Customer Support Ground Zero for AI
Front vs Intercom
AI hit customer support first because support had the cleanest before and after math. A company could replace a $10 to $15 human handled ticket with a roughly $1 to $1.50 AI resolution, using the same help docs and back office systems it already had. That let buyers justify rollout fast, cut or freeze support hiring, and immediately weaken the seat expansion engine that older help desk software depended on.
-
The product change was concrete, not theoretical. Older bots mostly routed tickets or matched keywords. New systems like Fin, Sierra, and Decagon could read knowledge bases, hold open ended conversations, and resolve 60% to 80% of cases without a human, often with CSAT at or above human teams.
-
This split the market into two models. Intercom bundled AI into an existing help desk, keeping seats but adding usage pricing at $0.99 per AI resolved ticket. Sierra and Decagon sold more like BPO replacements, priced on outcomes and designed to take full ownership of resolutions.
-
Front sits in the middle of that transition. Its core business still comes from per seat shared inbox software, and it last posted 137% net dollar retention in the pre AI expansion era. As support teams flatten, Front has to make money from automation and cross team workflows, not just from adding more agents.
The next phase is a shift from AI answering tickets to AI running the whole customer conversation layer across chat, email, and voice. That favors platforms that can combine automation, human escalation, and deep system actions in one workflow. Support was the beachhead, but the winners will expand into sales, success, onboarding, and retention from there.