readiness_checklist.runliveillustrative

Problem and value

Answer for how things are today, not how they are planned to be.

Have you picked one specific task or decision you want AI to help with, rather than "use AI somewhere"? Yes No

Do you know what that task costs today, in hours, errors, delays or lost revenue? Yes No

Have you agreed what result would count as success, such as an accuracy level or hours saved, and how you would test it? Yes No

Data

Is the data this use case needs held digitally, rather than on paper or in people’s heads? Yes No

Is that data reasonably clean and consistent, and can it be brought together without weeks of manual work? Yes No

If the model needs labelled examples, such as tagged images, categorised tickets or approved answers, do you have them or know who will create them? Yes No

People and process

Is there a named person in the business who owns the project and will make decisions about its results? Yes No

Will the people who do the task today give time to test the AI and point out where it gets things wrong? Yes No

Have you decided what happens when the AI is unsure or wrong, including who reviews its output and how errors are corrected? Yes No

Technology and governance

Can the systems involved be reached through APIs or regular exports, so AI can read from and write to them? Yes No

Do you have a written policy on what data may be sent to AI models and where it may be stored? Yes No

Does someone in IT or security have time to review access, logging and security for a new AI system? Yes No

$ runs in your browsernothing stored or sentassumptions ↓

about this tool

How it works and what it assumes.

Each "yes" scores one point. The questions cover the four areas where AI projects most often stall: a clear problem and payoff, usable data, the right people and process around the system, and the technology and governance to run it safely.

9 – 12: ready to pilot. Pick the use case with the clearest payoff and measure it against a baseline.
5 – 8: ready for a focused proof of concept on your strongest use case, fixing gaps along the way.
0 – 4: start with groundwork. A short discovery and data review before any build will save money later.

Every "no" is listed with the step that addresses it. Data gaps are the most common; our data analytics consulting, data annotation and labelling and business intelligence teams handle that groundwork. When you are ready to put a figure on a build, try the AI project cost estimator. Nothing you enter leaves your browser.

explore

More free tools

Free consultation

Walk through your results with us

Enquiry formFields marked * are required.

Is your business ready for an AI project?