Why we measure willingness to pay

Our company is called PXQisR. Read it aloud and it says P × Q is R: price times quantity is revenue.

We chose the name because it is the one statement in pricing nobody can argue with. It holds whatever your customers do, whatever market you sell into, whatever model you charge by. Add one more, that profit is revenue less cost, and you have nearly everything in pricing that is certain.

Almost everything else is an assumption. Assumptions can be wrong, and in pricing they usually go untested.

You choose P. Your customers decide Q.

Of the two numbers on the left of that equation, only one is yours. You set the price. How many customers buy at it is decided by them, one at a time, by their willingness to pay.

Willingness to pay is the highest price at which a customer would still choose your offer over their best alternative. That alternative might be a competitor, a spreadsheet, an internal build, or doing nothing at all.

Each part of that definition matters:

• The highest price: a limit, not an average.
• Would still choose: a decision, not a feeling. A customer can love your product and still not pay for it.
• Their best alternative: the thing your price is really competing with, which is often not another vendor.

So willingness to pay is not what the product cost to build. It is not what a competitor charges. It is not what a customer says the product is worth. It is the point at which their choice changes.

Every price leaves money on both sides

Two facts make this hard. Customers differ: for the same product, some would pay several times what others would. And the number is hidden: you can see company size, usage and what customers choose, but never what each one would have paid.

So any single price loses revenue twice. Customers who would have paid more buy anyway and keep the difference. Customers who would have paid less, but still more than it costs to serve them, walk away. In software and data, serving one more customer costs almost nothing, so nearly every customer who walks is revenue you could have had.

Your revenue report never shows either loss. It shows what you captured, not what you left.

What most prices rest on instead

In practice, new product pricing is usually set in the last few weeks before launch, on one of three things:

1. What it cost to build. Your customer never sees that number, and it does not change what they would choose.
2. What a competitor charges. That tells you about one alternative. It does not tell you what your customer will do.
3. What customers said. Naming a number in a survey or an interview costs the customer nothing, so stated answers tend to run higher than what people go on to pay.

None of these is evidence of willingness to pay. Each is a guess about it, dressed as a method.

Every price is judged against another price

There is a second reason willingness to pay is hard to pin down. Customers rarely judge a price on its own. They judge it against a price they already hold in mind: a list price, a competitor’s quote, what they paid last year.

That reference moves willingness to pay even when nothing about the product changes. And a loss against it hurts more than an equal gain pleases. Which reference the customer holds depends on the moment you are pricing for, and that gives product and marketing teams four distinct pricing moments:

• A new product. There is no reference yet, so the launch price becomes it. Price too low at launch and you pay for it at every renewal.
• A new feature. The reference is the current plan. Every feature is a pricing decision: it earns, expands, keeps customers, or it is given away.
• A pricing refresh. The reference is the last price paid. Every price rise is measured against it, not against your new value.
• A successor or a sunset. The reference is the plan being retired. Price the successor so staying beats switching.

Whatever the moment, the price then has to survive finance, sales and the board. A price you can’t evidence is a price you can’t defend.

So: measured

If P × Q is the only certainty, and willingness to pay decides Q, then the job is to measure it rather than assume it. Measured means four things:

1. What customers do, wherever they have done it. Purchases, upgrades, renewals and departures show what customers will pay. Once customers are paying, that is the evidence we use.
2. Before anyone has bought, trade-offs, not direct questions. The questions are built around the alternative customers use today and what they already spend on it, and the answers are corrected for the fact that saying costs less than paying.
3. A range, not a point. As the pricing author Tim J. Smith argues, there is no single optimal price; there is a range customers will accept. Every result comes with its margin of error.
4. Assumptions in the open. Each one is written down with where it holds, where it fails and what would disprove it, so when one falls you know which conclusions fall with it.

Most pricing decisions today rest on the weakest kind of evidence: assumed. Some reach asked. Very few are observed or measured. Moving a decision up from assumed is the work.

That is what PXQisR is for. If you have a price you are least sure of, we would like to hear about it. Get in touch