Tool: AI Readiness Assessment

Discover

Is the organization ready for a production AI path?

56/ 100

Recommended path

Workflow redesign
  • Strategy
  • Data
  • Technology
  • People
  • Governance
  • Processes
  • Adoption
  • Security
  • Workflow maturity
  • Agent readiness

Workbench

Score the ten conditions that decide whether a first bet can ship.

Operating conditions
3 / 5
No ranked betsNamed workflow and a no-list
3 / 5
No usable baselineCase data is available
3 / 5
No system of recordWrite-back is possible
3 / 5
No ownerAn operator can change the process
2 / 5
No rulesApprover and kill switch exist
3 / 5
The path is tribalA completed case can be mapped
3 / 5
Old path will stayManagers will ask for the new artifact
3 / 5
Data cannot be usedApproved path is clear
3 / 5
Exceptions are the processVolume and quality are known
2 / 5
No control pointsAn outcome could be assigned

56/ 100

Recommended path

Workflow redesign
  • Strategy
  • Data
  • Technology
  • People
  • Governance
  • Processes
  • Adoption
  • Security
  • Workflow maturity
  • Agent readiness

Blockers

2The process is not yet clean enough for an agent, but it is worth mapping and changing one expensive path. Governance, Agent readiness
Adnan Boz
Adnan Boz

How I see it

Score the operating conditions, not the enthusiasm

Teams often score themselves on interest. Readiness is whether a workflow has an owner, data, a system of record, a control point, and a way to operate what you build.

Score ten conditions from 1 to 5. The result is a readiness score, the dimensions that will block a first bet, and a recommended path: adoption, workflow redesign, a contained pilot, or agent architecture.

A high score is not permission to skip discovery. A low score is not a reason to buy training for the whole company. It tells you which condition has to move before the economics matter.

Economic model

Readiness

average of ten operating conditions, with any score of 2 or below treated as a blocker

High value with two blockers is a later bet. Medium value with no blockers is often the first win.

When this is the wrong next step

Do not use this tool to decorate a decision.

  • You want a certification, not a diagnosis.
  • Nobody who runs the work will score the conditions honestly.
  • The only acceptable result is 'ready for agents.'
Adnan Boz

A useful next step

Bring the numbers into a conversation.

The tool is a first cut. If the result is material, we guide the implementation and stay hands-on through operations — or tell you when a simpler answer is better.

Discuss an AI opportunity