Build, buy or configure? Why 76% of enterprise AI is now bought, not built

A 2026 build vs buy AI analysis finds 76% of enterprise AI use cases were purchased in 2025, up from 53%, while in-house builds fell from 47% to 24%. For UK mid-market firms the question has become what thin custom layer is worth building.

Searches for “build vs buy AI” are rising among UK mid-market firms, and the numbers behind them are stark. A 2026 analysis of enterprise AI spending reports that 76% of AI use cases were purchased in 2025, up from 53% in 2024. Internal builds fell from 47% to 24% in a single year.

The surprise in the data

You might read that and conclude custom software is dying. The same analysis points the other way: 35% of surveyed teams have replaced at least one SaaS tool with custom software, 78% of builders expect to build more in 2026, and 51% have shipped production AI software. Buying and building are both growing, because they are no longer the same decision.

The four buckets

The framework in circulation sorts every workflow into one of four groups:

The recommended pattern is layered: buy the model and assistant platform, configure it to your workflows, and build only the thin custom layer that encodes what makes your business different.

The cost most plans miss

Custom builds on bought models typically take around 90 days from start to production for a defined workflow. The same source warns that maintenance costs exceed initial development. A budget that only covers the build is the wrong budget.

Another figure should worry any CTO: 60% of teams report creating software outside IT oversight, and 25% do so frequently. Ungoverned building is the real risk, not building itself.

So what

Do not ask “should we build or buy AI?” Ask which of your workflows sit in the “build it” bucket, and whether the answer depends on data or rules a vendor cannot see. If the honest answer is a small number of them, you need a focused custom layer, properly maintained, not a platform. That is the work behind our custom software development service, and if you want to test a specific workflow against the four buckets, start a project.

← All research

More research.

GitHub Copilot adds Supabase, Webflow, Wix and five more connectors in public preview

GitHub Copilot now connects to eight more services, including Supabase, Webflow, Wix and Mixpanel, through verified MCP connectors in public preview. They join more than 25 existing ones.

Read ›

Gemini 4 Neon is generally available in GitHub Copilot, aimed at lower-cost agentic coding

Gemini 4 Neon is now generally available in GitHub Copilot for agentic coding. Google says its quality is close to Gemini 4 Carbon at a lower token cost per task.

Read ›

Claude Code 2.1.328 brings auto mode to Bedrock, Vertex and Foundry, and narrows /code-review to bugs

Claude Code 2.1.328 extends auto mode to Claude Opus 4.6 and Sonnet 4.6 on Amazon Bedrock, Google Cloud and Microsoft Foundry. /code-review now reports only bugs and CLAUDE.md violations.

Read ›

GitHub Copilot Evals: repeatable tests for AI agents, now in public preview

GitHub Copilot Evals lets teams write repeatable tests for AI coding skills, custom agents and MCP tools. It checks both the result and the path the agent took, and is in public preview.

Read ›

GitHub Copilot can now drive your browser: browser use enters public preview

GitHub Copilot browser use lets an AI coding agent control Chrome to debug web apps and run browser tests. It is in public preview across the CLI, the Copilot app, VS Code and JetBrains.

Read ›

Claude Code 2.1.327 refuses some symbolic links and lists broken setup in /status

Claude Code 2.1.327 tightens file and package permissions again and adds a /status list of broken agent, skill and hook settings. A short read for teams using AI coding tools on a live product.

Read ›

Start a project

Building with AI? Let's talk.

Tell us what you are building. We'll give you an honest view of the right approach.

Start a projectSee our work