Is there an AI readiness checklist you can use yourself?

Yes. Public checklists can organise questions about data, processes, tools and people. Checkmarks record your answers; they do not by themselves certify technical or legal readiness for a particular task.

Which categories can you check yourself?

Pro-How’s GitHub repository organises questions into five areas. You can start with these, recording evidence or missing information alongside the answers:

Data and information: where is current information and who maintains it?
Processes and documentation: are steps and exceptions described?
Tools and software: what do you use and how can it connect if needed?
Team and culture: who will use the change and what help do they need?
Goals and expectations: which task do you want to improve and how will you check the outcome?

Pro-How defines its own percentage categories: 90–100% for starting implementation, 70–89% for closing gaps, 50–69% for groundwork and below 50% for basic conditions. These are provider definitions, not independently confirmed readiness thresholds or certification. A high overall score does not remove a particular problem with access, quality or data use.

What are Amazon Web Services’ five steps?

AWS’s guide for small and medium businesses proposes these steps, briefly paraphrased here:

Data: record sources, owners, updating and access rights.
Use: choose a bounded task with a checkable outcome and human review.
Technology: check necessary connections, permissions and current systems’ limits.
Governance: record rules for privacy, protection and output review.
Measurement: define a baseline, metric and quality check.

AWS suggests a 30–60 day pilot and an illustrative 6–8 week readiness sprint. This is a provider plan, not a measured standard for every business or a deadline for the AI Opportunity Map. All data does not have to move into one system. APIs or webhooks concern necessary integration; they are not a condition of every possible AI use.

Which nine topics offer a closer look?

RTS Labs covers nine dimensions. Use these as discussion topics suited to your size and task:

Strategy: connection to a particular business objective.
Data: availability, quality and context.
Infrastructure: resources and protection needed.
Skills: knowledge and people responsible.
Governance: applicable rules, accountability and controls.
Change: learning, adoption and ways of working.
Finance: budget, costs and a benefit criterion.
Process: AI’s place in daily work.
Maintenance: monitoring, changes and expansion.

These topics do not mean every small business must hire a machine learning team or buy a new cloud environment. Check what the chosen use needs. A list of desired conditions is not evidence that those conditions have been met.

What needs checking beyond the boxes?

A checklist can also help with priorities. For a decision in your business, however, connect answers to particular work:

Is the data sufficient for this task, including difficult cases?
Which tasks cost time, what benefit do you seek and what constrains delivery?
Are there undocumented workarounds to discuss with the people doing the work?

Record what is established by an example and what is assumed. A conversation may clarify gaps; it does not find them automatically just because someone external is involved. You need a checkable method and evidence appropriate to the scope.

Questions

Is a checklist useful before speaking to a consultant?

Yes. Mark what is known, what is unclear and the source of important answers. The conversation can then start with particular questions. Completing a checklist does not oblige you to buy consulting or implementation.

Which of the three checklists should we choose?

You can use AWS’s five steps for a short start. Pro-How organises questions by operational area, while RTS Labs offers a broader set of topics. Choose for the task and the people answering; no one option is established as best for every small business.

Does a checklist replace a professional assessment?

Self-assessment and independent checking are different. Ask what evidence is reviewed, what limits the method has and what falls outside scope. Neither an overall percentage nor a paid conversation alone establishes readiness.

Sources

Connect the gaps identified to the order of tasks.

The AI Opportunity Map orders tasks by the hours they cost you today; for the first three it gives a proposed tool, requirements, an owner and a build-effort range in hours, without a build price. A completed checklist can prepare you for the conversation. The result is a prioritised task list, not readiness certification or a promise of implementation.

Ask about the AI Opportunity Map

or write to contact@ergostrat.com