Phase 1 – Behavioral & Data Assessment

Establishing AI and process improvement readiness across people, processes, and data.

Goal Establish AI and process improvement readiness across people, processes, and data.
1

Data readiness scan

  • Quick inventory of key systems supporting 5–10 priority processes.
  • Assess data quality for those processes:
    • Completeness
    • Consistency
    • Accessibility
  • Identify where “tribal knowledge” lives only in people’s heads versus in documented processes, SOPs, or systems.
2

Behavioral & change-readiness assessment

  • Individual assessments for key managers, supervisors, and project leaders, probing for strengths and blind spots:
    • Balancing change vs. structure.
    • Balancing experimentation vs. persistence.
    • Effectiveness at providing clarity.
    • Balancing collaboration vs. decision velocity.
    • Balancing direction-setting vs. planning ahead.
    • How they view the future and balance risk-taking, optimism, and what-if analysis.
    • Process discipline vs. ad hoc working.
    • Comfort with ambiguity and technology.
  • Identification of blind spots related to process thinking, change management, and collaboration.
3

Prompting / AI mindset baseline

  • Establish a baseline of prompting behavior and AI mindset for each manager.
  • Categorize prompting maturity (e.g., Beginner, Emerging, Advanced) by role and responsibility.
  • Identify high-leverage use cases where better prompting would immediately improve quality and throughput.
4

Competency & job fit analysis for managers

  • Map current managerial competencies against the demands of AI-enabled process improvement.
  • Highlight strengths that will accelerate implementation and adoption.
  • Surface risks and derailers that could slow or block AI-related initiatives.
5

Phase 1 deliverables

  • Organization-wide AI Readiness Heatmap across people × process × data.
  • Manager profiles and development plans outlining key opportunities and threats.
  • Shortlist of 5–8 high-potential workflows for immediate prompt support and future automation.
Return to Main Program Overview
2
Next Phase
Role-Based Prompting & Knowledge Management
Focus shifts from readiness to role-specific AI enablement.
Phase 2 – Role-Based Prompting & Knowledge Management

Equipping each manager with role-specific prompt patterns that solve real problems.

Goal Equip each manager with role-specific prompt patterns that solve real problems.
1

Prompt pattern design by role

  • Develop prompt “recipes” for 3–5 core manager roles (e.g., Sales, Operations, Finance, HR).
  • Anchor design on recurring, high-friction tasks such as:
    • Reporting and analysis
    • Emails and stakeholder updates
    • Meeting preparation and follow-through
    • Coaching and performance conversations
    • Documentation and knowledge capture
2

Context & constraints built in

  • Embed organizational context, policies, and constraints into prompts so outputs are usable, not theoretical.
  • Provide practical toolkits:
    • Templates and guardrails for safe, consistent prompting.
    • Role-specific prompt libraries and reference sheets.
    • Development of mini “playbooks” for each manager.
3

Hands-on sessions

  • Facilitated working sessions where managers apply their prompt patterns to real, current work.
  • Live refinement of prompts based on outcomes, edge cases, and manager feedback.
  • Focus on “in-the-flow-of-work” usage rather than abstract training.
4

Phase 2 deliverables

  • Role-specific Prompt Playbooks for each participating manager/role.
  • Before/after examples highlighting time saved, improved quality, and better decision support.
  • A growing, living repository of prompts that can later be integrated into agentic AI and automation in Phase 6.
Return to Main Program Overview
3
Next Phase
Organizational Integration
AI shifts from experimentation into embedded, repeatable workflows.
Phase 3 – Organizational Integration

Moving from “AI as a toy” to AI embedded in business processes.

Goal Move from “AI as a toy” to “AI embedded in business processes.”
1

Process mapping for priority workflows

  • Map 5–10 priority workflows where AI can materially improve speed, quality, or consistency.
  • Identify exactly where prompts fit into each step (inputs, decisions, handoffs, summaries).
  • Define or update Standard Operating Procedures (SOPs) to explicitly include AI steps.
2

AI-enabled SOPs

  • Create or refine SOPs so AI usage is:
    • Documented, not ad hoc.
    • Integrated into checklists and workflows.
    • Clear about when humans review, override, or approve outputs.
  • Align roles and responsibilities so managers, analysts, and frontline staff know how AI supports their work.
3

Tool & system touchpoints

  • Identify where AI interactions live today (chat interfaces, internal tools, data platforms).
  • Clarify how prompts, outputs, and decisions connect to:
    • CRMs, ERPs, HRIS and other core systems.
    • Document repositories and knowledge bases.
    • Task and project management tools.
  • Flag future integration points for agentic AI and automation in later phases.
4

Measurement, governance & deliverables

  • Establish basic measurement and governance for AI-in-process:
    • Define success metrics (time saved, errors reduced, quality uplift).
    • Decide who reviews AI usage and outcomes.
  • Deliverables include:
    • AI-enhanced process maps and AI-enabled SOPs for key workflows.
    • A simple measurement plan for impact tracking.
Return to Main Program Overview
4
Next Phase
Behavioral / Effectiveness Facilitation & Coaching
Turning AI usage into a disciplined leadership habit.
Phase 4 – Behavioral / Effectiveness Facilitation / Coaching

Making AI usage a disciplined leadership habit, not a one-off experiment.

Goal Make AI usage a disciplined leadership habit, not a one-off experiment.
1

1:1 coaching for key leaders

  • Review each manager’s prompting style, behavioral profile, and Phase 1 assessment results.
  • Address specific tendencies that impact AI effectiveness, such as:
    • Over-reliance on AI vs. chronic under-utilization.
    • Over-analysis that blocks experimentation and iteration.
    • Poor instructions or unclear problem framing.
  • Co-create practical behavior shifts that align with their role and team realities.
2

Live practice & reflection

  • Managers bring real use cases and prompts from the past week for live review.
  • Walk through recent AI interactions:
    • What was the goal?
    • What prompt was used?
    • What worked and what did not?
  • Introduce micro-habits and nudges:
    • “Before you do X, try one iteration with your role prompt.”
    • Scheduled 5-minute “AI warm-up” practices to start the day or key tasks.
3

Team coaching & peer labs

  • Facilitate peer sessions where managers share prompts, successes, and failures.
  • Normalize experimentation while reinforcing standards for quality and governance.
  • Capture “from the field” examples that feed back into prompt libraries and SOPs.
4

Phase 4 deliverables

  • Individual Development Plans for each participating manager, focused on AI-enabled leadership behaviors.
  • Improved prompting examples with documented behavioral shifts (before/after patterns).
  • A clearer picture of which leaders can champion AI-enabled process improvement in later phases.
Return to Main Program Overview
5
Next Phase
Prompt Library as Knowledge Management
Turning ad hoc prompting into a reusable, evolving asset.
Phase 5 – Prompt Library as Knowledge Management

Turning ad hoc prompting into a reusable, evolving asset for the organization.

Goal Turn ad hoc prompting into a reusable, evolving asset for the organization.
1

Prompt library design

  • Design a simple, searchable structure for prompts and patterns (by role, process, and outcome).
  • Define standard fields (e.g., purpose, context, inputs, version, owner) so prompts are understandable and reusable.
  • Decide where the library “lives” (knowledge base, shared drive, internal site) and how it is accessed.
2

Contribution & curation

  • Define how managers and teams contribute new prompts and improvements.
  • Establish light curation:
    • Which prompts are “approved” or recommended.
    • How duplicates and outdated prompts are handled.
  • Identify a small group of “prompt stewards” or librarians responsible for oversight.
3

Integration & feedback loops

  • Integrate key prompts and patterns into:
    • New manager onboarding.
    • Role-specific training and playbooks.
    • Ongoing learning sessions and peer labs.
  • Build feedback loops:
    • Quick ways to flag prompts that do not work well.
    • Channels to suggest improvements or new use cases.
  • Treat the library as a living asset that evolves with the business, not a static document.
4

Phase 5 deliverables

  • A working Prompt & Pattern Library for the organization, aligned to priority roles and workflows.
  • Basic governance and maintenance guidelines so the library stays alive and useful, not a “graveyard.”
  • Clear integration points to support future automation and agentic AI in later phases.
Return to Main Program Overview
6
Next Phase
Planning & Initial Agentic AI Automations
Moving from enhanced human work to carefully chosen automation.
Phase 6 – Planning & Setting Up Initial Agentic AI Automations

Moving from enhanced human work to carefully chosen, trusted automation using agents.

Goal Move from enhanced human work to carefully chosen, trusted automation using agents.
1

Automation opportunity assessment

  • Review Phase 2–5 outputs (prompt patterns, workflows, SOPs, readiness) to identify strong automation candidates.
  • Prioritize workflows where:
    • Steps are well defined and repeatable.
    • Data quality is sufficient.
    • Risk can be controlled via human review or guardrails.
  • Align priority automations with business outcomes (cost, speed, quality, risk reduction).
2

Agent role design & handoffs

  • Define clear agent roles: what each agent is responsible for and what it is not.
  • Clarify human–agent handoffs:
    • Where humans provide inputs, approvals, or overrides.
    • What decisions stay human-owned vs. agent-supported.
  • Translate existing prompts and patterns into agent behaviors and flows.
3

Technical planning & pilot agents

  • Identify technical requirements:
    • Data and system connectivity.
    • Security, access, and permissions.
    • Monitoring and logging needs.
  • Plan and stand up 1–2 pilot agents:
    • Start with low-to-moderate risk, high-visibility workflows.
    • Define clear success criteria and safety checks.
  • Capture lessons learned from pilots to refine design and governance.
4

Scale, roadmap & deliverables

  • Define a scale-out approach for successful agents (where to expand, in what order, and under what conditions).
  • Outline a 6–18 month Agentic Automation Roadmap that aligns with:
    • Business priorities and capacity.
    • Process maturity and data readiness.
    • Change management and training capacity.
  • Deliverable: a clear, prioritized Agentic Automation Roadmap connecting today’s pilots to future automation.
Return to Main Program Overview