For engineering leaders

Why it matters

Usage numbers are easy to collect and easy to misread. High usage can hide shallow skills and practices that haven't changed, and a dashboard can't see a tool someone pays for themselves. To know whether AI is making your engineering better, you need skill, practice and results, measured next to usage.

• Mostof about 600 engineering leaders call token-usage metrics ineffectiveLeadDev AI Impact Report, 2026
• 90%+of the companies MIT NANDA studied have workers using personal AI tools for work, often without IT knowingMIT NANDA, The GenAI Divide, 2025

Questions it answers

For CTOs, VPs of Engineering and the leads under them: the questions usage numbers leave open, and where the report answers each one.

• Are engineers getting results from AI, or just using it?

  The 360° has its own Delivery & Engineering questionnaire, for engineering, QA, DevOps, design and data. It places every team by Usage against Impact and flags practices that rest on one person.

  360° AI Maturity Assessment

• How good are we, really?

  A quiz sets the level: three questions in each of ten areas. Where confidence and results part ways, the report marks a blind spot or a hidden strength. n8n is live; Claude, with extra questions for Claude Code and the API, is coming soon.

  n8n Skills Assessment · Claude Skills Assessment

• Which practices is each project missing?

  One project's practices in 14 areas, routed by role, with a score out of 10 per area and what's unmet, most important first.

  Software Quality Management (SQM)

• Are we keeping up with what shipped this quarter?

  Engineers pick the Technical path: each quarter's developments for technical work, from Not aware to Regular use, each set against the stage we expect the market to be at.

  AI Market Pulse

• What stops engineers from using AI more?

  Barriers with who can fix each one, and a heat map by department or by how much people use AI.

  AI Adoption Barriers

• Which tools do people actually use?

  Who has hands-on experience with each of 125 AI tools and models, including the ones the company doesn't provide.

  AI Skills Distribution

Next to your dashboards

Engineering-intelligence platforms and vendor dashboards count activity: seats, prompts, suggestions accepted. They can't see skill, what people find hard, or a tool someone uses on their own account. AI Adoption Score asks people directly, so there's nothing to install, no repository or tool to connect and no admin access to grant. It runs alongside the dashboards you already have.

Where to start

1. Run the 360° AI Maturity Assessment on the Delivery & Engineering questionnaire: about 15 minutes per person.
2. Check skills with a quiz. The n8n Skills Assessment for people who build workflows takes about 20 minutes; the Claude Skills Assessment follows when it launches.
3. Review each project with Software Quality Management. One run per project, answered by its team, with the findings as a PDF.
4. Run AI Market Pulse every quarter on the Technical path: about 8 minutes per person.

What it won't tell you

Anything beyond what people report, though we check it where we can. The skills assessments score a quiz, not a self-rating. The 360° sets each self-rating against what peers can see, and one question on its engineering questionnaire lists project configuration files, three of which don't exist: picking one flags the response. None of this catches everything.

Whether one engineer is qualified. Three questions per area show where a team stands and where to start. It isn't an exam, or an official n8n certification.

Who's right about a practice. In SQM, several people answering the same practice can expose a disagreement, but not settle it, and there's no benchmark from other companies.

Anything certain about a small team. Results show from the first answer, with no minimum group size, so a team of three is three people's answers.