AI Readiness Scorecard
Fifteen questions on whether your support operation is ready for AI, or whether AI would just scale your existing problems.
Data & Documentation
Knowledge Base
Tooling & Tech
Process & Governance
Leadership & Strategy
Your readiness score
—
Dimension breakdown
Your priority
Based on your own assessment. A directional guide, not an audit.
Nothing on this page is saved, sent or recorded.
Why most AI support rollouts disappoint
The failures rarely look like technology failures. The tool works, the integration finishes on time, and the deflection rate still lands nowhere near the number in the business case. Whatever went wrong happened long before anyone signed a contract.
AI in support multiplies what you already have. Point it at a clean knowledge base, a coherent ticket taxonomy and clear escalation rules and it performs roughly as advertised. Point it at documentation that fell behind the product two releases ago and it will answer confidently, incorrectly, at volume, to your customers.
Readiness is mostly unglamorous
The five dimensions here are documentation and data quality, knowledge base coverage, tooling integration, process governance and leadership alignment. None of them are exciting and none of them are what gets demoed to you. They're just what separates a pilot that gets extended from one that quietly gets dropped.
The most common pattern I see is a team strong on strategy and weak on knowledge. Genuine executive appetite, a clear business case, and a knowledge base nobody has audited in a year. That combination gets approved fast and disappoints slowly.
Where to start
Fix your lowest dimension before you evaluate a single vendor. Almost everything on the recommendation list is work you'd need to do whether or not you ever buy an AI tool, which is exactly why it's worth doing first. It improves the operation either way.
Common questions
How is the readiness score calculated?
Fifteen questions across five dimensions, each answer worth 0 to 3 points. Each dimension converts to a percentage of its own maximum and the five get averaged, so every dimension carries equal weight regardless of how many questions it holds. The result is a score out of 100.
Why does each dimension count equally?
The alternative weights them by question count, which is an accident of how the questions got drafted rather than a judgement about what matters. Knowledge base quality decides what an AI can actually answer. It shouldn't count for less than leadership alignment just because it takes fewer questions to assess.
What does a low score actually mean?
That AI would amplify problems you already have. These tools work from your existing data and documentation, so gaps in taxonomy, coverage or process consistency don't get smoothed over by automation. They get reproduced at scale and at speed.
Which dimension should I fix first?
The lowest one. The recommendations under your score are generated from your weakest dimension for that reason. Readiness behaves like a chain, and a strong strategy can't compensate for a knowledge base an AI can't use.
Do you store my answers or ask for my email?
No. Everything runs in your browser, nothing is transmitted or recorded, and there's no email form on this page.