OpenAI and Deepseek Drive AI Costs Up, Forcing MSPs to Rethink Pricing Models
The structural mechanism driving current changes for MSPs is a shift from seat-based software revenue toward variable, usage-based AI consumption, resulting in pronounced margin pressure and operational complexity. This shift is being shaped by enterprise software vendors, including Atlassian and HubSpot, moving away from flat per-user AI fees in favor of metered pricing models tied directly to consumption. The episode also identifies increased rework and governance burdens for MSPs, particularly as automation and AI adoption reduce traditional seat counts but introduce new variability and labor demands around oversight, exception handling, and security remediation.
The most consequential development highlighted is the transition by a growing number of vendors to usage-based AI pricing, treating AI as a metered utility rather than a bundled feature. The Information reports that by the end of 2025, 79 out of 500 tracked software companies are expected to have implemented some form of usage-based AI fee. This adjustment is driven by vendors’ need to offset the potential revenue loss resulting from AI agents reducing seat license counts. Org View data cited in the episode suggests that 55% of companies who laid off staff in favor of AI later regretted the decision, underscoring the unexpected operational burdens and instability introduced when automation is rushed or incomplete.
Additional developments reinforce this structural shift. Semaphore describes open-source models like Deepseek offering lower-cost, competitive AI, which increases adoption even beyond premium vendor ecosystems. The CIA’s deployment of AI-generated intelligence reports—expected to be ubiquitous in analytics platforms within two years—signals the integration of AI into core workflows. Vendor activity, such as Appdirect’s acquisition of Partner Stack, reflects a market trend favoring platforms capable of provisioning, governing, and managing diverse AI toolsets and workflows for customers who lack internal capability.
For MSPs and IT service leaders, these trends introduce direct pricing pressure, unpredictable pass-through costs, and expanded liability exposure. The transcript emphasizes the need to separate AI rework pricing from security incident response, implement controls on AI usage and licensing, and reframe AI engagements around workflow governance rather than tool deployment. Failure to formalize and price these activities increases the risk of unbilled labor, contract ambiguity, lender skepticism, and downward pressure on margins, especially as the gap widens between shrinking seat-based revenue and volatile AI consumption charges.
00:00 Metered AI
03:34 Governance Is Margin
05:17 Seat Drop Math
08:36 Why Do We Care?
Supported by:
Upcoming event:
The Pivotal Point of IT: Building Services for the AI-First Era
Date: May 13 at 1p.m. EDT
Register: https://go.acronis.com/davesobelaiera
💼 All Our Sponsors
Support the vendors who support the show:
👉 https://businessof.tech/sponsors/
🚀 Join Business of Tech Plus
Get exclusive access to investigative reports, vendor analysis, leadership briefings, and more.
<
