How to price a SaaS product
There are only three honest ways to arrive at a price, and every pricing consultant in the world is selling some arrangement of them. You can price from what it costs you, from what everyone else charges, or from what the customer gets. The calculator above runs all three at once, because the interesting number is not any one of them — it is how far apart they are.
A product whose value price is ten times its cost price is a product with room to grow into. A product whose cost price already exceeds the market rate is a product with a business-model problem that no amount of clever packaging will fix. You want to know which one you have before you print a pricing page.
The three methods
Cost-plus — your floor
Work out what it costs to serve one customer for a month: your fixed costs divided across however many customers you have, plus whatever each additional customer costs you directly in infrastructure, support and payment fees. Then gross that up to the margin you want.
Software businesses typically run 70–85% gross margin, which means the price is roughly three to seven times the cost to serve. If yours comes out much lower, something in your cost base is behaving like a services business — usually hand-holding support, or per-customer infrastructure that never amortises. That is worth fixing before it is worth pricing around.
Cost-plus is a terrible way to set a final price and an essential way to find your floor. Nobody has ever bought software because it cost the vendor a lot to make.
Competitor-anchored — the expectation
Buyers do not evaluate your price in a vacuum. They evaluate it against the last similar thing they paid for, which means a competitor's pricing page has already trained them before they reach yours. You get three positions against that anchor:
- Undercut. Roughly 30% below. Wins deals on price, attracts the customers most likely to leave for the next cheaper thing, and makes raising prices later very hard.
- Match. The safe default. Moves the conversation off price and onto features, which is where you would rather have it.
- Premium. Roughly 30% above. Only works if the buyer can articulate why — a specific capability, a specific audience, a specific outcome. "We are nicer" is not a reason.
The trap is picking your comparison badly. If you anchor against a tool that solves an adjacent problem for a different buyer, you will inherit their price ceiling for no reason.
Value-based — your ceiling
The most defensible price is a share of the money the customer stops spending. Ask three customers what they did before you existed and how long it took. Multiply those hours by their loaded cost — salary plus overhead, not take-home pay — and you have a monthly figure they will recognise.
Charging 10% of the value you create is easy to justify and easy to say yes to. Charging 25% is where procurement starts negotiating. Above that you are asking the customer to share the upside, which works in some markets and gets you thrown out of others.
Value pricing is also the only one of the three that gets better as you improve the product. Cost-plus rewards you for being cheap to run; value-based rewards you for being worth more.
What founders get wrong
- Launching too low to avoid rejection. A low price does not remove the objection, it moves it. Cheap software gets evaluated as cheap software, and the buyer wonders what is missing.
- Assuming they can raise it later. Raising prices on existing customers is one of the hardest conversations in the business, and most founders never actually have it. Your launch price is closer to permanent than you think.
- Copying a competitor's number without their cost base. They may be subsidising that price with funding, a different product mix, or a much larger customer count.
- Pricing per seat on a tool that needs adoption. Per-seat pricing makes the buyer ration who gets access. If your product only works once the whole team uses it, you have priced against your own retention.
- Having only one plan. A single price forces every buyer into the same box. Two or three tiers let people self-select, and the top tier makes the middle one look reasonable.
Choosing what to charge for
The unit you price on matters more than the number. Pick something that grows when your customer succeeds — projects shipped, contacts managed, revenue processed — rather than something that grows when they merely use you more, like storage or API calls. The first makes a price rise feel like a sign of progress. The second makes it feel like a penalty.
Whatever you land on, put a number on the page. "Contact us for pricing" costs you every founder who was ready to buy at 11pm and is not going to fill in a form to find out whether they can afford you.
Test it on people, not on a spreadsheet
Once you have a number, the next step is not more modelling. Quote it to the next ten prospects and watch what happens. Nobody flinching means it is too low. Everybody flinching means it is too high, or you have not made the value legible. A few flinching and buying anyway is roughly right.
Then you need people to find the thing at all. That is what AlphaShot is for — put your product on the board, hold it for a week, and get it in front of founders who go looking for new tools on purpose.