Workflow audit and use-case selection
Choose a useful AI task and count the work it really saves.
In this lesson · A1 of 8
All lessons in this track →Progress for this track is saved in this browser only. No completion badge is issued. View your learning.
Start with the idea
Choose a useful AI task and count the work it really saves.
Pick a task with a clear output and a person responsible for it. Include checking time when you compare effort. Consider whether a simple checklist would do the job better.
Baseline: The current way of doing the task, used for comparison.
See the example and check your understanding
A made-up teaching example
In the made-up meeting example, manual drafting and review take 150 minutes a week. AI-assisted drafting and review take 130. The improvement is 20 minutes, and only if quality is equal.
Ask yourself
Should you count drafting time but leave out review time?
Show the explanation
No. Count the whole task. A faster process is not useful if it produces unacceptable work.
Start with these checks
- Include review in both comparisons.
- Respect tasks reserved for human decisions.
- Consider a non-AI alternative.
Use the different practice case below and check every requirement in its full instructions.
The video explains the main idea. Use the sources, instructions and checks below to complete the practice.
Captions appear here during playback.
Workflow audit and use-case selection
Choose a useful AI task and count the work it really saves.
Video playback is unavailable. Read the descriptive transcript below, go directly to practice, or open the video file.
Source-based teaching example · Fictional data · Reviewed 2026-10-03. Microsoft Mark · locally synthesized. On a small screen, use fullscreen for diagram details or read the transcript.
Read the descriptive transcript
0:00 · What you will learn
Choose a useful AI task and count the work it really saves.
What appears on screen
Choose a useful AI task and count the work it really saves.
0:06 · The idea
Pick a task with a clear output and a person responsible for it. Include checking time when you compare effort. Consider whether a simple checklist would do the job better.
What appears on screen
Pick a task with a clear output and a person responsible for it. Include checking time when you compare effort. Consider whether a simple checklist would do the job better.
0:20 · One useful term
Baseline. The current way of doing the task, used for comparison.
What appears on screen
The current way of doing the task, used for comparison.
0:27 · A worked example
In the made-up meeting example, manual drafting and review take 150 minutes a week. AI-assisted drafting and review take 130. The improvement is 20 minutes, and only if quality is equal.
What appears on screen
In the made-up meeting example, manual drafting and review take 150 minutes a week. AI-assisted drafting and review take 130. The improvement is 20 minutes, and only if quality is equal.
0:45 · Your turn to think
Should you count drafting time but leave out review time? Pause and decide before continuing.
What appears on screen
Should you count drafting time but leave out review time? Pause the video if you need more time.
1:02 · Why this answer works
No. Count the whole task. A faster process is not useful if it produces unacceptable work.
What appears on screen
No. Count the whole task. A faster process is not useful if it produces unacceptable work.
1:12 · Check 1
Include review in both comparisons.
What appears on screen
Include review in both comparisons.
1:17 · Check 2
Respect tasks reserved for human decisions.
What appears on screen
Respect tasks reserved for human decisions.
1:23 · Check 3
Consider a non-AI alternative.
What appears on screen
Consider a non-AI alternative.
1:28 · Now try a different case
Open the lesson and try its different practice case. Use every check in the full task. Watching alone does not mark the lesson complete.
What appears on screen
Open the lesson Try the new case Check and revise The full task and its requirements are in the linked lesson.
Watching explains the example. Complete the lesson practice to check what you can do. Explore all visual lessons →
Before you start
Prerequisite: F8 source-bounded field exercise and O2 authorization. Offline/no-cost route: a notebook, calculator and supplied synthetic facts. Optional tools must already be authorized. No real workplace data, paid API, new account or external action is required.
What you will be able to do
- Map inputs, decisions, outputs, owners, controls, and failure costs.
- Score value, feasibility, sensitivity, reversibility, and evidence.
- Create a defensible no-go list.
Part 1 Audit the current work before proposing AI
Map inputs, transformations, queues, handoffs, review and outputs. Measure total effort, quality and waiting separately: faster drafting can increase correction burden. Define a baseline before intervention and include a simple non-AI alternative such as a checklist or deterministic rule. Rank use cases by value, reversibility, evidence availability, data permission and consequence. Technical novelty is not a value metric. A no-go decision is useful when scope is unknown or reliable checking is impossible. Model and tool selection follows the selected task; it cannot repair a workflow with no owner or source of truth.
Part 2 Worked synthetic example
Synthetic inventory: meeting summary, 10/week, manual draft 12 minutes + review 3; request sorting, 40/week, 4 minutes each; personnel eligibility, 5/week, reserved human authority. Meeting summary baseline is 150 minutes/week. Candidate draft 5 + review 8 is 130 minutes/week: only 20 minutes improvement, assuming equal quality. Eligibility automation is excluded. A checklist may solve sorting without generation. These are invented measurements, not promised outcomes.
Part 3 Guided practice and feedback
Compute a second candidate with draft 4 + review 12: 160 minutes/week, worse than 150 despite faster drafting. Identify missing evidence: whether quality, task mix and review standards were held constant. Reject a value claim until those conditions are checked.
Review and next step
Self-review is practice, not independent certification or operational approval. A2 packages the selected bounded task for reproducible use.
Substantive review: 2026-10-02. Use the configuration guide when selecting tools. NIST AI RMF is voluntary risk-management guidance and grants no course or operational authorization.