CRM Battleground for AI Support

Salesforce buying Fin turned AI support from a fast growing startup category into a core CRM battleground. Fin already proved that a support agent could sell on outcomes, not seats, with per resolution pricing and cross channel automation. Inside Salesforce, that model now gets pushed through a much larger installed base, which raises the pressure on Sierra and Decagon to win with deeper workflows, faster deployment, and expansion beyond classic support.1 2 3 4 5

• Fin had already expanded beyond Intercom’s own inbox before the deal, selling as a standalone agent on top of Zendesk or Salesforce and through API access. That mattered because it showed AI support could detach from the old help desk bundle and become its own buying decision.2 4
• The acquisition also reset the scale bar. Fin closed the Salesforce deal at about $3.6B, after reaching roughly $419M in ARR by April 2026, while Sierra reached an estimated $400M ARR by October 2026. The category is no longer early experimentation, it is a race among companies already operating at major software scale.1 2 3 5
• Competition is shifting from who has a chatbot to who can actually finish the job. Decagon built around backend actions like refunds and cancellations, while Sierra is extending from service into sales workflows through Horizon. That means the next wedge is not answering questions, it is completing multi step customer work across systems.5 6 7

From here, AI support is likely to split into two durable lanes. Large suites like Salesforce will bundle strong default agents into the system of record, while independents push outward into broader customer facing agents that handle service, sales, and account changes in one flow. The winners will be the ones that can take action across more systems with less setup.1 4 5 6 7