How Should You Price a Product Nobody Has Compared Yet?

The pricing page is the last thing left before launch and it has been blank for a fortnight. Everything else had a reference to work from. The pricing page has none, because the usual method is to look at three competitors, sit somewhere between them and move on, and there are no three competitors. That is the situation this piece is about.

The relief, once you see it, is that your buyer is not pricing in a vacuum either. They have the problem today and they are already paying for it, in money or in somebody’s hours. That existing spend is your reference price, whether or not either of you has said it out loud. SaaS pricing in a new category is not an act of invention. It is a comparison against a line item that already exists in their budget, and your first job is to find out which one.

Our position: pick the alternative you displace, price against it, and put all of your remaining energy into choosing the unit you charge by. The number is adjustable later and everybody knows it. The unit is not, and almost nobody treats it with the seriousness that asymmetry deserves.

Nobody prices in a vacuum, including the buyer

When a buyer cannot compare you to a rival, they compare you to their status quo. This happens automatically and mostly silently, and it is the single most useful fact available to you. They are not asking whether your price is fair against a market rate. They are asking whether switching to you beats what they are doing on Monday.

Which means the pricing question sits downstream of the positioning question, and the two get answered together or badly. If you cannot say in one sentence what a customer stops doing after they buy, you do not have a price problem yet; you have a positioning problem. Deciding whether the thing is a product, a service or a feature belongs in the same conversation, and our parent company has set that out in product, service or feature.

Do not skip past the cheapest competitor on the list, either. Doing nothing costs nothing and carries an incumbency advantage no product can match. Plenty of first launches lose to it while believing they lost on price.

Find the line item you replace

The alternatives you are actually against

• A person and a spreadsheet. The most common incumbent by a distance. The cost is hours of somebody whose salary is already known, plus the errors nobody has counted.
• An agency or a contractor. The easiest comparison to make, because it arrives as an invoice every month and nobody has to estimate anything.
• Three tools held together by hand. Each one cheap, the total not cheap, and the maintenance invisible until somebody adds it up on your behalf.
• An internal build. Sometimes real, more often a half-finished script somebody defends in meetings. Either way it has an owner and an opportunity cost.
• Nothing at all. The problem is tolerated. Your price then has to clear the cost of the change itself, which is higher than any figure on your page.

The question that gets you the number

Do not ask what someone would pay. The answer is invented politely on the spot and it is worthless. Ask what they are doing about this today, who does it, how long it takes them, how often it goes wrong, and what happens when it does. Every one of those has a real answer, and together they produce a defensible figure that came from the buyer rather than from a workshop.

The value metric is the pricing decision

The value metric is the unit you charge by: the thing that goes up as the customer gets more out of you. Seats, projects, transactions, gigabytes, managed sites, monitored endpoints. It matters more than the number attached to it, because it decides who can afford to start, who grows into paying more, and what your customers do to avoid paying you.

Three tests a metric has to pass

It has to track value, so the bill rises roughly as the customer succeeds. It has to be predictable, so a buyer can estimate next year without a spreadsheet and a phone call. And it has to be countable without argument, because any metric you and the customer can measure differently becomes a monthly dispute, and disputes cost more than the revenue they defend.

The metric that punishes adoption

The worst metric is the one that charges for the behaviour you need in order to be useful at all. Per-seat pricing on something collaborative is the classic: the buyer restricts access to control the bill, half the organisation never touches the product, and it fails to embed for reasons that look like a product problem and are a pricing problem. Charge for the result the buyer wants more of, never for the participation that produces it. A publishing workspace such as Acrosite shows the shape of that choice plainly: its value sits in the sites it publishes and the deployments it triggers, not in how many people are allowed to open the editor, and a unit that follows the first will always be easier to defend than one that taxes the second.

Packaging: what each tier is for

Tiers are not sizes of the same thing. Each one answers a different question, and a tier that answers none of them is a column on a table that makes the decision harder. The entry tier answers "can I start without asking anyone for permission". The middle tier answers "does this work for my whole team". The top tier answers "will this survive our procurement, our security review and our contract".

Which is why the top tier is usually sold on control rather than capacity: permissions, audit trails, support commitments, invoicing that finance will accept. Splitting a core capability across tiers to force an upgrade is the tempting mistake, and it teaches buyers that the cheap plan is a trap. Publish what you have decided not to build while you are at it, because publishing your non-goals makes the tiers legible and saves the sales conversation that ends in disappointment.

How people get through the door belongs to the same decision. A trial and a free tier make different promises and attract different people, and the trade-offs are laid out in free trial or freemium. In a category nobody has heard of, the entry route carries more weight than usual, because it is where the buyer works out what the thing is at all. The page has the same job: sell the work being replaced rather than the feature inventory, which is the argument in a landing page that sells the job.

What getting the unit wrong actually costs

Changing a number is an email. Changing the unit is a migration. Every existing contract was written against the old unit, so each one has to be renegotiated or honoured indefinitely. Invoices change shape. Forecasts break, because last year is no longer comparable to this year. Sales compensation has to be rewritten. Somebody has to rebuild the billing logic and reconcile the customers straddling both models during the switch.

And there is a cost that never lands in a spreadsheet: buyers who chose you partly because the pricing was easy to explain internally now have to go back to their own finance team and explain that it changed. Some of them will not bother. That is why we treat the unit as an architectural decision rather than a commercial one, in the same family as the standing cost of every feature you ship: cheap to add, expensive forever afterwards.

The first price is a hypothesis, not a verdict

You will not get it right, and treating the first price as a test rather than a commitment changes what you do next. Its job is to be wrong in a way you can read. So publish it, sell against it, and pay attention to a handful of signals rather than to the revenue line, which tells you almost nothing this early.

Nobody flinching is a signal you are underpriced, and a suspiciously pleasant sales process usually means the same thing. Every deal stalling at the same step points at the packaging rather than the amount. Deep discounts closing quickly mean the number is wrong but the value is real. Small customers churning while large ones stay means the metric is scaled for the wrong end of the market. And a buyer who cannot explain your pricing to a colleague will not champion it, whatever they said on the call.

Reading those signals honestly takes the same care as any other measurement, because the story a team tells itself about why deals close is rarely the one the evidence supports. It is the same failure described in why your best performing channel is probably being miscredited, and it is worth someone owning the measurement side before the anecdotes harden into policy.

A price nobody argues with is not validation. It is a discount you decided to give before anyone asked for one.

Raising it later without burning the first cohort

Assume the first price is too low, because it usually is, and plan the increase before you need it. The mechanics are unglamorous and they work: new pricing applies to new customers first, existing customers keep their terms for a stated period, and notice goes out early enough that nobody hears about it from an invoice. Say what changed and why, in plain language, without the paragraph about continued investment in the platform.

Here is the honest concession. Grandfathering is a promise with a very long tail, and a founding cohort held on old terms forever becomes a real constraint on the shape of the business two years out, when their expectations no longer resemble anything you sell. Give the early cohort a genuinely good deal, put an end date on it, and say so at the start. Vagueness now is a negotiation later, and you will be negotiating from the weaker position.

One more thing, cheaper than it sounds: publish the awkward questions on the pricing page. What counts as a unit, what happens if we exceed it, what happens when we cancel, what support is included. In a category with no review sites and no comparison articles, your page is the only place those answers exist, and a buyer who cannot find them assumes the answer is one they will not like.

The sequence we would run

Six steps, in order, before anything is published. Most of the work sits in the first two, which is exactly where teams are most tempted to guess.

1. Name the alternative for each buyer type. Not the market. The specific thing this person does today, and who inside their organisation currently owns it.
2. Price that alternative from their side of the table. Invoices, hours, rework, the cost of the errors. Build it from what they told you, not from what you hoped they would say.
3. Choose the value metric, and write down why. Run it against the three tests, then picture a customer trying to avoid paying you and check that the metric still holds.
4. Design three tiers around three questions. Can I start, does it work for my team, will it survive procurement. Anything that answers none of those does not need a tier.
5. Instrument the metric and watch it before billing on it. A month of counting tells you more about the real distribution than any amount of modelling.
6. Publish, then put a review date in the calendar. Decide in advance which signals would move the price and in which direction, while you are still capable of being objective about it.

If you are heading into a first release, the pricing decision belongs beside the rest of the launch work rather than at the end of it, because it changes the page, the trial and the first conversation with every customer. Start from the line item you replace, defend the unit harder than the number, and set the review date before you need it. If you want a second read on a model before it goes public, tell us what you are replacing and who currently pays for it.