Pick the right use case · E-Learning
AI Use-Case Finder
When to use it: before anyone writes a prompt. Teams that start with the tool end up with random acts of AI. This finder starts with the recurring work that fills the week and asks, task by task, whether AI genuinely helps.
Practice it first: the Use-Case Finder simulation — six fictional AI ideas built around your marketing, scored and debriefed.
Working area — map the week
| Recurring task | Channel | Frequency | AI fit | Why |
|---|
Use-case shortlist
Pick the right use case · Individual Coaching
Use-Case Prioritization Matrix
When to use it: once the shortlist is longer than the time you have. The matrix scores business value, feasibility, data readiness, and risk, so effort goes to the use case that can prove something — not the one that demos best.
Working area — score, then let the ranking argue for you
| Use case | Business value | Feasibility | Data readiness | Low risk | Total | Priority |
|---|
Priority logic: total of 16–20 = Start now · 12–15 = Pilot with guardrails · 8–11 = Park · below 8 = Drop.
Load the context · Individual Coaching
Brand Voice & Context Pack
When to use it: before the first real prompt. Generic output comes from a model that was never told who you are, who you serve, or what you sell. The pack captures it once, so every prompt starts from the same verified facts.
Working area — build the pack
Context pack — paste at the top of every prompt
Load the context · Live Online Simulation
Customer Insight Prompt
When to use it: when you need to understand a segment, a competitor, or the questions buyers ask. AI can synthesize fast — and invent just as fast. This prompt asks for sources and separates facts from hypotheses, and the checklist makes you verify before anything is used.
Working area — research, then verify
Insight prompt
Verification checklist — before any insight is used
Prompt and produce · E-Learning
Campaign Prompt Builder
When to use it: every time you brief AI on a deliverable. Inspired by the CO-STAR framework (GovTech Singapore), the builder makes you state the context, objective, style, tone, audience, and expected response — the brief you would give a new team member.
Practice it first: the Prompt Lab simulation — judge six prompts for the same task before you write your own.
Working area — brief the model
Assembled prompt
Prompt and produce · Live Online Simulation
Content Variant Planner
When to use it: when one campaign has to work across channels and segments. AI makes variants cheap, which is exactly how the core message gets lost. The planner fixes the message first, then plans what changes per channel — and turns the plan into one prompt.
Working area — fix the message, vary the angle
| Channel | Segment | Variation angle | Call to action |
|---|
Variant request prompt
Review and verify · E-Learning
Human Review Checklist
When to use it: before anything AI-assisted ships. A draft is not a deliverable. A named reviewer checks accuracy, brand, bias, privacy, intellectual property, and disclosure, and owns the result as if they had written every word.
Working area — review the piece
Accuracy
Brand
Bias
Privacy
Intellectual property
Disclosure
Launch and test · Live Online Simulation
Test & Measurement Plan
When to use it: before launch, not after. Teams that feel faster but cannot show it never earn the right to scale. The plan names the KPI, the test, and the baseline up front, then calculates time saved and KPI change from your own numbers.
Working area — plan the test, read the result
Hours saved = (before − with AI) × deliverables per month. Time per deliverable = change vs. before.
Measure and scale · Individual Coaching
AI Workflow Card
When to use it: when a workflow has run on real work and has a result. The card documents it so a manager can approve it, the team can reuse it, and anyone can check the results. It is The Engine's final deliverable.
Working area — document the workflow
AI Workflow Card — for manager review