The Consulting Prompt Mindset

Develop the prompting approach that works for consulting's specific demands — where structure, analytical rigour, and client-readiness are the standard for every output.

In practice: Desk research: 3 days → 4–8 hours

Step 1 of 4

Read

0% done

Why Consulting Prompting Is Different

Consulting output is expected to be structured, logical, evidence-based, and client-ready. Generic AI output rarely meets this standard. Consulting prompts need to specify: the analytical framework, the level of rigour, the client context, and the output format.

The Consulting Prompt Elements

1. Project context. What is the engagement? What is the client situation? What decision are you trying to inform?

2. Analytical framework. What structure should the output follow? MECE breakdown? McKinsey problem solving? Porter's Five Forces? Specify the framework explicitly.

3. Evidence standard. How specific should claims be? Should they be supported by data? What level of assertion is appropriate here?

4. Output format. Bullet points for speed, or prose for a client memo? Slide-ready text, or analytical working notes?

5. Consulting voice. Client-ready language, or working analysis? Should it read like a senior consultant wrote it?

Prompt Patterns for Consulting Tasks

Issue tree prompt: "I am working on a [type] engagement for a [client type]. The central question is [state it]. Build a MECE issue tree that breaks down this question into the key sub-questions a consultant would need to answer. Three levels deep."

Hypothesis generation: "For a [industry] client facing [problem], generate eight potential hypotheses about the root cause. Organise them into three to four theme buckets. For each hypothesis, suggest one data source or analysis that would test it."

Framework application: "Apply Porter's Five Forces to the [industry] sector. For each force, assess the current intensity (high/medium/low) and provide three specific supporting observations. Then summarise the overall industry attractiveness."

Calibrating for Client-Ready Output

AI defaults to a generic professional voice. Consulting output needs to be more precise, more assertion-heavy (so what, not just what), and more structured. The calibration prompt: "Rewrite this in the voice of a senior McKinsey consultant writing for a CEO audience. Lead with the insight, not the analysis. Use pyramid principle structure: answer first, then supporting evidence."

Competitive landscape: generic prompt vs. consulting-grade prompt

The prompt

Give me a competitive landscape of the UK private equity market

Claude's response

The UK private equity market is a dynamic and competitive landscape with several key players:

**Major firms:** Apax Partners, CVC Capital Partners, Cinven, and BC Partners are among the largest UK-headquartered PE firms.

**Market trends:** The market has seen increased competition, with more capital chasing fewer quality deals. ESG considerations are becoming more prominent.

**Challenges:** Rising interest rates have impacted deal economics, and exit markets have been more challenging.

Overall, the UK PE market remains one of Europe's most active, though conditions have become more selective in recent years.

Quick check

You ask AI to "analyse the retail banking sector." The output is well-written but generic. What is the most likely cause?

Key Takeaways

1Consulting prompts require five elements: project context, analytical framework, evidence standard, output format, and consulting voice
2Specify the framework explicitly — MECE, Porter's Five Forces, issue trees — for structured analytical output
3AI defaults to generic professional voice; calibrate for consulting's answer-first, assertion-heavy standard
4The pyramid principle prompt ("answer first, then evidence") produces more client-ready output than default AI style
5Build a consulting prompt library for your most common task types to accelerate future engagements

Your challenge this week

Apply what you learned in a real task

Write one consulting-grade prompt for a task you do on every engagement. Test it, refine it, and save it as the first entry in your personal prompt library.

Starter prompt · paste into Claude

I am working on a [type of engagement] for a [client type] in [industry]. The central analytical question is: [state the question]. Apply the [framework name] framework to this question. For each component, provide: (a) an assessment (high/medium/low or specific rating), (b) three specific supporting observations relevant to this client context, and (c) the strategic implication for the client. Conclude with the two most important overall insights. Write in the voice of a senior consultant for a C-suite audience: assertion-led, specific, and action-oriented.

Before you practise

On your last engagement, where did the analytical output — whether from AI or your own drafting — fall short of client-ready standard? Was it the structure, the evidence, the specificity, or the voice? What would a better prompt have specified?

Was this lesson helpful?

Next step

Put it into practice

You've read the lesson — now apply it in a guided hands-on exercise. It takes about 5 minutes.

Ask the AI Tutor