
How do the four models compare?
Each model ties price to a different measure of value, and each carries a distinct risk. Here is the short version.
| Model | You charge for | Best fit | Main risk |
|---|---|---|---|
| Per-seat | Each named or active user | Collaboration tools, CRMs, team software | Customers share logins to cut costs |
| Usage-based | Units consumed: messages, API calls, storage, transactions | Infrastructure, communication, AI and data products | Unpredictable bills scare buyers; revenue swings month to month |
| Tiered | Packages with feature or limit differences | Most B2B SaaS with distinct customer sizes | Too many tiers confuse; wrong limits push people to the cheapest plan |
| Freemium | Nothing, until users need more | Products that spread virally or self-serve at low cost | Free users cost money to serve and rarely convert |
Hybrids are common: a tiered plan with a seat allowance and a usage cap, or a platform fee plus metered usage.
When does per-seat pricing make sense?
Per-seat works when each extra person using the product genuinely gets value from it. Team chat, project management and sales tools fit naturally, because a ten-person team gets more out of the product than a two-person team.
Its strengths are predictability and easy maths for the buyer. Its weakness is that it punishes adoption: every new user is a new cost, so customers ration access. Watch for these warning signs:
- Customers invite only one or two "power users" and export data for everyone else.
- Shared logins appear in your session data.
- Your product's value comes from automation, not from people logging in.
A common fix is charging only for active users each month, or offering free "viewer" seats while charging for editors.
When does usage-based pricing make sense?
Usage-based pricing fits when the customer's value grows directly with volume, and the volume is easy to measure. If you send WhatsApp messages, process documents, run AI queries or move money, each unit has a cost to you and a value to the customer.
It lowers the barrier to starting, since small customers pay little, and revenue grows automatically as accounts grow. The downsides are real, though:
- Buyers, especially finance teams, dislike bills they cannot forecast.
- Your monthly revenue becomes harder to predict.
- You must build accurate metering, usage dashboards and alerts before you can charge a cent.
If your own costs are mostly variable, such as AI model or messaging fees, some usage component protects your margins. Many AI products now pair a base subscription with included credits and charge for extra usage on top.
When does tiered pricing make sense?
Tiered pricing suits products whose customers fall into clear groups with different needs and budgets. A freelancer, a growing agency and a large company might all use the same core product, but want different limits, support and controls.
Good tiers follow a few rules:
- Three tiers is usually enough. A fourth "Enterprise: contact us" option can sit alongside them.
- Separate tiers by value, not by artificially crippling the cheap plan. Limits on projects, records, team size or advanced features work better than removing basics.
- Put the features larger customers must have in the higher tier. Single sign-on, audit logs, advanced permissions and priority support are classic examples.
- Make the middle tier the obvious choice for your ideal customer.
Tiers also make annual discounts and upgrade paths easy to explain, which helps self-serve sales.
When does freemium make sense?
Freemium makes sense only when serving a free user costs you very little and free users bring in paying ones. Products that spread by sharing, such as documents, forms or design tools, can justify it because every free user shows the product to others.
For most niche B2B tools, a free trial is the better choice. A 14- or 30-day trial of the full product creates urgency; a permanent free plan lets people stay comfortable on it forever. Before offering freemium, answer three questions honestly:
- What does one free user cost you each month in hosting, support and third-party fees?
- How does a free user lead to revenue: by upgrading, inviting a paying colleague or referring others?
- What specific limit will make a growing user want to pay?
If you can't answer the second question, freemium is a marketing expense rather than a pricing model.
How does your pricing model change what you need to build?
Pricing is not just a page on the website; it is a set of features in your product. Founders often underestimate this, so budget for it explicitly.
| Pricing element | What developers must build |
|---|---|
| Per-seat billing | Seat counting, invite limits, proration when seats change mid-cycle |
| Usage-based billing | Event metering, usage dashboards, spend alerts, overage invoicing |
| Tiers | Feature flags or "entitlements" that switch features on per plan |
| Free trial | Trial expiry, reminder emails, graceful downgrade |
| Freemium | Hard limits, upgrade prompts, abuse protection |
| Annual plans | Discounts, renewals, refunds and plan-change rules |
Billing platforms handle the charging, but your product still needs to know what each customer is allowed to do. Our feature-by-feature SaaS cost breakdown shows how billing and account features fit into the total budget.
How should you set prices for your first customers?
Start higher than feels comfortable, with fewer plans than you think you need, and talk to every early customer about value. Underpricing is the more common mistake, and raising prices for existing customers later is awkward.
A practical sequence:
- Find the value metric. What does the customer measure success by: leads captured, hours saved, orders processed? Price along that line if you can.
- Anchor to the alternative. Compare your price with what the customer pays today, whether that is staff time, spreadsheets or a competitor.
- Launch with one or two plans and an enterprise option for larger buyers.
- Offer founding-customer pricing, locked for a period, in exchange for feedback.
- Review after the first 20–30 paying accounts and adjust before you scale marketing.
This fits naturally with the build-small-first approach in our guide to building an MVP without wasting money. Pricing is one chapter of a longer story, told in full in our 2026 SaaS playbook, from first idea to first paying customer.
Frequently asked questions
Which SaaS pricing model is best for a new B2B product?
For most new B2B products, simple tiered pricing with a seat or usage limit in each tier is the safest start. It is easy for buyers to understand and easy to build. Add pure usage-based charges only if your costs or customer value clearly scale with volume.
Should I show prices on my website?
For self-serve products under a few hundred dollars a month, yes; hidden prices create friction and lose sign-ups. For complex enterprise deals, show starting prices and offer a "talk to us" option. Showing at least a starting point filters out buyers who could never afford you.
How often can I change my pricing?
You can test pricing for new customers as often as you learn something meaningful. Changing prices for existing customers should be rare and well communicated, ideally with notice and a grandfathering period. Many companies keep early customers on their original plan as a goodwill gesture.
Is usage-based pricing risky for customers in India and Australia?
It can be if bills are unpredictable. Spend caps, usage alerts and prepaid credit packs make usage pricing much easier to accept. Pricing in local currency, rupees or Australian dollars, also removes exchange-rate surprises from the buyer's side.
Do I need a billing platform from day one?
Not always. Some early SaaS companies invoice their first handful of customers manually and add automated subscription billing once the plans settle. Once you have more than a few dozen accounts, or any self-serve sign-ups, automated billing becomes essential.
Getting pricing right before you build
Your pricing model affects your database, your permissions and your onboarding, so it is worth settling before development starts rather than after. As part of our SaaS and web app development services, we help founders turn a pricing idea into the entitlements, metering and billing flows a real product needs. If you'd like a second opinion on your model and a scoped build plan, talk to our team about your SaaS idea.