Most SaaS companies leave money on the table not because their product is weak, but because their pricing was never designed to scale.
Getting your SaaS pricing strategy right is one of the highest-leverage decisions you'll make, and most founders get it wrong in the same direction: they set a number early, never revisit it, and wonder years later why their best customers feel like a bargain. The mechanics aren't complicated. The discipline is.
There are four pricing models you'll encounter in SaaS, and the right one depends less on your category than on what your customers actually measure.
Flat-rate pricing, one price for the whole product, is the simplest to explain and the hardest to sustain. Basecamp famously charges $299 a month flat, regardless of how many users or projects you have. It works for them because simplicity is part of the brand promise. For most companies, especially ones selling into businesses of different sizes, flat-rate pricing either leaves enterprise revenue uncollected or prices out the SMBs who would have grown into bigger accounts.
Per-seat pricing, the model Salesforce and most of the legacy CRM market built their fortunes on, scales cleanly with usage and is easy for procurement to approve. The downside: it creates perverse incentives. Teams share logins, limit rollout, or actively resist expanding the tool precisely because every additional user is another line item. Slack ran into this before introducing free tiers. It's a real constraint on viral adoption inside an org.
Usage-based pricing ties the bill to what the customer actually consumes, whether that's API calls, data processed, messages sent, or contracts generated. Twilio and Snowflake are the canonical examples. Twilio's revenue grows with its customers' businesses without requiring a renegotiated contract every time usage doubles. The trade-off is unpredictability: customers nervous about surprise invoices sometimes cap usage in ways that artificially limit the value they extract. That's a retention risk.
Value-based pricing SaaS is the hardest to implement and usually the most profitable. It ties price to the business outcome you deliver, not the feature set or consumption volume. Veeva Systems, which sells into life sciences, doesn't price like a generic CRM. It prices against the cost of regulatory non-compliance, which runs into the millions. When you can anchor price to the outcome the customer cares about, you stop competing on features and start competing on ROI. Most SaaS companies say they want to do this. Very few actually do the customer research required to make it credible.
When to Switch Models as You Scale
Here's the thing: the model that got you to $1 million ARR probably won't get you to $10 million, and definitely won't get you to $50 million. The inflection points are predictable.
Early on, simplicity beats precision. A flat-rate or low-friction per-seat structure removes friction at the sales layer when you can't afford a long evaluation cycle. Notion started with a freemium tier that put almost no limit on what a solo user could do, using individual adoption to build bottom-up pressure inside organizations. That's a deliberate strategy: land with low or no cost, expand when the team is already hooked.
The first real pressure to change your pricing model usually comes when you start selling upmarket. An SMB paying $49 a month and a 500-person company with five departments using your product should not be on the same plan. If they are, you've already left a significant amount of revenue uncollected. The move here is almost always toward tiered pricing, three to four plans that create natural migration paths as usage or organizational complexity grows.
Structuring those tiers well is where most companies stumble. The mistake is building tiers around feature gates, locking capabilities behind the Enterprise plan to force upgrades. Customers resent it, and it creates support overhead when someone on the mid-tier plan wants to know why a single feature requires jumping two price points. Better practice is to identify your expansion metric, the one signal that correlates most tightly with the customer getting more value, and build your tiers around that. For a platform like HubSpot, it's the size of the contact database and the number of marketing automations running. The tiers don't feel arbitrary because the thing you're paying more for is the thing making you more money.
Structuring Tiers That Convert SMBs Without Losing Enterprise
The tension in SaaS pricing tiers is real: price too low to attract small teams and you anchor the perception of your product's value for the enterprise buyer who finds you later. Price for enterprise from the start and you can't build the user base that creates social proof and organic referrals.
The cleanest resolution is a freemium or self-serve entry point that genuinely delivers value on its own, paired with a clear and non-punishing upgrade path. Figma did this. The free tier lets individual designers do real work. Teams need the paid plan. Organizations need the enterprise contract. At no point does the free tier feel deliberately crippled; it's just limited by the right variable, number of collaborators, which happens to be the exact thing enterprise buyers care about most.
For enterprise pricing, custom contracts are usually unavoidable above a certain revenue threshold, but that doesn't mean pricing should be completely opaque. Publishing a starting price, even a floor, reduces the number of discovery calls that go nowhere because a prospect assumed you were out of their budget. Intercom has moved back and forth on transparency here, and the periods of published pricing have generally produced shorter sales cycles for their mid-market segment.
The annual contract discount is worth taking seriously. Offering 15 to 20 percent off for annual prepayment is standard, but many companies underuse it as a conversion tool. A monthly subscriber who hasn't churned in six months is a strong candidate for an annual offer. The math works for the customer, and the cash flow certainty is worth more to you than the discount costs. Chargebee data from several years of SaaS billing analysis has consistently shown that annual customers churn at roughly half the rate of monthly customers.
One thing that almost never works: discounting to close a deal and calling it a special offer. It trains the buyer to wait for discounts, signals that your list price is fictional, and creates internal pricing inconsistencies that become messy at scale when your sales team compares notes. If you need to close a deal, add implementation support or extra seats before you touch the base price.
Frankly, most SaaS pricing problems are research problems in disguise. Companies guess at what the market will bear because they never asked. A customer willing to pay $200 a month and one willing to pay $2,000 a month for the same tool are not in the same conversation, and no amount of tier-naming creativity will substitute for knowing which one you're actually talking to.
Also read: How to Build a Pitch Deck That Gets You Into the Room With Tier-1 VCs • Build a SaaS Customer Feedback Loop That Actually Moves Your Roadmap • SaaS Landing Page Best Practices for Turning Cold Traffic Into Trials
