Enterprise AI Operating Capability

Guide to Enterprise AI Operating Capability
Start with One Workflow

Connect stakeholder value to workflows, responsibilities, human–AI work and evidence in a real pilot.

Discuss a Pilot 简体中文

Methods and Further Reading

The Operating Problem
Start with Actual Delivery

Start with a quotation or delivery review. Identify who benefits, who bears the risk and how the result reaches the next step.

Name the operating goal and the people who use the result. Choose one workflow with observable outputs and manageable risk.

Trace a real deliverable back through the workflow. Record its inputs, owner, acceptance criteria and downstream recipient.

Manage Handovers

Use the same process definition for role responsibilities, playbooks and performance criteria. Describe who handles exceptions, what authority they have and how decisions are reviewed.

Discuss a Pilot
After producing the work, check whether the recipient can actually use it.
Record Actual Results

( 02 )

Who Should Be Involved
Start with Actual Delivery

Record execution, acceptance, downstream use and whether the overall goal improved within the observation window.

Goal

Compare quality, time, rework and cost before and after the change. Track manual intervention and exceptions alongside generation speed.

Priority Workflow

An AI skill brings together the knowledge, tools and rules for a task. The human playbook and the skill share acceptance criteria and specify when a person must take over.

Ownership

Set the scope, owner, baseline and stop conditions before the pilot. Continue observing if evidence is unclear, and roll back if risk grows.

Process Owner

Ask another colleague to repeat the task using the playbook or skill. Note gaps that still need verbal explanation. Expand only when the process can be repeated reliably.

Separate observed causes from hypotheses in the review. Write tested changes back into SOPs, responsibilities, playbooks and skills, with version ownership and rollback.

Discuss a Pilot

( 03 )

How to Map the Workflow
Start with Actual Delivery

Use four levels to describe results: business outcomes, processes, functional responsibilities and individual actions. Then verify that the work at each level contributes to the overall goal.

Agree on interfaces, delivery, acceptance and dispute handling with external partners. Each organization can keep its own process while sharing a standard at the handover.

Bring a Business Problem

Possible deliverables include a process map, a human–AI responsibility matrix, acceptance criteria or a pilot review. Confirm deliverables and timing after discussing scope. Define what counts as acceptable first.

V2.0 is an interim research framework. Test implementation conditions and business impact on real work, then use the evidence to decide whether to expand.

Log waiting and rework separately so the right owner can address each.

EXAMPLE 02 · PILOT OBSERVATION

( 04 )

Delivery, Authority and Review
Start with Actual Delivery

Manage Handovers

Set a Human Fallback

Name the operating goal and the people who use the result. Choose one workflow with observable outputs and manageable risk.

Record Actual Results

Trace a real deliverable back through the workflow. Record its inputs, owner, acceptance criteria and downstream recipient.

Test Repeatability

Use four levels to describe results: business outcomes, processes, functional responsibilities and individual actions. Then verify that the work at each level contributes to the overall goal.

Use the same process definition for role responsibilities, playbooks and performance criteria. Describe who handles exceptions, what authority they have and how decisions are reviewed.

Discuss a Pilot

( 05 )

Scope and Deliverables
Start with Actual Delivery

Check two things at handover: whether the upstream output passes, and whether the next step is ready. Recording both shows whether work needs correction or is waiting on a dependency.

Give the method to someone who did not design it and see whether the result still passes.
EXAMPLE 03 · PILOT OBSERVATION

Passing upstream review does not mean the next step has authority or resources.

EXAMPLE 04 · PILOT OBSERVATION

Knowing when to stop is part of a good pilot.

EXAMPLE 05 · PILOT OBSERVATION

After producing the work, check whether the recipient can actually use it.

EXAMPLE 06 · PILOT OBSERVATION

Log waiting and rework separately so the right owner can address each.

EXAMPLE 07 · PILOT OBSERVATION

See Use Cases

( 06 )

Frequently Asked Questions

Compare on the Same Basis

  1. Define what counts as acceptable first. Compare quality, time, rework and cost before and after the change. Track manual intervention and exceptions alongside generation speed.
  2. Define what counts as acceptable first. An AI skill brings together the knowledge, tools and rules for a task. The human playbook and the skill share acceptance criteria and specify when a person must take over.
  3. Define what counts as acceptable first. Set the scope, owner, baseline and stop conditions before the pilot. Continue observing if evidence is unclear, and roll back if risk grows.
  4. Define what counts as acceptable first. Ask another colleague to repeat the task using the playbook or skill. Note gaps that still need verbal explanation. Expand only when the process can be repeated reliably.
  5. Define what counts as acceptable first. Separate observed causes from hypotheses in the review. Write tested changes back into SOPs, responsibilities, playbooks and skills, with version ownership and rollback.

1

Workflow
to Pilot

4

SOP
Levels

2

Handover
Checks

4

Evidence
Checks

( 07 )

Bring One Workflow to Discuss
Start with Actual Delivery

Record execution, acceptance, downstream use and whether the overall goal improved within the observation window.

Compare quality, time, rework and cost before and after the change. Track manual intervention and exceptions alongside generation speed.

An AI skill brings together the knowledge, tools and rules for a task. The human playbook and the skill share acceptance criteria and specify when a person must take over.

Set the scope, owner, baseline and stop conditions before the pilot. Continue observing if evidence is unclear, and roll back if risk grows.

Bring a brief outline of one workflow.

( 08 )

Connect the Steps
Start with Actual Delivery

Manage Handovers

Separate observed causes from hypotheses in the review. Write tested changes back into SOPs, responsibilities, playbooks and skills, with version ownership and rollback.

Give the method to someone who did not design it and see whether the result still passes.

Set rules for data scope, access and retention before using sensitive information. Escalate to an authorised person when access, inputs or risk limits are unclear.

( 09 )

Make the Method Reusable
Start with Actual Delivery

Agree on interfaces, delivery, acceptance and dispute handling with external partners. Each organization can keep its own process while sharing a standard at the handover.

Bring one live workflow to the discussion. Describe the desired outcome, the bottleneck, the roles involved and the samples available for comparison.

Confirm Fees After Scoping

Possible deliverables include a process map, a human–AI responsibility matrix, acceptance criteria or a pilot review. Confirm deliverables and timing after discussing scope.

Discuss a Pilot

( 10 )

Keep a Record and Review
Start with Actual Delivery

Make the Method Reusable

Use four levels to describe results: business outcomes, processes, functional responsibilities and individual actions. Then verify that the work at each level contributes to the overall goal.

Define what counts as acceptable first. Name the operating goal and the people who use the result. Choose one workflow with observable outputs and manageable risk.

Trace a real deliverable back through the workflow. Record its inputs, owner, acceptance criteria and downstream recipient.

( 11 )

Frequently Asked Questions

Give each step a purpose and an output. When connecting steps, name who consumes the earlier output and under what conditions the next step may start.

How Do We Choose the First Workflow?

Name the operating goal and the people who use the result. Choose one workflow with observable outputs and manageable risk.

Do We Need New AI Tools First?

Start with tools you already have after checking data permissions and acceptance standards. Evaluate purchases only if there is a genuine capability or integration gap.

How Should Sensitive Data Be Handled?

Set rules for data scope, access and retention before using sensitive information. Escalate to an authorised person when access, inputs or risk limits are unclear.

How Do We Know the Pilot Helped?

Compare quality, time, rework and cost before and after the change. Track manual intervention and exceptions alongside generation speed.

When Is It Ready for a Wider Rollout?

Ask another colleague to repeat the task using the playbook or skill. Note gaps that still need verbal explanation. Expand only when the process can be repeated reliably.

What Should We Bring to the First Discussion?

Bring one live workflow to the discussion. Describe the desired outcome, the bottleneck, the roles involved and the samples available for comparison.

How Are Scope and Fees Agreed?

Possible deliverables include a process map, a human–AI responsibility matrix, acceptance criteria or a pilot review. Confirm deliverables and timing after discussing scope.

Define Authority

Use the same process definition for role responsibilities, playbooks and performance criteria. Describe who handles exceptions, what authority they have and how decisions are reviewed.

Discuss a Pilot