Repricing a services-heavy SaaS product without burning trust
The analysis shifted the debate from "how much should prices rise?" to a more useful question: which pricing metric better reflected use, and how could existing customers cross the bridge without a trust shock?
- Client profile
- Vertical-software company
- Pricing context
- No price change in four years
- Commercial constraint
- Concentrated, annual-contract customer base
- Selected structure
- Usage bands with a 12-month bridge
The decision behind the brief
Leadership suspected under-pricing, but a poorly designed change could have damaged revenue and customer trust at the same time.
The company had not changed prices for four years. Its customer base was concentrated, contracts renewed annually, and two previous price-update emails had triggered escalations. The risk was not limited to whether a higher list price would be accepted. The team also needed to know whether per-seat pricing still matched how customers used and valued a services-heavy product.
The work therefore had to separate three questions: what customers appeared willing to pay, which pricing metric fit the value pattern, and how to migrate the existing base without treating every account as identical.
Constraints that shaped the work
- Customer concentration raised the cost of error. A small number of escalations could matter commercially even if the average response looked acceptable.
- Annual contracts delayed feedback. Each renewal cycle revealed only part of the customer response.
- Prior communication had already created sensitivity. Two earlier price-update emails had triggered escalations.
- Structure and level were different decisions. The evidence ultimately supported changing the pricing metric more strongly than applying only a headline increase.
Evidence used in the engagement
The published case identifies several research inputs and models. It does not disclose sample sizes, response rates, customer identities, contract values, model coefficients, or statistical-test details.
- Van Westendorp range-finding across the customer base provided evidence about perceived price boundaries.
- Choice-based conjoint covering the three most-used modules examined how customers traded off package attributes.
- Interviews with recently churned and recently won accounts added context from both sides of the switching decision.
- Price-volume-mix models under three elasticity assumptions tested how different customer responses could affect the economics.
- A margin waterfall exposed unmanaged discounting that was separate from the list-price question.
Analytical approach
Measure perceived price boundaries
Van Westendorp range-finding was conducted across the customer base. This moved the conversation away from internal opinions and toward a structured view of prices customers might perceive as too cheap, expensive, or prohibitively expensive.
General method note: Van Westendorp analysis is a range-finding technique, not a complete pricing answer. Its result depends on the audience, framing, product definition, and how hypothetical responses compare with actual purchase behavior.
Test tradeoffs among the most-used modules
Choice-based conjoint was used with the three most-used modules. The purpose was to understand tradeoffs among configurations, rather than asking customers to state a single preferred price in isolation.
General method note: Conjoint can reveal relative preference and simulated choice under the tested design. It remains sensitive to sample composition, attribute selection, and the realism of the choices presented.
Use edge cases to explain the mechanism
Interviews included recently churned and recently won accounts. These conversations helped surface how customers described the offer, switching decision, and likely objections to a change.
General method note: Interviews can explain why a pattern may exist, but they do not establish its prevalence without appropriate quantitative evidence.
Separate the pricing metric from a headline increase
The analysis supported moving from per-seat pricing to usage bands more strongly than applying a headline increase alone. That distinction matters because the metric determines how price changes as customer use changes.
General method note: A useful pricing metric should track customer value closely enough to feel intelligible, remain measurable, and avoid creating incentives that suppress healthy use.
Model uncertainty and design the migration
Price-volume-mix outcomes were modeled under three elasticity assumptions. The migration included grandfathering with a 12-month bridge, and communication was drafted around the two objections the research indicated would actually occur.
General method note: Elasticity scenarios are conditional cases, not forecasts. Their value is showing which result depends on customer response and where monitoring should focus during renewal.
How the evidence changed the decision
| Question | Published signal | Decision implication |
|---|---|---|
| Was the issue only price level? | Willingness-to-pay and module tradeoff research supported a metric change more strongly than a headline increase alone. | Move from per-seat pricing toward usage bands. |
| How should the base migrate? | The company had annual contracts, customer concentration, and a history of escalations. | Use grandfathering and a 12-month bridge rather than an abrupt universal switch. |
| What could undermine realized economics? | The margin waterfall exposed unmanaged discounting worth about three margin points. | Add a simple approval rule alongside the pricing change. |
| What should communication address? | The research identified two objections expected to occur. | Draft migration communication around those objections rather than a generic announcement. |
Recommendation and rollout logic
The recommendation combined four elements: move from per-seat to usage-banded pricing, model the result under multiple elasticity assumptions, give existing customers a 12-month bridge, and shape communication around the objections surfaced by the research.
The margin waterfall revealed a second lever that did not require the same customer-facing change: unmanaged discounting. A simple approval rule addressed that leakage alongside the broader migration.
Observed outcome
Revenue per account increased while reported churn remained in line with the prior baseline
Over two renewal cycles, net revenue per account rose by approximately 14%. The published case reports that logo churn was statistically indistinguishable from baseline. The discount approval rule closed unmanaged discounting worth approximately three points of margin.
Limitations and attribution
The facts are anonymized, lightly generalized, and rounded. The public case does not provide the research sample, response rates, price points, customer segments, renewal-cohort size, churn baseline, test statistic, confidence interval, or absolute revenue and margin values.
Because the statistical method and power are not disclosed, the statement about churn cannot be independently assessed from the published information. The reported outcomes followed the pricing and governance changes, but the public record does not isolate their effects from customer mix, product changes, sales execution, market conditions, or other factors.
Transferable lessons
- Test the pricing metric separately from the headline price level.
- Use preference research, behavioral context, and scenario economics together rather than asking one method to carry the decision.
- Treat migration design and communication as part of pricing strategy, not as post-analysis administration.
- Inspect discount governance because realized price can matter as much as list price.
Considering a pricing change?
Describe the current metric, renewal structure, and risk you are trying to manage. We will tell you which research and modeling would materially improve the decision.