The short version
- A pricing model is wrong when the buyer has to be taught it before they can evaluate the product.
- Per-unit models ask a prospect to forecast their own volume in a category they have never bought.
- A flat tier asks the buyer for nothing they do not already know about themselves, which is why they can place themselves on it without a rep in the room.
- Usage pricing works when the unit is already a line in the buyer's budget. Where the unit exists because it is easy for the vendor to count, the model ships with a training requirement attached to it.
A pricing model is wrong when the buyer has to be taught it before they can evaluate the product. On one account, per-deal pricing confused every sales conversation, and the model was replaced with flat tiers.
What makes a pricing model hard to explain?
The buyer has to supply a number before they can get a price. Per deal, per transaction, per document, per active project: each asks a prospect to forecast their own volume in a category they have not bought before.
A model is explainable when the buyer can price themselves in one pass from something they already know. Headcount works, and so does a revenue band. The unit has to be a figure the buyer was carrying around before you arrived.
How the common models behave in a first sales call:
| Model | What the buyer must work out | Where the call stalls | Right when |
|---|---|---|---|
| Flat tier | Which tier they sit in | Rarely | Value does not track a countable unit |
| Per seat | How many people need access | Seat counting and access rationing | Every named user gets value |
| Per unit or per deal | Their own volume | Forecasting | The unit is already billed elsewhere |
| Usage or consumption | A rate times an unknown | Budget approval, the ceiling is invisible | The buyer already buys infrastructure this way |
| Quote on request | Nothing, until you tell them | Before the call, on the pricing page | Deal sizes vary by an order of magnitude |
Why did per-deal pricing stall every conversation?
It moved the first decision from us to them. A deal only becomes a countable unit once the buyer is working inside the product, so the prospect was being asked to forecast volume under a process they had not adopted.
It also made the price unpublishable in any useful form. A pricing page carrying a rate and a unit answers nothing, so every serious buyer became a call, and part of each of those calls existed to explain a formula rather than a product.
What does a flat tier ask the buyer for?
Nothing they do not already know about their own company. A tier is a shelf the buyer puts themselves on, and once the formula goes, the arithmetic that used to open the call goes with it. The price can then sit where a buyer expects to meet it, after they have worked out whether they want the thing at all, and the comparison happens on the website while nobody is in the room.
When is usage-based pricing the right call?
When the unit is already a line item in the buyer’s budget. Companies that buy compute, messages or transactions are practised at forecasting them and have the number to hand. Metering reads as fair because it matches how they already think about cost.
It goes wrong when the vendor picks the unit for accounting reasons rather than buying reasons. If the unit exists because it is easy for you to count, and your buyer has never counted it, the model ships with a training requirement attached.
If the pricing model needs explaining before the product does, it is the wrong model.
How do you test a pricing model before you ship it?
Hand the pricing page to someone in your ICP and ask them to price themselves out loud, with no help from you. Count the questions they have to ask before they reach a number. A page priced per deal sends the reader looking for the definition of a deal before they can start.
The number itself is a separate decision from the mechanic: what the price level tells a buyer and why pricing sits with marketing.
Frequently asked questions
What are the main SaaS pricing models?
Flat tiers, per seat, per unit or per transaction, usage-based metering, and quote on request. They differ in what the buyer has to work out before they can price themselves, which is the difference that shows up in a sales call.
How do you know your pricing model is wrong?
The sales call opens with the model rather than the product. If reps spend the first part of every conversation explaining how the price is calculated and prospects still cannot say what it will cost, the mechanic is the obstacle.
Is per-seat or usage-based pricing better for B2B SaaS?
Per seat is better when every named user gets value and the buyer can count those users today. Usage-based is better when the buyer already budgets for that unit, such as transactions or compute. Both fail when the unit is one the buyer has never tracked.
Why does per-deal or per-transaction pricing fail?
It asks the prospect to forecast their volume under a process they have not adopted yet.
Should you publish your prices?
Publish a tier and a number, or publish a floor. A pricing page carrying only a rate and a unit answers nothing, so every buyer becomes a call. A buyer who screens on price has already started evaluating you against an alternative.
How do you test a pricing model before launching it?
Hand the pricing page to people in your ICP and ask them to price themselves out loud, with no help from you. Count the questions they need before they reach a number, and listen for whether they can work out what happens to the price when they grow.
Sources
- Chua Network delivery data across 8 client accounts (internal fact bank)
- Chua Network engagement records, anonymized (internal experience bank)
