SaaS pricing tools: how software companies set the right price
SaaS has no landed cost to mark up - every price is a judgment about value. The value metric, the tooling that gets used, and why packaging beats price points.
Software pricing is strange for one specific reason: the marginal cost of another customer is close to zero. There is no landed cost to mark up, no freight, no shrink. Every number on a SaaS pricing page is a judgment about value, which is why so many of them are picked in a meeting and never revisited.
That absence of a natural floor is what makes SaaS pricing tools a distinct category from retail pricing systems - the vocabulary overlaps (our pricing glossary covers the shared terms), but the underlying problem is different. Retail asks "what will the market bear against my cost." SaaS asks "what is this worth, and how do I charge in a way that grows with the customer."
Start with the value metric, not the number
Before any tool matters, you need a value metric: the unit you charge for. Seats, API calls, contacts stored, gigabytes processed, transactions handled.
Get this wrong and no amount of price testing rescues you. A good value metric has three properties - it scales with the value the customer receives, the customer can predict it before signing, and it is hard to game. Per-seat pricing fails the first test in products where five people get most of the value from a hundred licenses. Pure usage-based pricing often fails the second, which is why procurement teams resist it.
Most mature SaaS companies land on a hybrid: a platform fee plus a usage component, with the usage metric chosen so that a customer growing their business naturally grows their bill.
The tooling that actually gets used
Willingness-to-pay research. Van Westendorp surveys, conjoint analysis, Gabor-Granger. Tools like Wynter, Maze, or a well-run customer interview program. Unglamorous and consistently more informative than internal debate.
Analytics and cohort tools. Amplitude, Mixpanel, or your warehouse. You are looking for which features correlate with retention and expansion - that tells you what belongs in which tier far better than intuition does.
Billing infrastructure. Stripe Billing, Chargebee, Metronome, Orb. If your billing system cannot express the pricing model you want, you will end up with the pricing model your billing system allows. This constraint is real and shapes more SaaS pricing than anyone admits.
Pricing page testing. Tier order, feature framing, whether you show a number at all on enterprise. Small changes here move conversion more than the prices themselves often do.
Experimentation frameworks. The hard part of SaaS price testing is that you usually cannot A/B test prices cleanly without fairness problems. Most companies test on new cohorts and grandfather existing customers.
The optimization logic underneath - margin constraints, elasticity, guardrails on how far an automated system can move - is structurally similar to what a price optimization engine does in retail, just with different inputs.
Packaging beats price points
Most SaaS companies underperform on packaging rather than on the numbers. Three tiers where the middle one is engineered to be the obvious choice. Feature gating on capabilities buyers can evaluate, not obscure ones. An enterprise tier without a public price, because at that size you are selling, not listing.
The common mistake is gating on features customers need to experience value rather than features that scale with value. Lock the wrong thing behind a paywall and you throttle adoption in the exact segment you needed to grow.
The thing SaaS teams skip
Price localization and periodic review. A price set at launch and never touched for four years is almost certainly wrong now - the product has more value, the market has moved, and inflation alone has eroded the real number.
The discipline that makes this manageable is the same one that works in retail: encode your constraints, review on a schedule rather than a crisis, and keep the reasoning behind each change auditable. If you want to see what that continuous-review model looks like when it is fully operationalized, agentic pricing is the retail expression of the same idea.