The artisan consultant recreates everything from scratch for every client. A consultant with a scalable practice enters the second project 30% faster and the third 50% faster because they have templates that work. Every deliverable you create well once can become an asset forever.
Illustrative diagram — sequence of deliverables in the AI consulting kit.
📋 Readiness questionnaire
The questionnaire is the first deliverable — sent before the first meeting. Arriving informed isn’t just about efficiency: is the first sign that you’re a professional. The client immediately realizes they’re not talking to someone winging it.
What the questionnaire covers
Data and infrastructure
- • Which systems are used (ERP, CRM, BI)?
- • Is the data structured or scattered?
- • Is historical data available? How many years?
Processes and pain points
- • Which process takes the most time?
- • Where do errors cost the most?
- • Have you tried AI before? What happened?
Culture and readiness
- • Does the team already use an AI tool?
- • Is there internal resistance to change?
- • Who are the decision-makers?
Expectations
- • What is the ideal result in 6 months?
- • What budget is available?
- • What is the most critical deadline?
💡 The Golden Rule of Questionnaires
Maximum 15 questions, answered in 20 minutes. If it takes longer, the client procrastinates and you lose momentum. Use open-ended questions for discovery and closed-ended questions for quick qualification.
🗺️ Opportunity map + value×feasibility matrix
The map lists all potential AI use cases identified during discovery. The matrix positions each one along two axes: business value × technical feasibility. The right quadrant is where you start.
The 4 quadrants
High value + High feasibility
→ Start here. Quick wins.
High value + Low feasibility
→ Medium-/long-term projects
Low value + High feasibility
→ Only if capacity allows
Low value + Low feasibility
→ Don't do it. Don't even plan it.
Why the matrix works
- •Makes a decision that is often political objective
- •Makes it easier for departments with different interests to align
- •Documents the prioritization logic — protects you and the client
- •It’s visual—any executive can understand it in 30 seconds
🌊 Wave-based roadmap + 1-page business case
The roadmap turns the opportunity map into an execution plan with time windows. The one-page business case translates each wave into the language of the people who approve the budget. Executives approve projects that fit on one sheet of paper.
🌊 Wave 1 — 90 days
Q1 matrix use cases: high value, high feasibility. Delivers visible results quickly and builds internal momentum.
🌊 Wave 2 — 6 months
More complex cases or those that depend on the wave 1 infrastructure. Lay the groundwork while harvesting results from the first wave.
🌊 Wave 3 — 12 months
Long-term vision. Projects that need cultural maturity and accumulated data. Anchors the ambition without committing to impossible timelines.
📊 One-Page Business Case Structure
what is costing or blocking us
what will change with AI
savings, revenue, or risk avoided
how many months to pay for itself
what could go wrong and how to mitigate it
what needs approval now
📊 Executive One-Pager
The one-pager is the document that travels without you — it’s sent to the CEO before a meeting, circulates among VPs, and reaches the board. It needs to be readable by someone looking at their phone between two meetings.
Executive one-pager template
Current situation
What’s happening today — concrete numbers where possible.
Opportunity
What can be changed with AI — the expected outcome in business terms.
Approach
How it will be done — phases, timeline, owners. Maximum 5 lines.
ROI + next step
The expected return and the decision that needs to be made now.
💡 The One-Pager Test
Read it aloud. It should take less than 2 minutes. If it takes longer, it's too long. The executive reads between meetings—if it doesn't fit their pace, it won't get read.
🛡️ Risk register and governance note
AI projects have specific risks that consultants in other fields often overlook: algorithmic bias, data privacy, silent failures, model drift. Documenting beforehand is more professional and less expensive than fixing things afterward.
📋 Risk register (5 columns)
⚖️ Governance note (3 sections)
⚠️ AI-specific risks every consultant should map
- •Data bias: model trained on historical data that reflects unfair practices
- •Silent failure: model that degraded without anyone noticing
- •Vendor dependency: API that changes pricing or policy without notice
- •Privacy: customer data used to train models without consent
♻️ Productize and reuse deliverables
The difference between a craftsperson and a scalable practice isn’t talent—it’s assets. Each deliverable customized for a client is a potential template. The first project funds the setup; every other project benefits.
✓ Practice with assets
- ✓Each project improves the existing template
- ✓New client onboarding: 30% faster
- ✓Consistency across deliverables (fewer errors)
- ✓You can outsource parts with clear instructions
✗ Practice without assets
- ✗Starts from scratch with every new client
- ✗Quality varies with fatigue and context
- ✗Can’t delegate — it’s all in your head
- ✗Hours increase with the number of clients
💡 How to Start Productizing Today
When you finish a project, open each deliverable and remove the client-specific data. What remains is the template. Add comments explaining the logic behind each section. After 6 projects, you’ll have a complete kit that gives you a real competitive advantage.
🎒 Module summary
Next module:
5.4 — Adoption, change management, and scaling the practice: the real bottleneck isn’t technology, it’s people