Thinking / Data readiness
Is your data ready for AI? A practical checklist
A business-led assessment of whether the information behind an AI product is reliable, usable and controlled enough to begin.
The answer in brief
- What it is
- A business-led assessment of whether the information behind an AI product is reliable, usable and controlled enough to begin.
- Best suited to
- Leaders who need to translate this issue into an investment, workflow, governance or capability decision.
- What useful progress looks like
- A user asks a natural-language question and receives a confident answer in seconds. Behind that interaction, the system may retrieve fragments from policies, contracts, customer records, messages and operational systems. Each source has an owner, version, permission and context. If those conditions are unclear, fluency can disguise uncertainty. Growing organisations often assume that readiness requires an expensive enterprise data programme. The more useful question is whether the information required by a specific product can support a reliable and defensible outcome.
The question leaders ask
How do we know whether our data is ready for AI?
Your data is ready enough when the required sources are accessible, current, understood, permitted, traceable and sufficiently reliable for the consequence of the use case. Readiness should be tested against one product or workflow.
01
Data can make AI look deceptively simple
A user asks a natural-language question and receives a confident answer in seconds. Behind that interaction, the system may retrieve fragments from policies, contracts, customer records, messages and operational systems. Each source has an owner, version, permission and context. If those conditions are unclear, fluency can disguise uncertainty. Growing organisations often assume that readiness requires an expensive enterprise data programme. The more useful question is whether the information required by a specific product can support a reliable and defensible outcome.
02
Test seven conditions before building
The checklist should be completed with the business owner, product lead and relevant technical or risk specialists. A green answer means evidence exists. Amber means a defined remediation is possible. Red means the proof should pause or narrow.
Purpose
Can the team state which decision, action or user need the information supports?
Availability
Can approved systems access the required structured and unstructured sources?
Currency
Are updates, superseded documents and retention cycles managed predictably?
Meaning
Do important terms, entities and definitions remain consistent across sources?
Quality
Are material omissions, duplication and known errors visible and manageable?
Permission
Are privacy, confidentiality, licence and access conditions understood?
Traceability
Can an output be connected to the source material and version that influenced it?
Source: McKinsey, AI data readiness03
Define good enough in context
Perfect data is an unhelpful ambition. A low-consequence internal drafting assistant can tolerate more uncertainty when users verify the result. A system recommending a customer action, interpreting contractual obligations or influencing employment decisions requires stronger source integrity, testing and human authority. Teams should document acceptable error, required freshness, excluded sources and the circumstances that trigger escalation. This creates a practical quality threshold rather than an abstract cleansing programme.
04
Use the proof to improve the foundation
A bounded product creates evidence about which information problems genuinely constrain value. Retrieval tests can reveal outdated documents, conflicting definitions and missing ownership. User feedback can show where context is absent or an answer needs explanation. Treat these discoveries as a prioritised data backlog. Solve the shared issue once where possible, assign an owner and make the improved source reusable. This connects foundation investment to a visible business need and reduces the temptation to create another isolated pipeline for every new idea.
FAQ
Questions leaders ask.
Direct answers to the questions that commonly shape an initial conversation.
01Do we need a data warehouse before using AI?+
No. The required architecture depends on the workflow, sources, scale and control needs. Many proofs can begin with governed access to a bounded set of reliable information.
02Who owns AI data readiness?+
Business owners define the required outcome and meaning. Data and technology teams establish reliable access and controls. Product leadership connects both sides around the use case.
03Can unstructured documents be used safely?+
Yes, where versions, permissions, sensitivity, provenance and retrieval quality are managed and outputs receive appropriate human review.
04What should we fix first?+
Fix the information issue most likely to change answer quality, user trust or the ability to prove value in the priority workflow.
Latest from the blog
Useful ideas for the decision in front of you.
Plate Spinner's PerspectiveFF＝01
Plate Spinner's Perspective · 8 minLearning to disagree with AI
Why a confident, persuasive answer should be the start of leadership judgement, not the end of it.
Read the articlePlate Spinner's PerspectiveFF＝02
Plate Spinner's Perspective · 7 minWhat does a Chief of Staff do in an AI business?
The operating role between technical possibility, commercial pressure and executive attention.
Read the articlePlate Spinner's PerspectiveFF＝03
Plate Spinner's Perspective · 7 minChief of Staff vs COO: where does the work split?
A practical distinction between enterprise operations and the executive agenda that cuts across them.
Read the articleApply the thinking
Assess the readiness behind your AI opportunity
Bring the live decision, workflow or commercial pressure. We will help translate the idea into a focused next step.
Discuss the implication