Module 5 · Lesson 3 of 4

Decide whether to adopt, adapt or stop

Use the pilot evidence to make a bounded decision instead of treating every trial as a success.

Go to the exercise

What you will leave with

A justified next decision with a clear scope and evidence that could change it.

Understand

The idea behind the work

Adopt means the tested workflow is useful within a stated scope and has an owner. Adapt means a specific change deserves another test. Stop means the approach is not useful or acceptable under the current conditions. These are operational decisions, not grades for the person running the experiment.

Compare the evidence with the success conditions established before the pilot. Check material failures and unresolved requirements before celebrating speed. State what the evidence covers and what it does not. If you change the workflow after observing a failure, test the corrected version again; the earlier evidence does not automatically validate the new one.

Inspect a worked example

Synthetic example: a fast but unreliable draft

This is a fictional teaching example, not a client result or a claim about Joseph Carroll's past work.

The scenario describes invented pilot observations.

The starting input

Three sample drafts are faster to produce, but two contain unsupported product claims and require extensive correction.

Work through it

  1. Record that the factual-fidelity condition was not met.
  2. Choose adapt: restrict the source boundary and require a claim map before review.
  3. Rerun the failed cases and a fresh case; keep publication manual until the checks pass.

The result

A targeted next experiment with a clear reason, rather than premature rollout.

A mistake to recognize

Adopting because the speed result looks attractive while ignoring claim failures.

How to recover

Treat the unmet quality condition as material and test the proposed correction.

Adapt the method

What changes in your business?

The same method can serve different settings. Data, allowed actions and operational support matter more than headcount alone.

Owner-led business or small team

Stopping a low-value experiment preserves time for a better task.

Department or growing team

Have the intended recipient and process owner participate in the decision.

Large or complex organization

Separate pilot acceptance from broader procurement, access or release approvals that have not yet occurred.

Try it, then inspect the result

Make the pilot decision

Choose adopt, adapt or stop. Cite the observations, state the approved scope, unresolved conditions, next owner and what evidence could change your decision.

Record the pilot decision

Use a sample or a description you are allowed to share. Your answer stays in this browser when storage is available; export it to keep a separate copy. 0 / 6,000 characters.

Check your reasoning

The workflow is accurate for public sample documents, but internal-data handling is unresolved. What adoption scope is supported?

Inspect your own output

Check each statement after you look at your answer. These evidence checks apply to this visit; your answer and completion status can be saved.

To mark completion, write an answer, check the reasoning scenario and inspect every evidence statement. You can continue to another lesson at any time.

Self-reported completion does not certify competence. Open My AI Plan to review your saved work.

Sources and further practice

The lesson is Carroll Consulting's teaching synthesis. These sources support the underlying capabilities and concepts; check the current product, plan and configuration before applying them.

You can move between lessons freely. Revisit the checks whenever your answer, evidence or business context changes.