The difficult work sits inside the noun resolution. Intercom had to define when an issue was actually solved, make that decision visible, and exclude outcomes the product could not control. The pricing innovation depends on measurement architecture before it depends on packaging.
THE INTERCOM / FIN SYSTEM
Intercom / Fin
An AI agent resolving more than 2M issues a week
Price the verified outcome, then reorganize the company around the product winning the new category.
Intercom chose a unit customers can audit: a support issue fully resolved without human intervention. It set the price against labor value rather than software seats. As Fin became the growth engine, the company elevated it from feature to brand.
2 plays worth stealing.
Open a play for the mechanism, the exact receipt, the failure mode, and a deployment brief Elena can put into your backlog.
26 Price the auditable outcome
Fin charges $0.99 when it fully resolves a customer issue without a human, instead of charging by seat or message volume.
The unit links price to labor value and lowers adoption friction: customers pay when the system produces the promised result.
Intercom reports Fin resolving more than 2M issues per week. The same outcome definition is measured across every conversation rather than selected case studies.
Choose an outcome both parties can observe, define the failure boundary, meter it transparently, and price below the customer’s next-best way to achieve it.
Do not price an outcome controlled mostly by the customer. Intercom rejected a pure percentage-of-sales model for that reason.
Company-reported. Intercom explains the resolution definition, price, rejected alternatives, and weekly volume.
Open the exact source ↗Turn this observation into a real experiment.
- Run it when
- The product produces an outcome both sides can observe and the vendor substantially controls.
- Owner
- Pricing lead, product analytics, finance, and customer success
- First sprint
- Define success and failure, audit 100 cases with customers, expose the meter, and test price against the alternative cost.
- Leading signal
- Verified outcomes, dispute rate, gross margin, expansion, and customer savings.
- Stop rule
- Do not launch outcome pricing while resolution disputes or external dependencies dominate the measured unit.
27 Let the winning product eat the parent brand
Intercom announced that the company would become Fin while retaining Intercom as the underlying service platform.
The rebrand concentrates memory, narrative, and investment on the product that defines the future category instead of forcing it to remain a sub-feature of the legacy suite.
The company positioned Fin as its future after the agent reached multi-million weekly resolution volume and expanded into an API platform.
When one product has a stronger category pull than the parent, test whether customers already use its name as the company shorthand. Align the brand only after the behavior is clear.
A rebrand does not create an inflection. It should acknowledge one that customers and revenue have already made undeniable.
Unfairly synthesis. The rebrand and product volume are reported. The category-concentration logic is our analysis.
Open the exact source ↗Turn this observation into a real experiment.
- Run it when
- One product already has stronger category pull and revenue momentum than the parent name.
- Owner
- CEO, brand lead, and customer research
- First sprint
- Measure customer shorthand, search behavior, win-loss language, and product revenue before testing a focused brand architecture.
- Leading signal
- Unaided recall, branded demand, qualified conversion, and retention through the transition.
- Stop rule
- Do not rebrand if the new name is not already carrying customer memory and commercial pull.
Do not copy Intercom / Fin. Adapt the system to your constraint.
Elena learns your product, customer, funnel, and current bets. Then she chooses the relevant pattern, scopes the first sprint, and watches the leading signal.