How do we know if we're actually ready for AI?
There is board pressure to do something with AI, several pilots have demoed well and deployed never, and nobody can articulate what is actually blocking production.
Readiness comes down to six things: data quality, process definition, governance, integration, adoption capacity and measurement. Most companies that stall fail on the same two — data quality and process definition — and both are prerequisites rather than phases. An AI system reasoning over inconsistent data produces confident wrong answers faster than a person would, and you cannot automate a process nobody has written down. Neither of those requires buying anything to fix.
The practical test takes an afternoon and no budget. Pick the single workflow you would automate first, then try to write it down as a sequence of steps a new hire could follow. If you cannot, the process is not defined yet. Then pull fifty records the workflow depends on and count how many have missing, stale or contradictory fields. Above roughly ten percent, fix the data before buying anything — an agent will inherit every one of those errors and apply them at speed.
What makes an organisation not ready
- Pilots are scoped to demo. A clean, narrow workflow proves very little about production, where variance and edge cases live.
- Data quality is assumed. Models inherit your data problems and express them with more confidence than a human would.
- No baseline exists. Without a before number, nobody can prove the pilot worked, so it cannot be funded further.
- Nobody owns the outcome. AI initiatives sponsored by everyone and owned by no one stall at exactly this point.
How to build AI readiness
-
01
Run the free AI Transformation Readiness assessment. Six dimensions, four minutes, ungated — it will name your weakest dimension.
-
02
Pick one high-volume, low-variance workflow. Resist the interesting problem; choose the boring frequent one.
-
03
Baseline it numerically before you build anything. Time taken, error rate, cost per unit.
-
04
Classify your data before choosing a model. Which classes may leave your boundary decides the architecture.
-
05
Name one accountable owner with authority to stop the project. Unstoppable projects are unmeasurable projects.
The prerequisites are unglamorous and nobody budgets for them after an exciting demo. Data cleanup and process definition are what actually determine whether an AI programme reaches production, and they are the two things a board presentation never covers. The organisations that succeed treat AI readiness as an operations problem with an AI application on top — not as an AI problem.
More on AI readiness
On a well-chosen workflow with a real baseline, six to twelve weeks. The constraint is almost never model capability — it is data readiness and the speed of decisions on your side.
Usually not at this stage. Most early value comes from applying existing models to well-defined problems, which is an integration and process discipline more than a research one.
Then that is the project, and it is worth knowing before spending on AI tooling. Data cleanup has a return independent of AI — better reporting, better forecasting, less manual reconciliation.
More on AI Transformation
Still not sure this is your problem?
A 20-minute fit check. We will tell you if it is something you can fix without us — that happens often enough that we lead with it.
Book a Strategy Session →Browse every answer.
The full Solution Desk — specific questions, straight answers, no gate.
Open the Solution Desk →