B2B SaaS pricing strategy: what product marketers need to know (and own)

Pricing is the most consequential number in a B2B SaaS business, and product marketing usually meets it last.

Finance runs the unit economics. Sales lobbies for less friction. Product argues for the features that justify a premium tier.

We get the calendar invite when the pricing page needs copy.

Every grocery store installed self-checkout so nobody would need help. Then they had to station somebody next to it anyway, because the machine can't handle a produce code, an age check, or an unexpected item in the bagging area.

A price is supposed to work the way that lane was supposed to work. It gets read alone.

When every deal needs a rep standing beside the number to explain it, discount it, and clear the error, we haven't set a price. We've built a machine that needs an attendant.

Call that the attendant test. The rest of this is about passing it.

The wrong anchors

Most B2B SaaS is underpriced, and the reason is rarely nerve. Nobody did the research to justify a bigger number, so we picked one that felt defensible.

Cost-plus is the first anchor. Add hosting, cost-to-serve, and the margin we want, then round up.

The output is honest and it has nothing to do with what the product is worth to a buyer. Cost-plus produces a floor. Prices live somewhere above the floor, and cost-plus can't say where.

Competitor benchmarking is the second. We land inside everyone else's range and call it market rate, without asking whether our competitors are underpriced too, selling different value to a different buyer.

Discounting is the third. When a price gets set without research, reps discover the real ceiling through negotiation, deal by deal, and list price quietly turns into fiction.

Tighter discount governance won't repair that. The repair happens earlier, at the moment the number gets set.

What they'd actually pay

Willingness-to-pay research is the structured version of a question we usually ask by accident. What will this buyer really pay, before we publish anything?

Van Westendorp is the cheap version. Four questions inside a buyer interview. At what price is this too expensive to consider, expensive but worth it, a good deal, or so cheap you'd doubt it works?

Fifteen to twenty of those interviews plot into an acceptable range and an optimal point.

Deal interviews are cheaper still and more neglected. Run win/loss with price as the actual subject. Not "was price an issue," but what their budget was and how our number compared to what they walked in expecting.

Conjoint analysis is the rigorous version. It needs a hundred-plus respondents and a research budget, and most of us don't need it yet.

We already sit on enough buyer conversations to run a version of this without spending anything. The real question is whether pricing is on the research calendar before the number gets set.

The unit we charge on

The value metric is the unit the price attaches to. It decides how revenue grows as customers get more value, which makes it the most structural choice in the whole strategy.

Per seat is the default. It fits when value comes from access and collaboration, and it fits badly when the value lands on the organization no matter how many people log in.

Usage fits infrastructure and developer tools, where value scales with volume. It costs the buyer predictability, and plenty of enterprise buyers will pay more for a flat number they can put in a budget.

Object-based metrics count the things being managed. Contacts, documents, properties, monitored endpoints.

Outcome-correlated metrics are the hardest to price and the closest to the truth. When a customer describes what the product is worth in the words they'd use with their CFO, they've just named our value metric.

One question sorts all of it. As a customer's business grows and they get more out of the product, does the number they pay on grow with them?

If our best customers get ten times the value on the same seat count, the expansion revenue is sitting there unclaimed.

The menu problem

A restaurant with eleven laminated pages has a kitchen that never decided. So we order the safe thing, or we ask the server what's good, which is another way of asking for an attendant.

SaaS packaging fails the same way. Starter, Professional, Enterprise, defined by which switches are off in each one, with no story about who any of them is for.

A tier is a positioning statement. Starter should be the whole product for a specific buyer at a specific stage, and it should solve that buyer's problem completely.

Starter tiers crippled enough to force an upgrade produce customers who churn because they never got an outcome. Price was never the reason.

The mid tier belongs to buyers who outgrew the entry motion, and the features in it should be the ones customers ask for after six to twelve months of use. Win/loss data already names them.

Enterprise is more than the same product with SSO bolted on. Security review, data residency, admin control, contract terms. Those buyers are organizationally different, and the price tracks that difference.

Most companies break the same rule. Every tier should stand on its own as a complete product for the buyer it targets, and a tier that forces an immediate upgrade is a trial with friction.

The frame around the number

Competitive price positioning means putting our number inside a frame where the comparison makes sense. Matching the market is beside the point.

Charging more works when the outcome is better or the total cost of ownership is lower. The work is making that difference explicit enough that a buyer can defend it to their CFO without us on the call.

Charging about the same is the most common position, and it demands the sharpest differentiation on everything except price. When the numbers match, the clearer story wins.

Charging less on purpose works for displacement and for land-and-expand, as long as we know how and when we intend to move up. Otherwise we've taught the market to expect a number we'll have to take back.

Competitor pricing also moves. Research from eighteen months ago is decoration.

The number we can't take back

New products need a price before there's any evidence for one. The temptation is to launch low, win adoption, and raise it once the value is proven.

Raising it later is harder than it sounds. Buyers anchor on the first number they ever saw.

Early customers who paid fifty dollars a month become the loudest objectors at a hundred and fifty, even when the product barely resembles what they bought. So the company discounts to keep them and underprices for everyone who would have paid the higher number on day one.

Price at the value we intend to deliver, then earn it with the experience. Run the buyer interviews before launch, find the range and the floor below which the product looks too cheap to trust, and open at the midpoint with the value story already written.

Time-limited introductory pricing does something a plain discount can't. It gives an early buyer a reason to act now instead of an anchor we'll spend two years trying to move.

The raise nobody resents

Every SaaS product eventually raises prices. A company that never does is giving away everything it has built since the last time it decided.

Increases go badly when they arrive as a billing note from Finance. The customer reads it as a policy change, gets defensive, and the renewal conversation turns adversarial.

The narrative is the whole job, and it answers one question. What has this product done since the last price was set? Nobody in the history of software has been persuaded by "our costs have gone up."

Twelve weeks is the minimum runway. An executive note with full context, then account-level conversations, then grandfathering decided on criteria we set in advance rather than handed to whoever pushes hardest.

We write the announcement, the conversation guide, and the objection responses, and we train them into CS before the first email goes out.

Then watch net revenue retention. A good increase lifts it, and a spike in logo churn is a verdict on the communication rather than on the number.

Where to start

Most of us inherit a pricing structure we didn't design and weren't asked about. Ownership doesn't start with an overhaul.

Run win/loss with price as the focus for one quarter. Ask what the budget was, and how our number compared to what they expected walking in. That data says more than any internal analysis.

Map the gap between list price and average selling price. When the gap is wide and consistent, list price is fiction, and that one chart gets attention in rooms we're usually not in.

Then interview five customers from each tier. Why that tier, what they never use, what they wish came with it. The answers tell us whether the packaging matches how people actually buy.

And build the ROI case for the price we charge today. If we can't say in specific numbers why the product is worth its cost, the buyer can't either, and the buyer is the one who has to say it out loud in a budget meeting.

A price should be readable standing alone, at the machine, with nobody there to help.

What to do next

If every deal closes below list, and the pricing conversation keeps happening without us in the room, the discount policy is downstream. The price is carrying a positioning job nobody assigned it.

A Bare Strategy positioning audit is built for that moment. We run the buyer research, pressure-test the packaging, and get the value story to a place where the number can stand on its own.

If that's where you are, start here. The first conversation is free.