SaaS Pricing [2026]: What I Learned From Getting It Wrong First

I've launched three SaaS products. I got pricing wrong in different ways each time, and the mistakes were expensive in different ways. The first product was underpriced and served the wrong customers as a result. The second was overpriced relative to perceived value and didn't convert. The third I got closer to right, and I can trace the improvement directly to understanding why I'd been wrong before. Here is what I actually learned.

The Fundamental Mistake: Cost-Plus Pricing

Most founders price their SaaS products by calculating their costs — infrastructure, team, overhead — and adding a margin. This produces a price that covers costs but has nothing to do with the value the customer receives or what they're willing to pay. The customer doesn't care what your AWS bill is. They care what the product does for them and how that compares to their next best alternative.

Value-based pricing starts from a different question: how much value does this product create for the customer, and what fraction of that value is reasonable to capture as price? A tool that saves a developer two hours per week at a $100/hour rate is worth $800/month in value. Charging $20/month for that captures 2.5% of the value created. Charging $100/month captures 12.5%. Both might be sustainable businesses; only the second one accurately reflects the product's contribution.

The problem with value-based pricing is that it requires understanding your customers' economics well enough to estimate the value you create. This requires customer conversations — real ones, not surveys — before you set a price. Most founders skip this step because the product isn't built yet and talking to customers feels premature. It's the opposite of premature; it's the most important early activity.

What Freemium Actually Does (and Doesn't)

Freemium is a marketing strategy masquerading as a pricing strategy. It's a customer acquisition tool, not a revenue model. The business logic: acquire a large number of users at zero cost, convert some percentage to paid. This works when the conversion rate and the monetized tier's pricing produce economics that justify the cost of supporting the free tier. It frequently doesn't work, which is why many companies that launched with freemium have quietly killed or severely restricted it.

The problems with freemium for small SaaS companies: the free tier users consume support, infrastructure, and attention while generating no revenue. If your conversion rate is 2-3% (industry average) and your paid tier is priced too low, you're essentially subsidizing a large number of non-customers. The product has to be valuable enough at the free tier to drive adoption, but that value can't be so complete that paying feels unnecessary.

When freemium makes sense: when the product has network effects that increase value as more people use it (collaboration tools, communication platforms), when the marginal cost of an additional free user is genuinely near zero, and when the conversion path from free to paid is natural and well-designed. When it doesn't make sense: single-player tools, products with meaningful per-user infrastructure costs, or early-stage companies without the resources to support a large free user base.

The Packaging Problem

Getting the price right is half the problem. Packaging — which features go in which tier at which price — is the other half, and it's equally important. The goal of packaging is to present options that feel natural and fair at each level, where the value clearly justifies the price at every tier. Poor packaging creates situations where customers feel either that the entry tier is too limited to be worth the trial or that upgrading doesn't add enough to justify the cost.

The packaging mistake I made most often: using seats as the primary differentiator when the customers' primary value driver was something else entirely. If a customer is choosing your product because it saves them time, they don't want to pay per seat — they want to pay for the time saved. If the value driver is data volume, per-seat pricing creates resentment. Align your pricing metric with how customers experience value.

Annual vs Monthly Billing

Offering annual billing at a discount (typically 2 months free, equivalent to ~17% discount) is standard practice for good reason: it dramatically improves cash flow, reduces churn (customers who pay annually are much less likely to cancel mid-year), and provides a committed revenue signal that's useful for planning. The standard discount is 15-20%. Discounts below 10% generally don't move the needle; discounts above 25% leave too much revenue on the table.

I now push annual billing in every onboarding interaction, offer it prominently in pricing pages, and make the savings concrete ("save $240/year"). The conversion to annual billing is one of the highest-leverage metrics to improve in an early-stage SaaS, because every annual subscriber is worth dramatically more in lifetime value than a monthly subscriber who churns at the median monthly churn rate.

The Test You're Not Running

Most SaaS founders set a price and leave it alone. The price should be an ongoing experiment. Test higher prices with new cohorts while honoring existing customers' pricing. Watch conversion rates carefully — if raising prices doesn't decrease conversion, you were underpriced. Talk to customers who didn't convert and ask specifically about price. Talk to customers who did convert and ask what other tools they were considering and what they pay for those. This competitive pricing intelligence is worth more than any pricing framework.

My honest take: Talk to customers before setting a price. Charge more than feels comfortable. Align pricing metric to value driver. Run pricing experiments. Almost everything else is secondary.