Most SaaS pricing gets set the same lazy way. You open three competitor pricing pages, average their numbers, add a "we're better" bump, and ship it. We have done it. It feels safe because everyone else did the homework, right? They did not. They copied someone too. So the whole category ends up priced off one founder's gut feeling from 2019.
Here is the stat that should sting. Paddle found that companies spend roughly six hours on their pricing strategy in the entire history of their business. Not six hours a year. Six hours, total, ever. Meanwhile pricing is the single most powerful lever you have: a 1 percent improvement in price can lift profit by around 11 percent, far more than the same effort spent on acquisition or retention.
This guide is how Sameer and Ankit actually set a price, minus the consultant fluff and the $40k pricing study. No sponsors, no affiliate links, nobody pays us to recommend anything. Just a repeatable way to pick a number you can defend, and the expensive pricing habits to cut. Let's strip it down.
◢Why is cost-plus the worst way to price SaaS?
Cost-plus pricing is the worst because it ties your price to your costs, and customers do not care what your costs are. They care what your product is worth to them. Adding a margin to your AWS bill produces a number that has nothing to do with the value you create.
This is the trap most founders fall into first. It feels rigorous because there is math involved. But cost-plus pricing fundamentally misaligns with how software economics work, since the marginal cost of one more user is close to zero. Price off cost and you systematically undercharge for software that saves someone ten hours a week.
The fix is value-based pricing: anchor the number to the outcome the customer gets, not your expense sheet. It is more work because you have to learn what that outcome is worth. It also pays. Pricing experts at Simon-Kucher's Global Pricing Study consistently find that companies with real pricing power, the ones that price on value, outgrow the ones that do not.
So before you pick a single dollar figure, throw out the cost spreadsheet. It is the wrong input. The right input is sitting in your customers' heads, and there is a cheap way to get it out.
◢What is a value metric, and why does it come first?
A value metric is the unit you charge for, and it should be the thing customers get more of as they succeed with you. Seats, contacts stored, emails sent, API calls, gigabytes, leads tracked: pick the one that grows when the customer wins. Get this right and your revenue rises naturally as accounts expand.
This decision matters more than the actual price. Charge per seat and you cap your revenue at headcount, even if a five-person team runs a million transactions through you. Charge on usage and your price scales with the value delivered. That alignment is exactly why usage-based pricing now appears in 43 percent of SaaS pricing models, up sharply year over year.
How do you find your metric? Look at why customers upgrade today. If your heaviest users hit a wall on the same thing every time, that wall is your value metric. We map this the same way we build any analytics setup: track what correlates with retention and expansion, then charge for that. A clean product analytics stack makes the metric obvious instead of a guess.
One warning: pick a metric customers can predict. Nobody wants a bill that swings wildly month to month. "Per active user" beats "per server CPU second" because the buyer can forecast it. If your usage metric scares people, soften it with a base fee plus overage, which is the hybrid most founders end up at anyway.
◢How do you find what customers will actually pay?
Ask them, with structure. The fastest, cheapest method is the Van Westendorp Price Sensitivity Meter: four questions sent to 30 to 50 target buyers that map the price range they accept. You get a defensible number in a week for the cost of a survey tool, not a quarter of consulting.
The four questions are deceptively simple. At what price is this too expensive to consider, expensive but acceptable, a bargain, or so cheap you'd question the quality? People struggle to name their willingness to pay directly, but they can easily tell you when a price feels wrong in either direction. Plot the answers and the crossover points reveal your acceptable corridor and a sensible launch price.
This is not theory. SurveyMonkey's guide to the method walks through running it in an afternoon, and Monetizely documents a SaaS launch where 250 responses pointed to a $15 to $40 range with an indifference point near $27. They shipped $28 and saw better conversion plus 17 percent longer retention. That is the payoff for asking instead of guessing.
Two cautions before you treat the survey as gospel. It captures stated intent, not real behavior, so it tends to run optimistic. And it struggles with multi-tier, recurring models. Use it to set the corridor, then confirm with real money: a price test, a few sales calls, an early cohort. Combining stated and actual data is what gets you close.
◢Which SaaS pricing model fits: usage-based, per-seat, or hybrid?
Match the model to your value metric, not to whatever is trendy. Per-seat fits tools where value scales with people logging in. Usage-based fits tools where value scales with what the product does. Hybrid, a base fee plus usage, fits most founders because it gives you predictable revenue and upside.
The momentum is clearly toward consumption. Around 38 percent of SaaS companies now use some form of usage-based pricing, up from 27 percent in 2021, and those companies post higher net revenue retention (120 percent or more) than subscription-only peers near 110 percent. Higher NRR means accounts grow on their own, which is the whole game. Industry benchmark data from High Alpha and OpenView shows the same drift as AI products tie price more tightly to value created.
But trendy is not a strategy. Per-seat still wins when seats are the value, like a team collaboration tool where ten users genuinely get more than two. The danger with seat pricing is it caps you and quietly invites password sharing. The danger with pure usage is bill shock and unpredictable revenue on your side. Hybrid threads the needle, which is why the average SaaS now runs about 3.2 public tiers plus a custom plan: a small base, then usage on top.
Whatever you pick, write the pricing logic so finance and ops can actually bill it. We have watched clever usage models die because nobody could invoice them cleanly. Keep the metering simple enough that your ops setup can track and charge it without a human reconciling spreadsheets at month end.
◢How many tiers should you ship, and how do you anchor them?
Ship three public tiers plus a custom enterprise option. Three is the proven default: a cheap entry plan, a popular middle plan you actually want people to buy, and a premium plan whose job is partly to make the middle look like a steal. That anchoring effect is real and it is free.
The math backs the gut feel here. The industry average is about 3.2 tiers, and 57 percent of companies still use per-user pricing as their primary model inside those tiers. Three options reduce decision paralysis while still letting buyers self-select by size. Stack on five or six and you stall the very decision your pricing page exists to speed up.
Design the tiers around your value metric, not random feature bundles. Each step up should unlock more of the thing customers want more of, with a couple of gating features that map to bigger teams. Name the middle plan something that signals "most people pick this," because most people will. This is the same self-serve motion we wire for go-to-market: let the product and the pricing page qualify buyers so a human only steps in for enterprise.
Then sanity-check the gaps. If your jump from the middle to the top tier is a cliff, you will push would-be upgraders to stay put or churn. Smooth the ladder so the next tier always feels like a small, obvious yes. And put a believable enterprise "talk to us" option at the top: it captures the whales without forcing you to publish a number you would rather negotiate.
◢What to cut: the pricing habits that cost you money
Here is where we earn our name. Most pricing pain is self-inflicted, and the fixes cost nothing but a little courage. Cut these habits and your existing traffic converts better, no extra spend required.
Cut competitor-copy pricing. Averaging three rivals' pages just inherits their mistakes and ignores your value. Cut cost-plus math entirely; it is the single most common pricing error and it always undercharges for good software. And cut the "set it and forget it" reflex. Companies with the most pricing power review every quarter and change roughly every six months, while everyone else freezes the number for years and bleeds margin.
Cut the urge to compete on being cheapest, too. A low price signals low value and torches the budget you need to actually grow. It also wrecks the other side of the ledger: weak pricing drives weak retention, and retention is where SaaS lives or dies. If churn is your real problem, fix the value and the onboarding first; our guide to reducing churn goes deeper than any discount ever could.
Finally, cut analysis paralysis. You do not need a perfect price, you need a defensible one you will revisit. Run the survey, ship a number, and put a 90-day reminder on the calendar. Want to see what your current tool spend is actually buying before you price your own? Run it through our stack cost calculator and cut what is not pulling its weight.
◢Conclusion
Pricing your SaaS is not a guess and it is not a competitor screenshot. It is a process: kill the cost-plus math, pick a value metric tied to customer success, ask 30 to 50 real buyers what they would pay, and set three tiers around the answer. Then ship a number and watch how it converts and retains.
The two habits that quietly cost the most are easy to drop. Stop copying rivals, and stop freezing the price for years. Review it every quarter, change it every six to nine months, and treat pricing like the highest-leverage feature you own, because it is.
Want the exact value metrics, survey templates, and tier structures we use when we set pricing for real products? That is what we publish every week. Subscribe to the Cut The SaaS newsletter and steal our playbook. Nobody pays us to recommend anything, so you get the honest version, including the parts the pricing consultants charge five figures for.
◢FAQ
How do you price a SaaS product without guessing?
Start with a value metric, the thing customers get more of as they grow, like seats, contacts, or events tracked. Then survey 30 to 50 target buyers with the four Van Westendorp questions to find the price range they accept. Set three tiers around that range, launch one number, and watch trial-to-paid conversion plus retention. Adjust within 90 days. The guess disappears once real buyers and real behavior tell you the number, instead of a spreadsheet of your costs.
What is the most common SaaS pricing mistake?
Cost-plus pricing, where you add up your costs and slap on a margin. Customers do not care what your servers cost. They care what the problem you solve is worth to them. The other big mistake is never changing the price again. Most companies spend roughly six hours on pricing in their entire history, then leave the number frozen for years while the product gets more valuable. Both mistakes leave money on the table you never have to give up.
How many pricing tiers should a SaaS have?
Three public tiers plus a custom enterprise option is the industry norm, and the average across verticals is about 3.2 tiers. Three works because it anchors a cheap entry plan, a popular middle plan most people pick, and a premium plan that makes the middle look reasonable. More than four tiers usually confuses buyers and slows the decision. Fewer than three removes the anchor that makes your main plan feel like the obvious choice.
Should an early-stage startup use usage-based or per-seat pricing?
It depends on your value metric, but usage-based pricing is winning for a reason. About 38 percent of SaaS companies now use some form of it, and those companies post higher net revenue retention than seat-only peers. Use per-seat if value scales with the number of people logging in, like a team chat tool. Use usage-based if value scales with what the product does, like API calls or emails sent. Many founders land on a hybrid: a base seat fee plus usage on top.
How often should you change your SaaS pricing?
Review it every quarter and expect to make a real change every six to nine months. Companies with the strongest pricing power re-evaluate on a schedule and adjust roughly twice a year, while most companies freeze their price for years and quietly lose revenue. You are not committing to a price for life. You are shipping a number, measuring how it converts and retains, and tuning it as the product and the market move. Treat pricing like a product feature, not a one-time decision.