ChatGPT Prompts for Consultants
Thirty battle-tested prompts that take you from a messy client problem to a structured analysis, a board-ready deck, and a signed proposal: framing, MECE trees, recommendations, and SOWs included.
In short: This page contains 30 copy-paste ready prompts, organized into 6 categories with a description and pro tip for each. The first 5 prompts are free instantly, no signup needed. Drafted with AI writing assistance and checked against our editorial policy.
Problem Framing
5 promptsSharpen the core question
1/30✨ What it does
ChatGPT turns the fuzzy complaint from [CLIENT], a [INDUSTRY] company, into one decision-oriented key question plus alternative framings of [PROBLEM]. You pick the question the sponsor will actually answer, then put it at the top of the kickoff.
<context> My client [CLIENT], a [INDUSTRY] company, describes their problem as: [PROBLEM]. This is the vague version they gave me in the kickoff. I need a precise, answerable question to anchor the engagement. </context> <task> 1. Restate the problem as a single, decision-oriented key question that starts with 'How should', 'Should', or 'What is the best way to'. 2. List the implicit assumptions baked into the client's framing and flag any that look risky. 3. Propose 3 alternative framings (broader, narrower, reframed-from-a-different-stakeholder) and note what each would change about the scope. 4. Recommend one framing and justify it in 2 sentences tied to what [CLIENT] can actually decide and act on. </task>
Pro tip: Paste the raw kickoff transcript before this prompt so ChatGPT frames against the client's actual words, not your paraphrase.
Build the SCQ storyline opener
2/30✨ What it does
ChatGPT writes a tight Situation-Complication-Question opener for the [CLIENT] kickoff in [INDUSTRY] so stakeholders align on [PROBLEM] before the work starts. You read it aloud in the room, then change any sentence someone will not accept.
<context> Engagement for [CLIENT] in [INDUSTRY]. The core problem is [PROBLEM]. I need a Situation-Complication-Question setup to open the kickoff deck and align stakeholders. </context> <task> 1. Write the Situation: 2-3 sentences of stable, agreed-upon context no stakeholder would dispute. 2. Write the Complication: the change or tension that makes the situation no longer acceptable. 3. Write the Question: the single question that naturally follows and that the engagement will answer. 4. Add a one-line 'so what' that previews why answering it matters financially or strategically to [CLIENT]. </task>
Pro tip: Ask ChatGPT to give three Complication options at different severity levels, then pick the one that matches the room's urgency.
Map stakeholders and their hidden agendas
3/30✨ What it does
ChatGPT maps likely [CLIENT] stakeholders on [PROBLEM] with influence and support ratings, plus a recommended engagement order. You confirm the names with your sponsor in [INDUSTRY], then book the high-influence conversations first.
<context> I am scoping an engagement for [CLIENT], a [INDUSTRY] firm, on [PROBLEM]. Decisions here are political as much as analytical. </context> <task> 1. List the likely stakeholder roles involved (sponsor, budget owner, blockers, influencers, end-users). 2. For each, infer their stated goal vs. their likely private incentive. 3. Rate each on a 2x2 of influence vs. support for change. 4. Recommend who to interview first, who to co-opt early, and one risk per high-influence/low-support stakeholder. </task>
Pro tip: Feed ChatGPT the org chart or LinkedIn titles first so the role inferences are grounded in the real team.
Define scope boundaries and out-of-scope
4/30✨ What it does
ChatGPT drafts crisp in-scope and out-of-scope lists for the [CLIENT] engagement on [PROBLEM], plus language you can use when someone tries to widen it. You put those lists in the unsigned SOW, then say the out-of-scope line when creep starts.
<context> Engagement for [CLIENT] in [INDUSTRY] on [PROBLEM]. Scope creep is my biggest risk and the SOW is not signed yet. </context> <task> 1. Draft an in-scope list: the specific questions and deliverables this engagement will cover. 2. Draft an explicit out-of-scope list of adjacent things the client may assume are included but are not. 3. Identify 3 grey-zone items likely to trigger creep and propose how to handle each (change order, fast-follow, parking lot). 4. Write one sentence I can say in the kickoff to set the boundary diplomatically. </task>
Pro tip: Have ChatGPT phrase the out-of-scope items as benefits ("so we can go deep on what moves the needle") to avoid sounding restrictive.
Stress-test the problem before you start
5/30✨ What it does
ChatGPT pressure-tests [PROBLEM] for [CLIENT] in [INDUSTRY] with 5 Whys and five disconfirming questions so you do not burn hours on the wrong issue. You ask those questions before staffing the team, then reframe if the root cause is different.
<context> Before committing the team to [PROBLEM] for [CLIENT] in [INDUSTRY], I want to pressure-test whether this is even the right problem to solve. </context> <task> 1. Apply 5-whys to the stated problem and surface the likely root cause. 2. Ask 5 disconfirming questions a skeptical partner would ask in a problem-validation review. 3. Identify what evidence would prove the problem is real and material vs. anecdotal. 4. Conclude: proceed, reframe, or push back to the client, with a one-line rationale. </task>
Pro tip: Run this on day one and keep the output; revisit it at the midpoint review to check the problem has not drifted.
Research & Analysis
5 promptsStructure a market sizing
6/30✨ What it does
ChatGPT lays out top-down and bottom-up market sizing for [PROBLEM] at [CLIENT] in [INDUSTRY], with an assumption ledger and a defensible range. You replace the thin inputs with better data, then quote the range, not a fake single number.
<context> I need to size the market relevant to [PROBLEM] for [CLIENT] in [INDUSTRY]. I have limited data and need a defensible estimate fast. </context> <task> 1. Build both a top-down and a bottom-up sizing approach, listing the inputs each needs. 2. State the assumptions for every input and flag which ones are weakest. 3. Produce a base / low / high estimate range and explain the driver of the spread. 4. List the 3 data sources I should chase to tighten the weakest assumptions. </task>
Pro tip: Ask ChatGPT to show the math as a calculation chain so you can drop the numbers straight into a model and audit them.
Synthesize interview notes into themes
7/30✨ What it does
ChatGPT clusters the [CLIENT] interview notes on [PROBLEM] into named insight themes with support, dissent, and open questions, not a recap. You check each theme against the raw notes, then take the open questions into the next interviews.
<context> I ran stakeholder interviews for [CLIENT] in [INDUSTRY] about [PROBLEM]. I will paste raw notes below. I need synthesis, not a summary. </context> <task> 1. Cluster the notes into 4-6 themes and name each theme as an insight, not a topic. 2. For each theme, cite which interviewees support it and note any dissent. 3. Flag the 2 most surprising or counter-consensus findings. 4. List the open questions the interviews did NOT answer and who could answer them. [PASTE NOTES BELOW] </task>
Pro tip: Tell ChatGPT to write each theme as a full-sentence finding ("Sales blames pricing, not product") so it can become a deck headline later.
Competitive teardown
8/30✨ What it does
ChatGPT produces a structured competitive teardown for [CLIENT] in [INDUSTRY] around [PROBLEM], plus one highest-leverage strategic move. You pressure-test that move with the sponsor, then build the rest of the work around it.
<context> [CLIENT] competes in [INDUSTRY] and is wrestling with [PROBLEM]. I need a structured competitive view to inform strategy. </context> <task> 1. Identify the 4-6 most relevant competitors and the basis on which they compete. 2. Build a comparison across dimensions that matter for [PROBLEM] (positioning, pricing, GTM, capability gaps). 3. Highlight where [CLIENT] has a defensible edge and where it is exposed. 4. Recommend the single competitive move that would most change [CLIENT]'s position, with the trade-off. </task>
Pro tip: Use ChatGPT search/browsing to pull current competitor pricing and positioning, then ask it to flag anything it could not verify.
Quantify the cost of inaction
9/30✨ What it does
ChatGPT estimates an annualized cost of inaction on [PROBLEM] for [CLIENT] in [INDUSTRY], with a punchy headline figure and the mechanisms behind it. You swap in their real numbers, then put that figure on the first slide.
<context> [CLIENT] in [INDUSTRY] is dragging on [PROBLEM]. I need to quantify the cost of doing nothing to create urgency. </context> <task> 1. List the mechanisms through which inaction costs money (lost revenue, churn, inefficiency, risk exposure). 2. For each, build a simple estimation formula and plug in placeholder assumptions. 3. Total an annualized cost-of-inaction range and identify the single biggest driver. 4. Phrase the headline number as a one-line statement I can use to open the recommendation. </task>
Pro tip: Ask ChatGPT to keep every assumption conservative and labeled, so a CFO challenging the number finds it bulletproof, not inflated.
Build the analysis plan
10/30✨ What it does
ChatGPT turns the key question for [CLIENT] on [PROBLEM] into a sequenced analysis plan with sources and a critical path. You staff the critical-path work first, then drop analyses that do not answer a sub-question.
<context> Engagement for [CLIENT] in [INDUSTRY] on [PROBLEM]. I have the key question and now need a plan for what analyses will actually answer it. </context> <task> 1. Decompose the key question into the 3-5 sub-questions that must be answered. 2. For each sub-question, specify the analysis required, the data needed, and the source. 3. Sequence the analyses by dependency and effort, flagging the critical path. 4. Identify the one analysis most likely to be the 'answer-first' driver and prioritize it. </task>
Pro tip: Save the output as a workplan table; ask ChatGPT to add an owner and a due-day column so it doubles as a team tracker.
Frameworks (MECE, Issue Trees, Hypotheses)
5 promptsBuild a MECE issue tree
11/30✨ What it does
ChatGPT builds a two-level MECE issue tree for how [CLIENT] should solve [PROBLEM], with overlap and gap checks and priority flags. You use the tree to assign workstreams in [INDUSTRY], then refuse branches that overlap.
<context> Key question for [CLIENT] in [INDUSTRY]: how to solve [PROBLEM]. I need a MECE issue tree to structure the analysis. </context> <task> 1. Decompose the key question into first-level branches that are mutually exclusive and collectively exhaustive. 2. Break each branch into a second level of sub-issues, keeping MECE at each node. 3. Flag any branch that overlaps or leaves a gap, and fix it. 4. Mark which leaf nodes are likely the highest-impact to investigate first. </task>
Pro tip: Ask for the tree as a nested bullet outline; it pastes cleanly into a slide or a Miro board without reformatting.
Generate and rank hypotheses
12/30✨ What it does
ChatGPT proposes ranked, testable hypotheses for [PROBLEM] at [CLIENT], each with kill criteria and a fast disconfirming test. You run the cheapest kill tests first in [INDUSTRY], then drop hypotheses that fail.
<context> [CLIENT] in [INDUSTRY] faces [PROBLEM]. I want to run a hypothesis-driven engagement instead of boiling the ocean. </context> <task> 1. Propose 5-7 candidate hypotheses that could explain or solve the problem, phrased as testable statements. 2. For each, state what evidence would confirm or kill it. 3. Rank them by likelihood x impact and recommend which 2-3 to test first. 4. Note the single fastest, cheapest test that could disprove the leading hypothesis. </task>
Pro tip: Lead with the hypothesis you already suspect and ask ChatGPT to argue the opposite, so confirmation bias does not pick your tests for you.
Apply the right framework
13/30✨ What it does
ChatGPT recommends the best-fit strategy framework for [PROBLEM] at [CLIENT] in [INDUSTRY] and pre-fills its structure so you do not force the wrong one. You fill the empty cells with evidence, then present from that structure.
<context> I am analyzing [PROBLEM] for [CLIENT] in [INDUSTRY] and want to avoid forcing the wrong framework onto it. </context> <task> 1. List 4-5 frameworks that could plausibly apply (e.g., Porter, value chain, 3C, jobs-to-be-done, profit tree). 2. For each, state what it would reveal and where it would fall short for this specific problem. 3. Recommend the single best-fit framework and one supporting framework. 4. Sketch how the recommended framework maps onto [PROBLEM] with the actual branches filled in. </task>
Pro tip: Push back if ChatGPT defaults to SWOT or a generic 2x2; ask it to justify why a sharper, problem-specific structure would not be better.
Build a driver/profit tree
14/30✨ What it does
ChatGPT builds a mathematically exact driver tree for the financial metric behind [PROBLEM] at [CLIENT], and surfaces the highest-leverage levers they can actually move. You pick the two levers [INDUSTRY] can change this year, then model those first.
<context> [PROBLEM] for [CLIENT] in [INDUSTRY] is fundamentally about a financial metric (e.g., profit, margin, CAC). I need a quantitative driver tree. </context> <task> 1. Decompose the target metric into its mathematical drivers, level by level, keeping the math exact. 2. At each leaf, note whether [CLIENT] can realistically move it (high / medium / low control). 3. Identify the 2-3 leaves with the best impact-times-controllability score. 4. Translate those leaves into the specific analyses needed to quantify the opportunity. </task>
Pro tip: Give ChatGPT the actual metric definition the client uses; a slightly different margin definition breaks the tree math downstream.
MECE-check my existing structure
15/30✨ What it does
ChatGPT audits your draft structure for [PROBLEM] at [CLIENT] in [INDUSTRY] for MECE violations and returns a minimally changed corrected version. You replace the overlapping buckets, then present the corrected tree.
<context> I already built a structure for [PROBLEM] at [CLIENT] in [INDUSTRY]. I will paste it below. I want a rigorous MECE critique before I present it. </context> <task> 1. Check every level for mutual exclusivity and flag overlapping buckets. 2. Check for collective exhaustiveness and flag missing categories. 3. Point out any bucket that mixes levels of abstraction or that is a solution masquerading as a category. 4. Propose a corrected version that keeps as much of my original as possible. [PASTE STRUCTURE BELOW] </task>
Pro tip: Run this the night before a partner review; the most common ding ("these buckets overlap") gets caught here for free.
Like these prompts? There are full tutorials behind them.
Learn the workflows, not just the prompts. Easy-to-follow tutorials inside AI Academy, and the library keeps growing.
Client Decks & Storylining
5 promptsBuild the pyramid storyline
16/30✨ What it does
ChatGPT organizes the [CLIENT] findings on [PROBLEM] into a Minto pyramid with one governing thought and tested supporting logic. You paste your findings below the prompt, then build the [INDUSTRY] readout from that pyramid.
<context> Final readout for [CLIENT] in [INDUSTRY] on [PROBLEM]. I have my findings (pasted below) and need a Minto pyramid storyline. </context> <task> 1. State the single governing thought (the one-sentence answer the whole deck supports). 2. Build the 3-4 supporting arguments directly beneath it, in MECE order. 3. Under each argument, list the evidence/slides that prove it. 4. Verify the logic flows bottom-up (evidence supports argument) and top-down (governing thought is fully answered). Flag any gap. [PASTE FINDINGS BELOW] </task>
Pro tip: Ask ChatGPT to render the storyline as a slide-by-slide list so each supporting argument becomes one section divider.
Write action-title slide headlines
17/30✨ What it does
ChatGPT rewrites the [CLIENT] slide titles on [PROBLEM] as action titles that tell the story alone, each a full-sentence takeaway. You replace the topic labels in the deck, then read only the titles to see if the story holds.
<context> I have a deck for [CLIENT] in [INDUSTRY] about [PROBLEM]. My slide titles are topic labels, not insights. I will paste them below. </context> <task> 1. Rewrite each title as an action title: a complete sentence stating the slide's single takeaway. 2. Ensure the titles, read in sequence, tell the whole story without the body content. 3. Flag any slide whose title cannot be made into a clear takeaway (it may not belong). 4. Return the original and rewritten title side by side. [PASTE TITLES BELOW] </task>
Pro tip: Read just the rewritten titles aloud to a colleague; if they understand the recommendation without the deck, the storyline works.
Design a single slide
18/30✨ What it does
ChatGPT specs one exec-level slide for [CLIENT] in [INDUSTRY] that proves [PASTE THE POINT] about [PROBLEM], with an action title, chart choice, content, and emphasis. You build that single slide, then cut anything that does not prove the point.
<context> I need one slide for [CLIENT] in [INDUSTRY] that makes this point about [PROBLEM]: [PASTE THE POINT]. The deck is exec-level. </context> <task> 1. Write the action title for the slide. 2. Recommend the single best chart or visual type to prove the point and why. 3. Specify exactly what goes on the slide: the data shown, the annotation/callout, and what to leave off. 4. Suggest the one number or phrase to emphasize visually. </task>
Pro tip: Tell ChatGPT the slide must pass the "squint test" so it keeps the visual to one chart and one message, not a dashboard.
Tighten an overloaded deck
19/30✨ What it does
ChatGPT cuts and re-sequences the bloated [CLIENT] readout on [PROBLEM] down to a tight spine, with appendix and Q&A backups. You drop the cut slides into the appendix, then present the 15-slide version to the partner.
<context> My readout for [CLIENT] in [INDUSTRY] on [PROBLEM] is 40 slides and the partner wants 15. I will paste the slide titles below. </context> <task> 1. Map each slide to the governing thought and flag any that do not directly support it. 2. Identify slides to cut, slides to merge, and slides to move to the appendix. 3. Propose the tight 15-slide spine in order. 4. Note which 2-3 cut slides to keep in back pocket for likely Q&A. [PASTE TITLES BELOW] </task>
Pro tip: Ask ChatGPT to justify every cut against the governing thought; "interesting but off-message" slides are exactly what bloats consulting decks.
Prep the executive Q&A
20/30✨ What it does
ChatGPT anticipates the 10 toughest questions [CLIENT] leadership will ask about [PROBLEM], including political ones, with on-message answers and honest handling of weak spots. You rehearse those answers before the [INDUSTRY] readout, then admit the weak spots instead of bluffing.
<context> I am presenting findings on [PROBLEM] to [CLIENT]'s leadership in [INDUSTRY] next week. I want to anticipate the hard questions. </context> <task> 1. Generate the 10 toughest questions execs are likely to ask, including the political/defensive ones. 2. For each, draft a 2-3 sentence answer that stays on message. 3. Flag the 2 questions where my evidence is weakest and suggest how to handle them honestly. 4. Note one question I should proactively raise myself to build credibility. </task>
Pro tip: Ask ChatGPT to role-play the most skeptical executive and grill you live before the meeting; it surfaces gaps a static list misses.
Recommendations
5 promptsDraft the recommendation
21/30✨ What it does
ChatGPT writes a single directive recommendation on [PROBLEM] for [CLIENT], backed by the three strongest reasons, risks, mitigations, and a next action. You attach the findings you pasted, then take that recommendation into the [INDUSTRY] decision meeting.
<context> Based on my analysis of [PROBLEM] for [CLIENT] in [INDUSTRY] (findings pasted below), I need a sharp, defensible recommendation. </context> <task> 1. State the recommendation as a single clear directive answer to the key question. 2. Give the 3 strongest reasons it is right, each tied to evidence from the findings. 3. Name the top 2 risks and the mitigation for each. 4. State explicitly what [CLIENT] must do differently on Monday morning if they accept it. [PASTE FINDINGS BELOW] </task>
Pro tip: Make ChatGPT commit to ONE recommendation; if it hedges with "it depends," ask it to state the recommendation for the most likely scenario.
Compare options with a decision matrix
22/30✨ What it does
ChatGPT builds a weighted decision matrix for the paths [CLIENT] can take on [PROBLEM], scores each option, and names the conditions that would tip the choice. You confirm the weights with the sponsor in [INDUSTRY], then recommend the option that wins under those weights.
<context> [CLIENT] in [INDUSTRY] must choose between several paths to address [PROBLEM]. I have the options (pasted below) and need an objective comparison. </context> <task> 1. Define the 4-6 decision criteria that actually matter to [CLIENT], weighted by importance. 2. Score each option against each criterion, with a one-line justification per cell. 3. Compute a weighted total and identify the leading option. 4. State the conditions under which the runner-up would win instead. [PASTE OPTIONS BELOW] </task>
Pro tip: Set the weights with the client before scoring; a matrix the client co-built is far harder for them to argue away later.
Build the implementation roadmap
23/30✨ What it does
ChatGPT generates a phased implementation roadmap for the accepted [CLIENT] recommendation on [PROBLEM], with quick wins, owners, and an early-warning risk indicator. You assign the 0-90 day owners in [INDUSTRY], then watch that indicator weekly.
<context> [CLIENT] in [INDUSTRY] accepted the recommendation on [PROBLEM]. Now they need a realistic roadmap to execute it. </context> <task> 1. Break execution into phases (quick wins / 0-90 days, build / 3-9 months, scale / 9+ months). 2. For each phase list the key initiatives, owners, and dependencies. 3. Identify the 2-3 quick wins that build momentum and credibility early. 4. Flag the single biggest execution risk and the leading indicator to watch for it. </task>
Pro tip: Ask ChatGPT to front-load a visible quick win in the first 30 days; early momentum is what keeps a recommendation from dying in committee.
Build the business case
24/30✨ What it does
ChatGPT builds a costed business case for funding the [CLIENT] recommendation on [PROBLEM], with ROI scenarios and a CFO-ready summary. You replace placeholder costs with finance numbers, then send the one-pager to [INDUSTRY] leadership.
<context> Leadership at [CLIENT] in [INDUSTRY] needs a business case to fund the recommendation on [PROBLEM]. </context> <task> 1. Lay out the costs (one-time and recurring) and the benefits (quantified where possible, qualitative where not). 2. Build a simple ROI / payback view with stated assumptions. 3. Run a base / conservative / optimistic scenario and show the swing. 4. Summarize the case in 3 sentences a CFO would forward to approve funding. </task>
Pro tip: Have ChatGPT label every benefit as "committed" vs "potential"; mixing the two is the fastest way to lose CFO trust.
Anticipate objections to the recommendation
25/30✨ What it does
ChatGPT maps likely objections to the [PROBLEM] recommendation across rational, emotional, and political lines for [CLIENT], with a pre-emptive response for each. You rehearse those responses before the [INDUSTRY] meeting, then use them when the pushback starts.
<context> Before I present my recommendation on [PROBLEM] to [CLIENT] in [INDUSTRY], I want to pre-empt the pushback. </context> <task> 1. List the 6 most likely objections, grouped into rational, emotional, and political. 2. For each, draft a respectful, evidence-based response. 3. Identify which objection, if unaddressed, is most likely to kill the recommendation. 4. Suggest one slide or talking point to neutralize that objection before it is raised. </task>
Pro tip: Ask ChatGPT to flag the political objections specifically; analytical consultants tend to over-prepare for logic and under-prepare for turf.
Go from copy-pasting to actually mastering AI.
AI Academy: hands-on tutorials on ChatGPT, Claude, Midjourney, and many other tools. New tutorials are added regularly.
Proposals & SOWs
5 promptsDraft the proposal
26/30✨ What it does
ChatGPT drafts a structured consulting proposal for [CLIENT] in [INDUSTRY] that restates [PROBLEM] in their language and runs through to a call to action. You swap in fees and dates, then send it only after the problem restatement sounds like them.
<context> A prospect, [CLIENT] in [INDUSTRY], wants a proposal to address [PROBLEM]. I need a persuasive, well-structured draft. </context> <task> 1. Open with a restatement of [CLIENT]'s problem and desired outcome in their language, proving I understood the brief. 2. Lay out the proposed approach in clear phases with the value each delivers. 3. Specify deliverables, timeline, and the team/roles involved. 4. Close with why me/my firm specifically, and a clear call to action to proceed. </task>
Pro tip: Make ChatGPT open with the client's own words from the brief; proposals that mirror the prospect's language convert far better than generic ones.
Write the scope of work
27/30✨ What it does
ChatGPT drafts a defensible statement of work for [CLIENT] on [PROBLEM], with measurable deliverables, acceptance criteria, exclusions, and a change-request process. You put those exclusions in the signed SOW, then point to them when [INDUSTRY] asks for extras.
<context> I need a tight statement of work for [CLIENT] in [INDUSTRY] on [PROBLEM] that protects me from scope creep. </context> <task> 1. Write specific, measurable deliverables with acceptance criteria for each. 2. List explicit exclusions and assumptions (client responsibilities, data access, availability). 3. Define the change-request process and how out-of-scope work gets priced. 4. Specify milestones, payment triggers, and what 'done' means for the engagement. </task>
Pro tip: Ask ChatGPT to write acceptance criteria as observable outcomes ("client sign-off on X"), not effort ("we will work on X"), to avoid endless revisions.
Price the engagement
28/30✨ What it does
ChatGPT compares fixed-fee, time-and-materials, and value-based pricing for the [CLIENT] engagement on [PROBLEM], sets a cost floor, and proposes a good-better-best tier. You pick the model that protects the floor, then quote the middle tier first.
<context> I am pricing an engagement for [CLIENT] in [INDUSTRY] to solve [PROBLEM]. I want to think beyond hourly rates. </context> <task> 1. Lay out 3 pricing models (fixed fee, time-and-materials, value-based) with pros/cons for this engagement. 2. Estimate the cost-to-deliver as a floor so I never price below it. 3. Anchor a value-based price to the quantified upside for [CLIENT]. 4. Recommend a model and a good-better-best tier structure with what changes between tiers. </task>
Pro tip: Ask ChatGPT to tie each tier to a different client outcome, not just more hours; outcome-anchored tiers justify premium pricing.
Respond to an RFP
29/30✨ What it does
ChatGPT maps each RFP requirement from [CLIENT] to a compliant, differentiated response on [PROBLEM], built around one win theme. You paste the RFP requirements, then fill any compliance gap before you submit in [INDUSTRY].
<context> [CLIENT] in [INDUSTRY] issued an RFP related to [PROBLEM]. I will paste the requirements below. I need a compliant, differentiated response. </context> <task> 1. Build a compliance matrix mapping each RFP requirement to where my response addresses it. 2. Draft responses that meet each requirement AND insert a differentiator where possible. 3. Flag any requirement I cannot fully meet and suggest how to position around it. 4. Identify the win theme that should run through the entire response. [PASTE RFP REQUIREMENTS BELOW] </task>
Pro tip: Ask ChatGPT to extract the evaluation criteria and their weights from the RFP first, then write hardest toward the highest-weighted ones.
Write the follow-up after a pitch
30/30✨ What it does
ChatGPT writes a post-pitch follow-up to [CLIENT] that recaps the one outcome they care about on [PROBLEM], handles the main concern, and proposes a next step. You send it the same day, then stop chasing if they decline the step.
<context> I just pitched [CLIENT] in [INDUSTRY] on solving [PROBLEM]. I want a follow-up that advances the deal without being pushy. </context> <task> 1. Draft a concise follow-up email that recaps the one outcome they care most about. 2. Address the main concern or question raised in the meeting. 3. Attach a clear, low-friction next step with a proposed date. 4. Include a one-line subject line and a softer alternative in case they went quiet. </task>
Pro tip: Tell ChatGPT the single objection you heard in the room so the follow-up resolves it directly instead of sending a generic "just checking in".
Free tool
Prompt Optimizer
Turn a rough idea into a structured, professional AI prompt.
Frequently Asked Questions
Prompts are the starting line. Tutorials are the finish.
A growing library of hands-on tutorials on ChatGPT, Claude, Midjourney, and many other AI tools. New tutorials are added regularly.
7-day free trial. Cancel anytime.
Related guides