AI projects generate more stakeholder anxiety than traditional projects:
Poor communication doesn't just annoy stakeholders—it kills projects. Good communication doesn't just inform—it builds the trust that lets you navigate inevitable setbacks.
Every status update should answer five questions:
AI PROJECT STATUS: [Date]
1. ARE WE ON TRACK?
Status: 🟢 On Track | 🟡 At Risk | 🔴 Off Track
2. WHAT DID WE ACCOMPLISH?
- [Milestone/deliverable completed]
- [Metric achieved]
3. WHAT'S NEXT?
- [Next milestone]
- [Key activities this period]
4. WHAT'S IN THE WAY?
- [Blocker/risk]: [Impact] → [Mitigation]
5. WHAT DO WE NEED?
- [Decision needed from whom by when]
- [Resource or access needed]
Executives have limited time and need different information than project teams:
| Executives Care About | Executives Don't Care About |
|---|---|
| Business outcomes | Technical details |
| Money and timeline | Individual task status |
| Strategic risks | Tactical issues |
| Decisions they need to make | Decisions you already made |
| What's different from the plan | Everything going as expected |
The 2-minute executive update:
AI projects are particularly prone to expectation gaps. Set expectations proactively:
Before the project:
During the project:
At milestones:
Every AI project eventually requires difficult conversations:
When delivering bad news, use this structure:
THE BAD NEWS CONVERSATION
1. STATE THE SITUATION
"I need to share some challenging news about [project]."
2. PRESENT THE FACTS
"Here's what happened: [objective description]"
3. EXPLAIN THE IMPACT
"This means: [timeline/budget/scope impact]"
4. TAKE OWNERSHIP
"I take responsibility for [what you control]"
5. PRESENT OPTIONS
"We have three paths forward:
- Option A: [description + trade-offs]
- Option B: [description + trade-offs]
- Option C: [description + trade-offs]
I recommend [X] because [reason]"
6. ASK FOR INPUT
"What questions do you have?"
"What am I missing?"
7. COMMIT TO NEXT STEPS
"I'll [action] by [date] and update you by [date]"
| Anti-Pattern | Problem | Correction |
|---|---|---|
| Hiding bad news | It always comes out, trust destroyed | Surface issues early |
| Technical jargon | Stakeholders tune out | Plain language, business focus |
| Status inflation | 🟢 until suddenly 🔴 | Honest, early warning |
| Blame shifting | Destroys relationships | Take ownership, focus on solutions |
| Over-promising | Credibility damage | Under-promise, over-deliver |
| Infrequent updates | Stakeholders fill silence with fear | Regular cadence, even if brief |
| One-way communication | Miss stakeholder concerns | Ask questions, seek feedback |
Different stakeholders need different frequencies and depths:
Executive Sponsors (Monthly)
Steering Committee (Bi-weekly)
Working Team (Weekly)
Stakeholders Affected by Change (As needed)
"We're behind schedule"
"The AI isn't performing as expected"
"We need more budget/time"
"We should stop this project"
STAKEHOLDER COMMUNICATION PLAN
PROJECT: [Name]
DURATION: [Weeks]
STAKEHOLDER MAP
| Stakeholder | Interest | Influence | Needs |
|-------------|----------|-----------|-------|
| [Name/Role] | H/M/L | H/M/L | [What they care about] |
COMMUNICATION SCHEDULE
| Audience | Frequency | Format | Owner | Content |
|----------|-----------|--------|-------|---------|
| Exec Sponsor | Monthly | 1-pager | PM | Strategic status |
| Steering | Bi-weekly | Meeting | PM | Phase progress |
| Team | Weekly | Standup | Lead | Task status |
| Affected users | As needed | Email | Champion | Change updates |
STATUS REPORT TEMPLATE
[Customize the 5-question template for your project]
ESCALATION PATH
- Level 1 (Working issue): [Who handles]
- Level 2 (Needs steering input): [Process]
- Level 3 (Executive decision): [Process]
KEY MILESTONES FOR COMMUNICATION
| Milestone | Date | Who to Inform | Message |
|-----------|------|---------------|---------|
| [Milestone] | [Date] | [Audience] | [Key message] |
Your AI project is 3 weeks behind the original 12-week timeline. The cause was unexpected data quality issues that required additional cleaning.
Write a 2-paragraph update to your executive sponsor that:
Scenario 2: The Performance Gap
Your AI classification system is achieving 78% accuracy. The target was 90%. You're 6 weeks into an 8-week pilot.
Write talking points for a conversation with your steering committee:
Scenario 3: The Stop Decision
After 4 weeks of a pilot, it's clear the use case isn't viable. The AI can't reliably handle the edge cases that matter most, and users aren't adopting.
Write a memo recommending project termination:
Complete the Phase 5 labs to practice selling AI to skeptics and assessing team capabilities. Then move to Phase 6: Enterprise AI Architecture where you'll learn to think at portfolio scale—designing vertical playbooks, balancing AI investments, and creating operating models for enterprise AI.
You've reached the end of this module. Review the checklist below to ensure you've understood the key concepts.