McKinsey: 32% of companies no longer buy software — they build it with AI agents

By Carlos Montiel | Enterprise AI Specialist
Leer en español →
Published: 2026-09-03 | By: Carlos Montiel | Reading time: ~5 min

McKinsey's "State of AI 2026" survey, run between May 4 and June 8, 2026 with 1,719 respondents, found that 32% of organizations decided against buying at least one piece of software because they could build it in-house using agentic coding tools. In the tech sector, that figure rises to 41%.

The shift from "buy" to "build"

The report's central finding is a measurable shift in how companies decide between acquiring commercial software and building it with AI agents. Among the highest-performing AI companies — the 6% of respondents who attribute at least 5% of their EBIT to artificial intelligence — nearly half already skipped a software purchase to build it with coding agents, versus 31% among the rest of organizations. The share of large companies scaling agents across one or more functions rose from 27% to 40% over the past year.

Operating cost is starting to bite

Adoption isn't friction-free: one in five respondents said their organization is limiting AI usage due to operating-cost considerations. This suggests that as agents move from pilots to production, compute and inference spending is becoming a real factor in scaling decisions, not just a theoretical concern.

Risk to watch: building instead of buying shifts the responsibility for maintenance, security, and updates from the software vendor to your own team. A coding agent can generate 80% of a feature quickly, but the remaining 20% — testing, error handling, compliance — still requires human judgment. Companies that underestimate that hidden cost end up with silent technical debt.
Impact for companies using or buying AI: if your organization is evaluating a SaaS purchase this quarter, it's worth first asking whether an internal team with coding agents (Claude Code, Copilot, Cursor) could cover 80% of the use case in less time than a procurement cycle takes. The build-vs-buy decision is no longer exclusive to companies with large engineering teams — with agents, a small team in Guatemala or Latin America can seriously evaluate building in-house what used to be viable only to buy.
Carlos Montiel
Enterprise AI Solutions Architect
LLMs, Agents & Orchestration Specialist
guatemalia.com/#contacto · info@guatemalia.com

Need to implement AI in your company?

Carlos Montiel is an enterprise AI solutions architect. He implements LLMs, Agents, RAG and orchestrators for companies across Guatemala and Latin America. Reach out for a consultation.

Contact Carlos Montiel

info@guatemalia.com