Calculate a fictional workflow comparison and explain which benefits are measured, assumed, or still unknown.
Choose a finished unit of work
A finished unit might be an approved briefing or a resolved request. Define completion before measuring how quickly the first draft appears.
The quality threshold must stay comparable across approaches. Otherwise, accepting weaker work can make a workflow look faster without making it better.
A baseline record captures the existing process under similar conditions. Include input preparation, drafting, checking, correction, and final handling.
Count all the work
Active effort includes time people spend preparing inputs and fixing mistakes. Record these stages separately so a faster draft cannot hide slower review.
Elapsed time also includes waiting between steps. Keep it separate from active effort because reducing one does not necessarily reduce the other.
Ongoing overhead includes maintaining sources, checking changes, and training reviewers. Record setup effort separately from recurring work to avoid confusing their effects.
Make the comparison fair
Comparable cases should resemble the work the process will actually receive. Include difficult inputs instead of selecting only convenient demonstrations.
A practical comparison can assign similar cases across approaches and reviewers. Avoid reusing an answer someone remembers when timing the second approach.
A small pilot provides a limited observation, not a universal prediction. Describe sample size, case mix, and anything that changed during the comparison.
Calculate a fictional result
This invented pilot uses two comparable batches of twenty briefs. Every brief is accepted after correction against the same stated standard.
The measured effort falls from 200 to 140 active minutes per batch. That is sixty minutes less, or thirty percent of the baseline effort.
The recurring overhead adds forty minutes per batch in this scenario. Therefore, the net measured reduction becomes twenty minutes, or ten percent.
| Stage | Manual baseline | AI-assisted |
|---|---|---|
| Prepare inputs | 40 min | 40 min |
| Draft | 100 min | 20 min |
| Review and correct | 60 min | 80 min |
| Task effort subtotal | 200 min | 140 min |
| Additional recurring AI upkeep | 0 min | 40 min |
| Total measured effort | 200 min | 180 min |
| Effort per accepted brief | 10 min | 9 min |
Keep benefits and assumptions separate
A time reduction creates capacity only if that time can be used. It does not automatically reduce payroll or create additional revenue.
Unmeasured costs remain open questions. The example excludes setup time, service charges, and any later corrections, so it is not a financial return calculation.
A useful recommendation states both the result and its limits. Continue testing if the operational benefit looks promising but quality or overhead remains uncertain.
Practice: challenge the improvement claim
Your exercise changes recurring upkeep from forty to eighty minutes per batch. Keep all other fictional values unchanged and recalculate the result.
Your recommendation should explain whether faster drafting still improves the complete process. Then name one missing measurement that could change your decision.
A strong answer distinguishes gross task savings from net operational benefit. Show the denominator and avoid treating capacity as cash savings.
Compare with the worked answer
Worked answer: The AI-assisted total becomes 220 minutes: 140 task minutes plus 80 upkeep minutes. That exceeds the 200-minute baseline by 20 minutes, or 10%. Faster drafting no longer produces a net effort reduction in this example. Investigate upkeep and later corrections before expanding.
Use the method on your own task
Editable worksheet: Download the Markdown template and open it in a text editor. Use fictional or approved information; no account or paid service is required.
Download the worksheet (.md) ↓What to carry forward
Calculate a fictional workflow comparison and explain which benefits are measured, assumed, or still unknown.
Show your reasoning with the exercise above. Keep the evidence, the decision, and the remaining uncertainty together.