How AI Is Rewriting the WealthTech Vendor's Business Model

11 minutes ago

2 min read

O'Connell, J. (2026, September 14). How AI is rewriting the wealthtech vendor's business model. The Oasis Group. https://theoasisgrp.com/article/how-ai-is-rewriting-the-wealthtech-vendors-business-model/

Per-seat licensing rests on the assumption that each additional login carries near-zero marginal cost, an assumption AI breaks because every AI-generated summary, automated workflow, and drafted client communication carries a variable cost that scales with usage rather than headcount.
Usage-based and hybrid pricing grew from 30 percent of software companies in 2019 to roughly 85 percent by 2024, while pure per-seat pricing fell from 21 percent to 15 percent.
Seventy-three percent of AI vendors now charge separately for AI features rather than folding them into a flat subscription.
Atlassian bundles a fixed number of AI credits into its standard plans and charges per conversation beyond the allotment, HubSpot charges by the block of AI credits above the subscription tier, and Zendesk prices its AI resolution agent per resolved conversation instead of per seat.
Buyers are advised to confirm three terms before signing an AI-enabled renewal: whether the AI feature is priced as unlimited, capped, or metered; what occurs when a cap is crossed, including throttling or automatic overage billing at an undisclosed rate; and who absorbs a mid-term rate increase from the vendor's own AI provider.
Seventy-eight percent of IT leaders reported unexpected charges tied to AI features or consumption in the past year, and most of those charges surfaced after the contract was signed.
Recommended negotiating steps include requesting firm-level consumption data before renewal discussions begin, securing a defined cap with a disclosed overage rate in place of an open-ended fair-use clause, and locking any consumption-based rate for the full contract term.

Knote: The Oasis team is spot-on again. We have looked at several companies where the unit economics actually break at scale due to AI costs.