There is no pricing rule that suddenly becomes valid at customer one, twenty or one hundred. Early pricing improves when each offer is treated as a dated hypothesis tied to a specific buyer cohort, then revised when the evidence changes.
Early pricing is a sequence of offers, not a ladder
The familiar low-to-high price ladder looks decisive, but it hides the questions that matter. Was the buyer evaluating a self-serve product or an assisted service? Did the package include data cleanup? Was the objection about price, trust, missing scope or lack of urgency? A number detached from those conditions teaches very little.
Keep the offer stable long enough to understand a coherent cohort, but do not wait for a universal sample size. A cohort can be defined by buyer situation, acquisition source, package, onboarding method and offer date. Change one important pricing hypothesis at a time so you can still explain what happened.
The pricing hypothesis log
| Field | What to record | Why it matters |
|---|---|---|
| Offer version | Date, plan, billing unit, billing cadence and any setup work. | Prevents later accounts from being compared against an offer they never saw. |
| Cohort definition | Buyer job, company shape, source and onboarding path. | Separates price response from a change in audience. |
| Value hypothesis | The old workflow cost or risk the buyer is expected to recognize. | Makes the reason for the price inspectable. |
| Package boundary | Included outcome, limits, service and exclusions. | Stops scope changes from masquerading as price changes. |
| Conversation evidence | Exact objection, question, acceptance condition or non-response. | Preserves what the buyer said instead of reducing everything to "too expensive." |
| Cost to serve | Setup effort, recurring support, processing and manual exceptions. | Shows whether an accepted price is economically usable. |
| Decision | Keep, revise, narrow, raise, separate service or reject the cohort. | Turns the log into an operating record rather than a diary. |
Illustrative cohort entries
The entries below show the shape of the record. They are fictional, contain no market benchmark and do not report customer results.
| Cohort | Offer hypothesis | Observed evidence to capture | Possible interpretation |
|---|---|---|---|
| Owner-operators using self-serve setup | One recurring plan covers the standard workflow. | Buyers understand the outcome, but setup questions repeat before purchase. | Keep the price test stable and fix the setup explanation before discounting. |
| Agencies importing client data | The same plan includes assisted migration. | Migration consumes founder time and every file needs a different cleanup step. | Separate migration as a bounded service or narrow the accepted input. |
| Teams requiring approval | A higher package covers roles and approval history. | The buyer asks about control and rollout rather than the headline price. | Test the package boundary; the concern may be operational risk, not price level. |
The point is not to manufacture tidy lessons. Real logs include ambiguous conversations, unsuitable buyers and changes that do not produce a clear answer. Record that uncertainty instead of forcing a result.
Revision triggers that do not rely on customer counts
- Value-language failure: qualified buyers cannot connect the offer to a current cost, risk or completed job.
- Package confusion: buyers repeatedly ask which plan covers the same ordinary workflow.
- Service leakage: accepted accounts require recurring manual work that the price and package never named.
- Cohort mismatch: the offer attracts buyers whose workflow is outside the intended product boundary.
- Metric distortion: the billing unit creates invoice surprises or discourages the action the product is meant to support.
- Operational strain: onboarding and support consume more founder capacity than the offer can sustain.
None of these dictates an automatic price increase. The right response may be a clearer promise, a smaller package, a setup fee, a new value metric, a different cohort or a deliberate no.
Change one variable and preserve the history
Before revising an offer, copy the current version into the log. Note the reason for the change and the cohort that will see it. Do not quietly edit old entries, because the comparison depends on knowing which buyers saw which terms.
Useful single-variable tests include changing the price while keeping scope stable, separating setup from recurring access, changing the billing unit while keeping the outcome stable, or narrowing the buyer definition. Changing the buyer, package, price and onboarding process together may improve sales, but it will not tell you which pricing hypothesis was wrong.
Next-offer record
What the first hundred framing gets wrong
Customer count is not a maturity model. A product can reach many low-touch accounts without learning how to price an assisted enterprise workflow. Another product can learn a great deal from a small number of complex pilots. The useful milestone is a clearer offer with a documented reason, not a round account total.