Measure
Is this useful enough to continue?
Learn to separate availability, quality and business value, compare economics on a common period and keep unknown benefits or costs explicit.
Required: your AI Operating Charter and either your own workflow or the fictional starter case. AI/GPT is not required for the core simulation.
Calculate the wrong-price rate and nominal time value, align the reporting period and decide what evidence would support stronger value claims.
A value review with checked arithmetic, assumptions, unknowns, trigger and next action.
This session uses supplied records and decision simulations. Keep the evidence label honest: designed, simulated, inspected or independently observed.
What you will leave with
A value review with a decision and unknowns. Add this to your AI Operating Charter.
Use your own workflow or the fictional quotation case. No coding or live customer action is needed.
Start with a decision · 10 minutes
The system ran all week and created 100 drafts. Ask what would make that good news or bad news for the business.
Bite A · Choose a signal that changes a decision
Question: What response follows an 8% wrong-price rate?
Input: S9 weekly: 8 of 100 drafts have wrong prices; 4 repair hours; available all week. Open the case cards.
Try · 12 minutes: Calculate the rate. Choose a quality trigger, owner and response. Explain why uptime and draft count cannot cancel the price defect.
Compare after you have decided
8/100 = 8%. Restrict sending, inspect the source/version issue and retest. Do not invent conversion or sales improvement.
Keep: Quality measure, trigger and action.
Bite B · Check the economics on a common period
Question: Does nominal time value cover operating cost?
Input: S9 monthly estimate: 60 hours at RM25; RM1,600 operating cost. Open the case cards.
Try · 12 minutes: Calculate nominal time value and the difference. Name two unknown benefits/costs and identify the period before comparing figures.
Compare after you have decided
60 × RM25 = RM1,500; RM1,500 − RM1,600 = −RM100 before other benefits/costs. Weekly repair hours cannot be subtracted from monthly value without alignment.
Keep: Period, assumptions, cost and decision.
Bite C · Separate capacity from cash
Question: When would “saved hours” become cash savings?
Input: A manager calls the 60 hours cash savings. Open the case cards.
Try · 12 minutes: State what evidence would support that claim. Name a possible non-cash benefit, how to observe it and a next review date or trigger.
Compare after you have decided
Released time may improve capacity or response time. Cash savings need an actual spending reduction or other defensible financial evidence.
Keep: Value hypothesis and next observation.
Put it together · 20 minutes
The review uses checked arithmetic, aligned periods, explicit unknowns and an action; it does not equate activity with value. Revise the weakest item and have a partner use it. Keep the result and one limitation.
Close your question · 10 minutes
Return to Is this useful enough to continue? and your own opening question. Answered, partly answered or still open? Show the evidence. Save the updated AI Operating Charter. A reasoned decision to keep a step human-led is useful work.
Common trap
Do not present nominal labour value as realised savings or infer causal business impact from this training packet.
Can a negative estimate still justify a pilot? Possibly, with a named additional benefit to test and a bounded cost; do not claim that benefit already exists.
Where this appears in real work
Gaia Eats/manufacturing: compare quality, rework and outcome cost; use synthetic data until a shareable real record is selected. The class uses the fictional packet; selected project demonstrations and technical walkthroughs can add detail later.