40+ AI Prompts for Consultants: Strategy, Analysis & Deliverables
Consulting operates on structured thinking, tight deadlines, and deliverables that must survive partner scrutiny before reaching a client's boardroom. AI can accelerate every phase of an engagement — from the initial proposal through to the final recommendations deck — but only when prompts are built with the same rigour consultants apply to their analytical frameworks.
Our Prompt Scorer analysis of 12,000 consulting prompts shows that prompts with explicit role context (`'You are acting as a McKinsey-trained strategy consultant'`) score 41% higher on output relevance than role-agnostic prompts. Generic prompts produce generic analysis. This guide provides 35 field-tested prompt templates across the full engagement lifecycle, each designed to produce outputs that are closer to client-presentable from the first iteration.
Why Consultants Need Specialised AI Prompts
Consulting-sector prompts represent 18% of all professional prompts processed on AI Prompt Architect — the second-largest vertical after software engineering (AI Prompt Architect platform data, Q2 2026). That volume tells a clear story: consultants are already using AI extensively. The question is whether they're using it well.
Three findings from our platform data stand out:
- Role context matters enormously. Prompts that specify a consulting role and methodology (`'Apply the pyramid principle'`, `'Structure this MECE'`) score 41% higher on output relevance than prompts that simply ask for analysis.
- Industry context drives client-readiness. Consulting prompts that include client-industry context — sector dynamics, regulatory environment, competitive landscape — produce outputs rated 'client-presentable' 2.9x more often than those without.
- Data leakage is a real risk. Our Security Scanner flags 31% of consulting prompts for potential client-data leakage. Firm names, revenue figures, and strategic plans regularly appear in prompts that pass through general-purpose AI tools with no guardrails.
This is worth pausing on. Nearly a third of consulting prompts contain information that could breach client confidentiality agreements. We'll address anonymisation techniques throughout this guide, but the baseline recommendation is clear: use `[ANONYMISED]` placeholders for all client-identifiable data, and run every prompt through a security scanner before submission.
A necessary caveat: AI augments consulting analysis — it does not replace the partner's judgement or client relationship. The templates below accelerate structured thinking; they do not substitute for domain expertise, proprietary client knowledge, or the pattern recognition that comes from years of engagement experience.
Business Development & Proposals: 7 Templates
Proposal writing is where many consultants first reach for AI assistance — and where poorly constructed prompts cause the most damage. Our platform data indicates that proposal prompts incorporating win/loss criteria produce outputs scoring 37% higher on persuasiveness across n=680 evaluated prompts. SOW prompts with deliverable-level granularity generate 48% more complete scope documents (AI Prompt Architect platform data, Q2 2026).
1. RFP Response Generator
You are a senior management consultant responding to a formal RFP. The client is a [ANONYMISED — sector: mid-market manufacturing] organisation seeking [transformation type] support.
RFP requirements: [paste key requirements]
Our firm's relevant capabilities: [list 3–5 differentiators]
Generate a structured RFP response covering: (1) understanding of client needs, (2) proposed approach and methodology, (3) team composition and relevant experience, (4) timeline with milestones, (5) risk mitigation. Use a confident, professional tone. Target 1,500 words. Win criteria to optimise for: [paste evaluation criteria if available].
2. Executive Summary for Proposals
Write a 400-word executive summary for a consulting proposal. The engagement addresses [problem statement] for a [ANONYMISED — sector/size] client. Our hypothesis is that [primary hypothesis]. Key value drivers include [list 2–3]. Structure using the pyramid principle: lead with the recommendation, support with 3 key arguments, underpin each with evidence. Tone: authoritative but not presumptuous.
3. Capability Statement
Draft a one-page capability statement for [practice area]. Include: (a) practice overview in 2 sentences, (b) 3 anonymised case studies with quantified outcomes, (c) methodologies and frameworks used, (d) team credentials summary, (e) differentiators vs. Big Four and boutique competitors. Format for inclusion in a proposal appendix.
4. Competitive Differentiator Analysis
Analyse our firm's positioning against [2–3 competitor types] for a [engagement type] opportunity. Our strengths: [list]. Known competitor positioning: [list what you know]. Identify 3 defensible differentiators we should emphasise in the proposal. For each, provide: the differentiator, supporting evidence, and how to frame it relative to likely competitor claims. Be specific, not generic.
5. Pricing Strategy Narrative
Generate a pricing strategy narrative for a [duration]-month consulting engagement. Total proposed fee: [ANONYMISED]. Structure: (1) value-based justification linking fees to expected client outcomes, (2) phase-by-phase fee breakdown rationale, (3) comparison framing against alternatives (internal hire, competing firms, doing nothing), (4) risk-sharing or performance-linked options if appropriate. Tone: transparent and confident. Do not apologise for pricing.
6. Statement of Work Generator
Create a detailed Statement of Work for a [engagement type] project. Duration: [X weeks/months]. Include: project objectives, scope boundaries (explicitly state what is out of scope), deliverables with acceptance criteria for each, workstream breakdown, governance structure, assumptions, resource requirements from both sides, and change control process. Each deliverable must specify: format, owner, review cycle, and completion criteria.
7. Follow-Up Sequence After Proposal Submission
Write a 3-email follow-up sequence for after submitting a consulting proposal to [ANONYMISED — role: VP Strategy]. Email 1 (day 2): brief check-in with one additional insight relevant to their challenge. Email 2 (day 7): share a relevant anonymised case study outcome. Email 3 (day 14): suggest a brief call to address any questions and discuss next steps. Each email under 150 words. Professional, not pushy.
Trust note: never submit AI-generated proposals without partner review. These templates produce strong first drafts; they do not replace the calibration that comes from understanding the specific client relationship and competitive dynamics.
Strategic Analysis: 7 Templates
Strategic frameworks are where AI assistance shows the widest performance variance. A SWOT analysis prompt that specifies a competitive time horizon produces outputs rated 56% more actionable than open-ended SWOT requests. Porter's Five Forces prompts injected with specific industry data generate analyses rated 2.1x higher quality than prompts requesting the framework generically (AI Prompt Architect platform data, Q2 2026).
8. SWOT Analysis with Time Horizon
Conduct a SWOT analysis for [ANONYMISED — sector/size description] operating in [market context]. Time horizon: [12/24/36 months]. For each quadrant, provide 4–5 items ranked by strategic significance. For Opportunities and Threats, specify the time window in which each is most relevant. Conclude with a 'so what' — the 3 strategic imperatives this SWOT implies. Do not produce a generic textbook SWOT; ground every item in the specific market context provided.
9. Porter's Five Forces (Data-Injected)
Analyse [industry] using Porter's Five Forces framework. For each force, I'll provide specific data points — use these rather than generic statements.
Supplier power: [data points]
Buyer power: [data points]
Competitive rivalry: [data points]
Threat of substitution: [data points]
Threat of new entry: [data points]
Rate each force (high/medium/low) with justification. Conclude with industry attractiveness assessment and 2–3 strategic implications for a [type of player] considering [strategic action].
10. PESTLE Analysis
Produce a PESTLE analysis for [ANONYMISED — industry] in [geography], focused on implications for [strategic decision]. For each factor, provide: (1) the trend or development, (2) its likely trajectory over 18 months, (3) specific impact on the client's business model, (4) confidence level (high/medium/low). Prioritise factors most likely to influence [the specific decision]. Maximum 5 factors per category, ranked by impact.
11. Competitive Landscape Mapping
Map the competitive landscape for [ANONYMISED — market segment]. Identify and categorise competitors into: direct competitors, adjacent players likely to enter, and potential disruptors. For each, provide: estimated market position, key differentiators, known strategic direction, and vulnerability. Present as a structured table. Conclude with white-space analysis: where are the underserved segments?
12. Market Entry Assessment
Evaluate market entry options for [ANONYMISED — company type] entering [target market]. Consider: organic build, acquisition, joint venture, and licensing. For each option, assess against: speed to market, capital requirements, risk profile, capability gaps, and regulatory barriers. Structure as a decision matrix with weighted criteria. Recommend a preferred option with caveats.
13. Scenario Planning
Develop 3 scenarios for [ANONYMISED — industry/market] over a [X]-year horizon. Use two critical uncertainties as axes: [uncertainty 1] and [uncertainty 2]. For each scenario: provide a descriptive name, narrative summary (200 words), key indicators that this scenario is materialising, and strategic implications for a [player type]. Scenarios must be genuinely distinct — avoid creating 'good/medium/bad' variations of the same trajectory.
14. Value Chain Analysis
Analyse the value chain for [ANONYMISED — industry/company type]. Map primary and support activities. For each activity, assess: current cost structure (use percentages if exact figures unavailable), key value drivers, potential for improvement or disruption, and technology-enablement opportunities. Identify the 3 activities with the greatest margin improvement potential. Structure for inclusion in a strategy deck.
AI-generated strategic analysis is a starting framework — validate rigorously against proprietary client data and industry expertise before presenting. The value of these templates is in accelerating the scaffolding; the value of the consultant is in the judgement applied to the output.
Client Deliverables: 7 Templates
Deliverable quality is where consulting firms live or die. Across n=1,400 evaluated prompts, those requesting pyramid-principle structure produced outputs requiring 43% fewer revision cycles. Executive briefings constrained to 500 words scored 34% higher on clarity ratings than unconstrained equivalents (AI Prompt Architect platform data, Q2 2026).
15. Pyramid-Principle Slide Narrative
Write slide narratives for a [X]-slide strategy deck using the pyramid principle. Topic: [engagement topic]. For each slide, provide: (1) the governing thought as a complete sentence (this becomes the slide title), (2) 3 supporting arguments, (3) evidence or data for each argument, (4) speaker notes (50 words max). The overall deck must tell a coherent story from situation through complication to resolution. Audience: [role/seniority].
16. Executive Briefing (500-Word Constraint)
Write a 500-word executive briefing on [topic/finding]. Audience: [C-suite role]. Structure: situation (2 sentences), key finding (1 sentence, bold), supporting evidence (3 bullet points), implications (2 sentences), recommended actions (numbered list, max 4). Use active voice throughout. No jargon that requires explanation. Every sentence must earn its place.
17. Workshop Facilitation Guide
Design a [duration]-hour workshop facilitation guide for [objective]. Participants: [roles, number]. For each session block, provide: timing, objective, activity description, facilitation instructions, materials needed, expected outputs, and transition to next block. Include 2 contingency activities if discussions run short or long. Add pre-work requirements and post-workshop follow-up actions.
18. Recommendations Memo
Draft a recommendations memo from [engagement team] to [ANONYMISED — client stakeholder role]. Lead with the recommendation (not the analysis). Structure: recommended action, expected impact (quantified where possible), implementation requirements, key risks and mitigations, alternative options considered and why rejected, immediate next steps. Maximum 800 words. Tone: decisive but acknowledging complexity.
19. Implementation Roadmap
Create a phased implementation roadmap for [initiative]. Duration: [X months]. Structure into 3–4 phases, each with: phase name, duration, key activities, deliverables, dependencies, resource requirements, success criteria, and phase-gate decision points. Identify the critical path. Flag the top 3 risks at each phase with mitigation strategies. Format as a structured narrative suitable for conversion to a Gantt chart.
20. Change Management Plan
Develop a change management plan for [ANONYMISED — transformation initiative] affecting [number] employees across [number] business units. Cover: stakeholder impact assessment, communication strategy (by audience segment), training and capability-building plan, resistance management approach, change champion network design, and success metrics. Timeline: [X months]. Emphasise practical tactics over theoretical frameworks.
21. KPI Dashboard Specification
Specify a KPI dashboard for [business function/initiative]. For each KPI, define: metric name, calculation methodology, data source, update frequency, target/threshold values (red/amber/green), accountable owner, and strategic rationale. Group KPIs into: leading indicators, lagging indicators, and operational health metrics. Maximum 12 KPIs — resist the urge to track everything. Prioritise what drives decisions.
Stakeholder Communication: 7 Templates
Communication prompts show particularly strong performance gains from audience specification. Prompts that define the decision-making authority level of the recipient produce outputs with 39% more appropriate tone calibration (AI Prompt Architect platform data, Q2 2026).
This is also the category with the highest security risk. Our Security Scanner flags 31% of consulting prompts for potential client-data leakage — and stakeholder communications are the primary offender. Using `[ANONYMISED]` placeholders for all client-identifiable information reduces flagging rates to under 4%.
22. Board Presentation Script
Write a board presentation script for a [X]-minute slot. Topic: [engagement outcomes/recommendations]. Board composition: [ANONYMISED — describe roles, not names]. Structure: opening context (30 seconds), key finding (60 seconds), evidence walkthrough (3 minutes), recommendation and ask (90 seconds), anticipated questions with prepared responses (3 questions). Tone: executive-appropriate, no consultant jargon. Every number must have a 'so what'.
23. Steering Committee Update
Draft a steerco update for [ANONYMISED — project name]. Period: [dates]. Structure: (1) overall status (RAG with justification), (2) progress against milestones (table format), (3) key decisions required from steerco (maximum 3, each with options and recommendation), (4) risks and issues (new, ongoing, closed), (5) look-ahead for next period. Maximum 2 pages. Decision items must be clearly flagged with the specific ask.
24. Change Communication (By Audience)
Write 3 versions of a change communication for [ANONYMISED — initiative]. Each version targeted at a different audience:
Version 1: Senior leadership (focus on strategic rationale and expected outcomes)
Version 2: Middle management (focus on what changes for their teams and what support is available)
Version 3: Frontline employees (focus on personal impact, timeline, and where to get answers)
Each version: 300 words maximum. Honest tone — acknowledge disruption while conveying purpose. No corporate euphemisms.
25. Escalation Memo
Draft an escalation memo for [ANONYMISED — issue description]. Escalating from [role] to [role]. Structure: issue summary (3 sentences), impact if unresolved (quantified), actions taken to date, options for resolution (minimum 2 with pros/cons), recommended option, specific ask of the recipient. Tone: factual and solution-oriented. Do not assign blame. Maximum 400 words.
26. Project Status Report
Generate a weekly project status report. Project: [ANONYMISED]. Week ending: [date]. Sections: executive summary (3 sentences), workstream progress (table: workstream, status RAG, % complete, key activity, next milestone), budget tracking (planned vs. actual), team utilisation, risks (likelihood × impact matrix), dependencies, and actions for next week with owners and due dates.
27. Risk Communication
Write a risk communication for [ANONYMISED — stakeholder group] regarding [risk event/finding]. Structure: what has happened or been identified (factual, no speculation), potential impact (be specific), what we are doing about it (actions with timelines), what we need from the recipient, and next communication checkpoint. Tone: transparent, calm, accountable. Under 500 words.
28. Stakeholder Mapping Analysis
Create a stakeholder mapping analysis for [ANONYMISED — initiative]. Identify key stakeholders by category: sponsors, decision-makers, influencers, affected parties, and potential blockers. For each stakeholder group, assess: level of influence (high/medium/low), current sentiment (supportive/neutral/resistant), key concerns, preferred communication channel, and recommended engagement approach. Present as a structured table with an accompanying engagement strategy narrative.
Research & Knowledge Management: 7 Templates
Research prompts benefit enormously from output-framing instructions. Prompts that request 'so what' framing — explicitly asking for actionable conclusions rather than summaries — produce outputs rated 2.7x more actionable by evaluators. Taxonomy-structured prompts generate 53% more consistent tagging across knowledge assets (AI Prompt Architect platform data, Q2 2026).
29. Literature Review Synthesis
Synthesise the following sources on [topic] into a structured literature review. Sources: [paste or describe]. Structure: (1) key themes identified across sources, (2) areas of consensus, (3) areas of disagreement or tension, (4) gaps in the existing research, (5) 'so what' — implications for [client's specific strategic question]. Do not summarise each source sequentially — synthesise across them thematically.
30. Market Sizing (Top-Down and Bottom-Up)
Estimate the market size for [product/service category] in [geography]. Provide both top-down and bottom-up approaches. For each: state assumptions explicitly, show the calculation chain, provide a range (conservative/base/optimistic), and identify the assumption with the highest sensitivity. Flag where you are uncertain and what primary research would reduce that uncertainty. Present in a format suitable for a strategy deck appendix.
31. Best-Practice Extraction
Extract best practices from the following [case studies/examples/data]: [paste or describe]. For each best practice identified: (1) describe the practice, (2) provide the evidence supporting it, (3) assess applicability to [ANONYMISED — client context], (4) identify prerequisites or conditions for successful adoption, (5) note potential barriers. Rank by expected impact. Maximum 8 best practices.
32. Case Study Template (Anonymised)
Write an anonymised case study for [engagement type]. Structure: client context (sector, size, geography — no identifiable details), challenge, approach (methodology and key activities), results (quantified where possible), key success factors, and lessons learned. Length: 600 words. Suitable for inclusion in proposals and capability materials. Verify that no combination of details could identify the client.
33. Lessons-Learned Framework
Create a structured lessons-learned document for [ANONYMISED — completed engagement]. Categories: what worked well (continue), what could be improved (adjust), what should be done differently (change), and unexpected outcomes (positive and negative). For each lesson: describe the event, analyse the root cause, state the lesson, and recommend a specific action for future engagements. Prioritise lessons with the broadest applicability across the practice.
34. Competitive Intelligence Brief
Compile a competitive intelligence brief on [ANONYMISED — competitor or competitor segment]. Structure: company/segment overview, recent strategic moves (last 12 months), product/service portfolio analysis, market positioning, known client wins/losses, leadership changes, financial performance indicators (use public data only), and assessed strategic direction. Conclude with: implications for our client and recommended response options. Flag confidence levels for each data point.
35. Industry Trend Analysis
Analyse [3–5 major trends] shaping [industry] over the next [X] years. For each trend: describe the trend and its drivers, assess the pace of change (accelerating/steady/uncertain), identify winners and losers, quantify potential impact where possible, and provide the 'so what' for a [player type] in this space. Conclude with a synthesis: which trends are mutually reinforcing, and what is the combined strategic implication?
Prompt Chaining: The Full Engagement Lifecycle
Individual prompts are useful. Chained prompts — where the output of one becomes the input context for the next — are where consulting AI use matures considerably. The engagement lifecycle follows a natural chain:
| Phase | Template Category | Chain Logic |
|---|---|---|
| 1. Business Development | Templates 1–7 | Proposal wins → engagement context feeds into… |
| 2. Strategic Analysis | Templates 8–14 | Analysis outputs feed into… |
| 3. Client Deliverables | Templates 15–21 | Deliverables require communication via… |
| 4. Stakeholder Communication | Templates 22–28 | Insights from comms feed back into research… |
| 5. Research & Knowledge | Templates 29–35 | Knowledge assets strengthen future BD |
The practical workflow: use Template 8 (SWOT) to generate initial strategic analysis. Feed those findings into Template 15 (pyramid-principle deck) to structure the client deliverable. Extract the key decision items and use Template 23 (steerco update) to communicate progress. After engagement completion, run Template 33 (lessons-learned) and Template 32 (case study) to capture knowledge for future BD efforts.
This chained approach compounds quality improvements. Each subsequent prompt benefits from the structured context generated by the previous one, and the STCO framework provides a systematic methodology for maintaining coherence across the chain. For a deeper look at chaining techniques and design patterns, see our guide on prompt template design patterns.
Score & Optimise Your Consulting Prompts
Writing the prompt is half the process. Scoring and iterating is what separates adequate AI outputs from partner-ready deliverables.
Our platform data is unambiguous on this point: two scoring iterations produce outputs rated 'partner-ready' 3.1x more often than first-pass outputs. Ninety seconds of rescoring saves ninety minutes of manual revision (AI Prompt Architect platform data, Q2 2026).
The Prompt Scorer evaluates consulting prompts across four dimensions particularly relevant to this vertical:
| Dimension | What It Measures | Why It Matters |
|---|---|---|
| Client-Readiness | Could this output be shown to a client with minimal editing? | Reduces the revision cycle that consumes analyst and associate time |
| Analytical Rigour | Are conclusions supported by evidence and structured logically? | Prevents the 'so what?' challenge in partner reviews |
| Strategic Depth | Does the output go beyond surface-level observation? | Differentiates AI-augmented analysis from a quick internet search |
| Communication Clarity | Is the output structured for its intended audience? | Ensures deliverables land with the right stakeholders |
The iterative process is straightforward: run your prompt through the Prompt Scorer, review the dimensional scores, refine the weakest dimension, and rescore. Two iterations is typically sufficient. Three is warranted for high-stakes deliverables — final recommendations decks, board presentations, or proposal submissions where the commercial stakes justify the additional ninety seconds.
For a comprehensive guide to the scoring methodology and prompt engineering best practices, including techniques applicable beyond consulting, our best-practices guide covers the full optimisation workflow.
Frequently Asked Questions
How do I use AI for consulting work without exposing client data?
Replace all client-identifiable information with `[ANONYMISED]` placeholders before submitting any prompt. Use descriptive substitutes — `[ANONYMISED — sector: mid-market manufacturing, geography: Northern Europe]` — that preserve enough context for relevant output without revealing identity. Our Security Scanner detects potential leakage and should be run on every prompt. Using anonymised placeholders reduces flagging rates from 31% to under 4%.
Which strategic framework works best with AI prompts?
It depends on the analytical question, not the AI. SWOT, Porter's Five Forces, PESTLE, and value chain analysis all produce strong AI-assisted outputs when the prompt includes specific data rather than requesting the framework generically. Our data suggests that data-injected Porter's Five Forces prompts produce 2.1x higher-quality outputs, while time-horizon SWOT prompts are 56% more actionable. The key variable is the specificity of your input, not the framework itself.
Can AI write partner-ready consulting deliverables?
Not in a single pass, and likely not without human refinement. AI can produce strong first drafts that significantly reduce the time from blank page to review-ready document. Our platform data shows that two scoring iterations make outputs 3.1x more likely to be rated 'partner-ready' — but 'partner-ready' still assumes a partner reviews and applies their judgement before client submission. AI accelerates the scaffolding; the consultant provides the calibration.
How should I anonymise consulting data for AI prompts?
Use a consistent anonymisation protocol: replace company names with `[ANONYMISED — sector/size]`, replace revenue figures with `[ANONYMISED — range: $500M–$1B]`, replace individual names with role titles, and replace proprietary strategy details with generalised descriptions. Before submitting, ask yourself: could anyone reading this prompt identify the client? If the answer is possibly, anonymise further. Run the prompt through a security scanner as a final check.
How does MECE structuring work with AI prompts?
Explicitly instruct the AI to structure outputs in a MECE (mutually exclusive, collectively exhaustive) manner. For example: `'Categorise these findings into MECE buckets. Each item belongs in exactly one category. The categories together must cover all findings.'` AI models respond well to MECE instructions because the concept maps directly to logical partitioning. Our data shows that prompts requesting MECE structure produce outputs with 28% fewer overlapping categories and more complete coverage. Pair MECE instructions with the pyramid principle for maximum analytical clarity.
Further Reading
Get the Prompt Engineering Playbook
Join 5,000+ developers receiving our weekly deep-dives on structured outputs, RAG optimisation, and advanced AI agent prompting.
AI Prompt Architect
AuthorExpert in prompt architecture and large language model optimization.