Bloom-aligned objectives

• Understand: how Copilot is grounded in work data and permissions
• Apply: prompt structures that reliably control output format and tone
• Analyze: when Copilot may be incomplete or incorrect and how to mitigate

Topics

• What Microsoft 365 Copilot is (in-app + Copilot Chat) and what it can do across Word/Excel/PowerPoint/Outlook/Teams
• Grounding and access control basics
  • Why Copilot responses reflect your existing permissions
  • Semantic Index for Copilot: what it enables and how it influences retrieval and answers
• Power-user prompting patterns (Copilot-specific)
  • Reference-first prompting: “Use this file/email/thread as source”
  • Output contracts: “Return in this structure, with sections and checklists”
  • Iteration loop: draft → critique → rewrite with constraints
• Practical quality controls
  • “Show assumptions” prompts
  • “What could be wrong?” prompts
  • “Give me missing questions to ask stakeholders” prompts

Lab 1A (35 min): “Prompt upgrade: from vague to deterministic”

• Rewrite 8 common work prompts into structured prompts (role, goal, sources, format, constraints).
• Deliverable: a reusable prompt set categorized by task type (drafting, summarizing, planning, decision memos).

Lab 1B (45 min): “Grounded response drill”

• Use a provided sanitized folder (emails + doc + slide) and prompt Copilot to answer questions only from those sources.
• Deliverable: response with explicit “source used / not found” notes (traceability habit).

Bloom-aligned objectives

• Apply: Copilot to draft, extend, rewrite, and summarize with business constraints
• Evaluate: tone, clarity, and completeness against a rubric
• Create: reusable document and email templates powered by prompt patterns

Topics

• Copilot in Word: drafting and adding content from existing context (files, notes)
• Editing and transformation workflows
  • Executive tone rewrites
  • Condense long sections into structured summaries
  • Turn bullets into a formal policy/one-pager
• Copilot for Outlook workflows (power-user patterns)
  • Thread compression into decisions + actions + owners
  • Customer-ready reply variants (formal, concise, empathetic)
  • Convert email thread → meeting agenda + pre-read

Lab 2A (60 min): “Executive-ready one-pager”

• Input: messy notes + a reference doc.
• Output: one-page brief with: objective, background, options, recommendation, risks, next steps.
• Deliverable: v1 + v2 after self-critique checklist.

Lab 2B (60 min): “Inbox-to-action workflow”

• Input: simulated email thread bundle.
• Outputs: (1) thread summary, (2) action tracker, (3) two draft responses with different tones, (4) meeting agenda.
• Deliverable: packaged set that can be reused weekly.

Bloom-aligned objectives

• Apply: Copilot to convert narratives into slides and speaker notes
• Analyze: whether slides match decision intent and audience needs
• Create: a repeatable “doc-to-deck” workflow

Topics

• Doc-to-deck storytelling
  • Build an outline first, then generate slides
  • Slide narratives, executive summaries, speaker notes
• Teams collaboration acceleration
  • Converting discussions into decisions and follow-ups
  • Preparing pre-reads and distributing actions
• Meeting productivity foundation (preparing for Day 2 meeting module)

Lab 3A (75 min): “Document-to-deck conversion”

• Input: a 2–3 page business brief.
• Outputs: 6–8 slide outline + slide titles + speaker notes + 1-slide exec summary.
• Deliverable: deck narrative + “what changed from doc” delta notes.

Lab 3B (45 min): “Stakeholder update pack”

• Input: a set of updates across 5 workstreams.
• Output: a weekly status update (RAG status, risks, dependencies, asks).
• Deliverable: Teams-ready update + slide-ready summary.

Bloom-aligned objectives

• Apply: Copilot to analyze data, create pivots, and explain formulas
• Analyze: trends, anomalies, and drivers with sanity checks
• Evaluate: when AI-generated analysis is unsafe for accuracy-critical reporting

Topics

• Copilot in Excel capabilities and current experience patterns (chat, agent-like analysis modes where available)
• Insight workflows
  • Summarize performance by segment
  • Driver analysis (what explains change)
  • Outlier detection + data quality questions to ask
• Formula workflows
  • Generate formula + explanation + test cases
  • “Explain this workbook” and “audit this logic” prompts
• Validation discipline
  • reproducibility checks, spot checks, reconciliations (human-in-the-loop)

Lab 4A (90 min): “Performance analysis sprint”

• Input: sales dataset (products, regions, months) with intentional anomalies.
• Tasks: summarize KPIs, identify drivers, detect anomalies, propose actions.
• Deliverable: 1-page insights memo + validation checklist completed.

Lab 4B (30 min): “Formula + pivot clinic”

• Create a PivotTable via Copilot and generate 3 formulas (one nested).
• Deliverable: formula set with explanation and 2 test cases each.

Bloom-aligned objectives

• Apply: Copilot to extract decisions, actions, and open questions from meetings
• Analyze: ambiguous items and assign ownership (RACI-style)
• Create: meeting-to-execution artifacts (minutes, action tracker, follow-up email)

Topics

• Using Copilot in Teams meetings to capture key points, action items, and outcomes
• Meeting prompt patterns (during and after)
  • “What decisions were made?”
  • “List action items with owners and due dates”
  • “What topics were unresolved and what info is missing?”
• Turning meeting output into execution
  • action tracker
  • stakeholder recap message
  • next meeting agenda auto-generation

Lab 5A (60 min): “Meeting recap to action plan”

• Input: meeting transcript excerpt (sanitized).
• Outputs: minutes, decisions, action items, risks, dependencies, follow-up email draft.
• Deliverable: action tracker + follow-up communication.

Lab 5B (30 min): “Hard questions generator”

• Input: a project status summary.
• Output: 12 stakeholder-ready questions grouped by risk area (scope, timeline, budget, dependencies, compliance).
• Deliverable: question bank to use in real governance forums.

Bloom-aligned objectives

• Understand: Microsoft 365 Copilot privacy and data handling commitments
• Evaluate: prompts and outputs for data risk (oversharing, sensitivity, leakage)
• Create: a personal Copilot playbook and team usage rules

Topics

• Privacy and data security fundamentals for Microsoft 365 Copilot (what happens to prompts/responses and training boundaries)
• Oversharing risk and data protection alignment
  • sensitivity labels, auditing, and oversharing controls impact Copilot experiences
• Safe-use rules for power users
  • “Never paste” categories (credentials, regulated data, client confidential)
  • How to ask for redaction, anonymization, and safe summarization
• Extensibility awareness (for power users)
  • What connectors/agents are in principle (so power users can request the right enablement from IT)
  • Conceptual overview: how copilots connect to enterprise tools (Graph/Connectors) vs emerging open tool-connection protocols (awareness only)

Workshop (75 min): “Power user playbook”

• Outputs:
  • Personal prompt library (by app + by task)
  • QA checklist (accuracy, traceability, sensitivity, approvals)
  • Team rules: approved usage, escalation paths, and review thresholds
• Deliverable: a reusable Copilot Operating Playbook (v1) ready for team rollout.

Final simulation (45 min): “One workday, end-to-end”

Teams complete a timed sequence:

1. Draft an email reply from a thread
2. Turn notes into a one-page brief
3. Convert the brief into a mini-deck outline
4. Produce Excel insights from a dataset
5. Generate Teams meeting recap outputs
6. Flag data risks and apply guardrails

Deliverables: final artifacts + a short “prompt strategy” explanation (what patterns worked and why).