Free tool from SherwoodCX

AI Readiness Scorecard

Fifteen questions on whether your support operation is ready for AI, or whether AI would just scale your existing problems.

0 / 15

Data & Documentation

Question 1 of 15
How well documented are your support workflows and processes?
Think: do new agents have written guides to follow, or do they learn by watching others?
Question 2 of 15
How structured is your ticket tagging and categorisation?
AI needs clean, consistent data to learn from. Random or inconsistent tags break training data.
Question 3 of 15
How much historical ticket data do you have available?

Knowledge Base

Question 4 of 15
How mature is your customer-facing knowledge base or help centre?
Question 5 of 15
How consistent is the tone, format, and quality of your knowledge base articles?
AI-generated answers are only as good as the source content they pull from.

Tooling & Tech

Question 6 of 15
How well integrated are your core support tools with each other?
e.g. helpdesk ↔ CRM ↔ product data. Disconnected tools create data silos that block AI effectiveness.
Question 7 of 15
What is your current helpdesk or support platform?
Question 8 of 15
How would you describe your team's comfort level with new tools and technology?

Process & Governance

Question 9 of 15
Do you have clear quality standards and a QA process for support interactions?
Question 10 of 15
How clearly defined are your escalation paths and routing rules?
AI needs to know when to hand off to a human. Unclear escalation paths cause bad AI experiences.
Question 11 of 15
How consistently does your team use macros, canned responses, or templates?

Leadership & Strategy

Question 12 of 15
How clear is leadership's understanding of what AI can and cannot do in support?
Question 13 of 15
Do you currently measure the business impact of your support team?
e.g. churn attribution, revenue influenced, cost per resolution, deflection rate
Question 14 of 15
Is there a designated owner or champion for your AI / automation initiative?
Question 15 of 15
How would you describe your organisation's overall readiness for change right now?

Your readiness score

0
out of 100

Dimension breakdown

Every dimension weighted equally

Your priority

Start with your weakest dimension

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.