The best AI prompts for professional services, 24 to copy (2026)

A good professional AI prompt for professional services states the role, the client, the goal of the document and the format, and gives you a proposal, executive summary or email ready to review. This library has 24 AI prompts to copy into ChatGPT, Claude, Copilot or Gemini, with examples and a template.

Add Founderz as a preferred source

What you’ll take from this

These AI prompts for professional services firms give you concrete starting points and the method to adapt them to each client:

• 24 AI prompts ready to copy: proposals, executive summaries, cover emails, follow-ups and client-specific personalization.
• Which tool fits which task: ChatGPT for general analysis, Claude for long documents, Copilot inside Word, Outlook and Teams, Gemini inside Google Workspace.
• The five-field method (role, context, task, format and limits) and a reusable template to build your own prompts.
• The limits that are not negotiable: validating figures and commitments, and protecting client data.

If you want to bring AI into your whole team’s processes, AI training for professional services firms at Founderz gives you the foundation. To generate a custom prompt in a minute, use the free generator.

Who these prompts are for and what information they handle well

You work at a law firm, a consultancy or an agency, and every commercial proposal costs you half a morning. These prompts change the starting point: instead of a blank page, you start by reviewing a draft that already has structure, tone and context. The time you save goes to what needs judgment: validating figures, adjusting the commitment to the client and closing the deal.

• Law firms preparing fee proposals and scope notes.
• Consultancies writing executive summaries from long reports.
• Marketing and communications agencies sending proposals and follow-up emails.
• Independent professionals who handle all their sales writing alone.

If you want the big picture before the prompts, this guide to AI tools for consulting firms sets out the full context.

Generative AI writes better when you give it concrete material instead of asking it to invent. These are the inputs it handles best:

• The client brief and the problem they want to solve.
• Meeting notes with the points you agreed.
• Project data: scope, timeline and deliverables.
• The target audience of the document and who will read it.
• Examples of earlier proposals that worked.

If you paste the notes from a 45-minute meeting together with the project scope, the draft already mentions the client’s sector, their problem and the agreed timeline, instead of describing a generic company with a generic challenge. AI does not verify what you give it: if the scope is incomplete, the draft will be too.

ChatGPT prompts for professional services

ChatGPT is the most versatile option to start with: it accepts attached files, tables and long conversations. The text in brackets is an example: replace it with your case and keep the rest. Library reviewed in October 2026. Before pasting anything, replace real names with markers such as [CLIENT].

Commercial proposals

Full proposal from a brief

Act as a senior consultant at [an operations consultancy] with experience writing proposals for [industrial SMEs].
Context: my client is [a logistics company with 120 employees] and wants to [cut the response time of its customer service]. Scope: [a six-week process audit]. Deliverables: [diagnosis, improvement plan and dashboard]. Indicative budget: [$20,000].
Task: write a commercial proposal the managing director can read in five minutes.
Format: executive summary, diagnosis, proposed work, timeline and investment, with a clear title per section. 700 to 900 words, professional and direct tone.
Limits: do not invent figures or success stories; if a data point is missing, mark it [COMPLETE].

Fee proposal and scope note

Act as a partner at a [corporate law] firm writing its fee proposal.
Context: the client is [a startup preparing an investment round] and needs [a shareholder agreement review and due diligence]. Meeting notes: [paste your notes here].
Task: write the scope note and the fee proposal, separating what the service includes from what is excluded.
Format: table with phase, activity, timeline and fees; then three paragraphs with assumptions and exclusions. Formal tone.
Limits: do not promise legal outcomes; use [FEES] where an amount is missing.

Executive summaries

Executive summary of an attached report or proposal

Act as a senior analyst preparing documents for leadership.
Context: I am attaching [a 12-page proposal] addressed to [the management committee of a healthcare client].
Task: summarize it in an executive summary of [200] words that highlights the client's problem, the proposed solution, the timeline and the investment.
Format: four short paragraphs with those headings in bold and a last line with the decision the reader is asked to make.
Limits: use only data from the document; if something is unclear, list it at the end as open questions.

Summary for a non-technical executive

Act as the VP of strategy explaining a project to a board member without a technical background.
Context: I am pasting the technical update of [a data migration project]. Audience: [the CFO].
Task: explain its strategic value and business impact.
Format: five bullets of one sentence each (value, main risk, timeline, cost, pending decision) and a final paragraph with next quarter's steps.
Limits: no technical jargon; if a figure does not appear in the text, do not include it.

Client emails

Personalized first-contact email

Act as the sales lead at [a procurement consultancy].
Context: I connected on LinkedIn last week with [NAME], head of purchasing at [a food company]. We manage procurement processes. Personal detail to mention: [they posted about their new warehouse].
Task: write an email that proposes a [20]-minute call.
Format: subject line under 8 words and a body of 90 words maximum, direct tone with no filler, one single call to action.
Limits: no clichés or sales adjectives.

Follow-up email without pressure

Act as a consultant who looks after a relationship with a prospect.
Context: I sent a proposal to [NAME], head of [operations], [a week] ago and they have not replied. In the meeting they showed interest in [the training plan].
Task: write a follow-up email that does not sound pushy, reminds them of the value without repeating the proposal and offers to answer questions.
Format: 100 words maximum, two specific time slots to talk ([Tuesday morning] and [Thursday afternoon]) and a subject line that does not start with "Follow-up".
Limits: friendly but professional tone.

Claude prompts for professional services

Claude works well with long documents and careful writing, so it is useful when the source material is long or the tone matters. Paste or attach the full text and ask it to cite the source section so you can check each point.

Long documents and formal writing

Analyze a request for proposal (RFP)

Act as the bid manager at [a strategy consultancy].
Context: I am attaching the tender for [a consulting services contract], [60 pages].
Task: extract the mandatory requirements, the scoring criteria with their weight, the deadlines and the points that require specific documentation.
Format: table with requirement, source page, difficulty level (high, medium, low) and what we need to prepare. Then a list of the five questions worth asking the buyer before we submit.
Limits: always cite the section; if a condition is ambiguous, do not interpret it: mark it [AMBIGUOUS].

Careful proposal from several sources

Act as a proposal writer with a formal, clear tone.
Context: I am pasting three sources: [minutes of the client meeting], [a similar earlier proposal] and [our working methodology]. The client is [a clinic chain] and its problem is [scattered patient data].
Task: write a new proposal that reuses the structure of the earlier one but speaks only about the current client.
Format: executive summary, diagnosis, approach, calendar by phase, team, investment. Short sentences, no unnecessary jargon.
Limits: do not copy data from other clients; mark with [VERIFY] any data that is not in the sources.

Executive summary of a 60-page report

Act as an analyst who synthesizes a report for leadership.
Context: I am attaching [the diagnostic report of a digital transformation project], [60 pages].
Task: write an executive summary of [250] words with findings, risks and recommendations, ordered by impact.
Format: three sections with a heading and, at the end of each claim, the page in parentheses.
Limits: do not add your own interpretations; at the end, say which parts of the report you could not summarize and why.

Find which questions my proposal leaves unanswered

Act as the director of [an industrial company] who receives this proposal and must decide whether to approve it.
Context: I am going to paste a commercial proposal for [process consulting services].
Task: read the proposal and tell me which questions remain unanswered, which commitments are ambiguous and which figures or deadlines are missing.
Format: numbered list of questions ordered by importance and, next to each, one sentence I could add to resolve it.
Limits: critique rigorously but do not rewrite the document.

Rewrite in the firm’s voice

Act as an editor at [a law firm] who unifies the style of its documents.
Context: I am pasting [a proposal draft written in a hurry] and [two sample paragraphs in our style].
Task: rewrite the draft imitating the style of the samples: short sentences, active voice, no sales adjectives.
Format: the full rewritten text and, below it, a list of the substantive changes you made, in case any alters a commitment.
Limits: do not change data, deadlines or amounts.

Copilot prompts for professional services

Copilot works inside Word, Outlook, Teams and Excel with the files you have access to. Write the prompt by referring to the open document, email or meeting instead of pasting the text. Features depend on your Microsoft 365 license: check with IT what data your version processes.

Inside Word, Outlook, Teams and Excel

Word: proposal draft from your notes

Act as a consultant writing a commercial proposal.
Context: use the file [Client_meeting_notes.docx] and the template [Proposal_template.docx] from my folder. The client is [a logistics company].
Task: create a draft in this document with the structure of the template and the data from the notes (problem, scope, timeline and deliverables).
Format: respect the template's heading styles; [two pages]; professional and direct tone.
Limits: mark with [COMPLETE] anything that is not in the notes.

Word: executive summary of the open document

Summarize the open document as an executive summary of [200] words for [the management committee]. Include the problem, the solution, the timeline and the investment. Format: four short paragraphs with the heading in bold. Do not add information that is not in the document. At the end, list the figures I should verify.

Outlook: cover email for the proposal

Write a reply to the latest email from [NAME] attaching the proposal [Client_proposal.docx]. Mention that it reflects what we discussed in the meeting on [Tuesday]. Friendly but professional tone, [90] words maximum. Close by inviting them to a [20]-minute call to answer questions and suggest two time slots from my calendar.

Teams: summary of a client meeting

Summarize the Teams meeting [Client meeting, Thursday] in three sections: decisions made, commitments with owner and date, and pending questions. Then write a follow-up email to the client with the agreements, in a professional tone, [120] words maximum. Do not attribute commitments to anyone who did not state them in the meeting.

Excel: fee scenarios for the proposal

Use the table [Project_hours] in this workbook. Create a table with three fee scenarios (basic, standard and extended) from the hours per phase and a rate of [$110 per hour]. Add total and margin columns. Explain in one sentence which assumption changes between scenarios. Do not modify the original data.

Gemini prompts for professional services

Gemini works inside Google Workspace: Docs, Gmail, Drive and Sheets. Name the file or thread you mean and make clear what it should produce. Check which features your Google Workspace plan includes.

Inside Google Docs, Gmail and Drive

Docs: proposal using the Drive template

Act as a consultant preparing a commercial proposal.
Context: use the template [Standard proposal] from my Drive and the notes in the document [Client notes]. The client is [a communications agency] and wants [to organize its quoting process].
Task: write the proposal in this document with the template's sections.
Format: template headings, [two pages], professional and clear tone.
Limits: do not invent timelines or amounts; use [COMPLETE].

Gmail: reply to a request for proposal

Write a reply to the email from [NAME] asking for a proposal for [a process reorganization project]. Confirm that you will send it before [Friday], ask about [available budget and desired start date] and suggest a [15]-minute call. Friendly and professional tone, [100] words maximum.

Docs: executive summary of a Drive document

Summarize the document [Diagnostic report] in my Drive as an executive summary of [200] words for [the managing director]. Structure: problem, key findings, recommendation and next step. Direct tone. If a conclusion depends on a doubtful data point, say so at the end.

Gmail: summarize a client thread

Summarize the email thread with [NAME] from [client company] over the last [three weeks]. Return: what they asked for, what we promised, what is pending and who needs to act. Then propose a draft email that resumes the conversation with a clear next step.

More examples of AI prompts for follow-up and business development

Prompts that work in any tool to keep the relationship alive after you send the proposal.

Three-email follow-up sequence

Act as the sales lead at [a training consultancy].
Context: I sent a proposal for [a training plan for 40 employees] to [NAME], HR director at [a retail company]. The main concern they raised was [the calendar].
Task: write a sequence of three follow-up emails, at [3], [7] and [14] days.
Format: for each email, a subject line and a body of 80 words maximum; each one adds something new (a data point, a case, a calendar alternative) instead of repeating.
Limits: no pressure or artificial urgency; the last email closes the door gracefully.

Reply to a price objection

Act as a sales consultant at [a strategy consultancy].
Context: the prospect, [managing director of a distribution SME], answers my proposal with the objection [the $20,000 investment seems high]. Main value of the proposal: [cut procurement costs by 20%].
Task: write an email that answers the objection without lowering the price right away.
Format: acknowledge the objection in one sentence, compare cost and impact with a data point from their case, offer an alternative (scope in phases) and suggest a call. 120 words maximum.
Limits: do not invent savings; use [DATA] where a figure is missing.

Discovery questions for the first meeting

Act as a consultant preparing the first meeting with a prospect.
Context: the client is [a 60-person real estate services company] and has asked me for a proposal to [improve its sales process].
Task: prepare the discovery questions I should ask so I can write a tailored proposal afterwards.
Format: [10] questions grouped by topic (current situation, goals, budget, decision makers, timeline) and, next to each, which part of the proposal it feeds.
Limits: open questions and no jargon.

Anonymized success story post

Act as the communications lead at [a marketing agency].
Context: I have a success story from [a client in the tourism sector] that cut [cost per booking by 22%]. Client name: not publishable.
Task: write a LinkedIn post that tells the story without identifying the client.
Format: one-line hook, three short paragraphs (situation, what we did, result) and a closing question. 150 words maximum.
Limits: do not exaggerate the result; use only the figure I gave you.

How I built these prompts

A prompt is the instruction you give an AI tool to generate a specific text. It is not a Google search or a loose question: it is a structured order with context, goal and format. The more precise it is, the better the result. Every prompt in this library follows the same five fields, and the difference between generic text and a usable draft almost always lies there.

Reusable template: copy this structure and fill in the brackets for each document.

Act as [ROLE: consultant, lawyer, account manager].
Context: my client is [SECTOR / COMPANY] and needs [PROBLEM]. The project data is [SCOPE, TIMELINE, DELIVERABLES].
Task: write [DOCUMENT TYPE] for [TARGET AUDIENCE].
Format: structure in [SECTIONS], length [X words], tone [PROFESSIONAL / FRIENDLY].
Limits: do not invent data; mark what is missing with [COMPLETE].

From a weak prompt to a strong one

Writing the second prompt takes two minutes and can save you half a morning of drafting.

Five tips for writing your own prompts

1. Choose the verb carefully: “summarize” returns a short text, “write” a full one, “rewrite” starts from something existing and “analyze” returns conclusions.
2. Give real material: meeting notes, the brief and the scope are worth more than any adjective.
3. Ask for one document per message and chain the steps in the same conversation.
4. Ask it to flag what it does not know instead of inventing it.
5. Save the prompt that works and improve it with every use: that is how your prompt library is born.

How the prompt changes by AI tool

The structure of the prompt is the same in every generative AI tool; what changes is where the material lives and how you present it.

• ChatGPT: accepts attached files and long conversations. The prompts on this page work as they are; ask for one change at a time.
• Claude: built for long texts and careful writing. Paste the full document and ask it to cite the source section.
• Copilot: works inside Word, Outlook, Excel and Teams. Refer to the open document, email or meeting instead of pasting the text.
• Gemini: integrates with Google Docs, Gmail and Drive. Name the file or thread and make clear what it should produce.

Comparison table: strengths of each tool by use case

The capabilities and plans of these tools change often: verify the current free and paid versions before you decide.

Beyond writing, these tools cover other use cases: a support chatbot answers frequent client questions and AI agents are starting to chain whole tasks together, which improves response times and customer satisfaction. For a deeper look, search for AI tools that integrate with your CRM and your sales team.

How to build prompts into your proposal workflow

AI adds the most when it fits your process instead of living apart. The point where it helps most is the draft: you go from a blank page to reviewable text in minutes.

1. Gather the inputs: brief, notes and project data.
2. Run the prompt with the template and get the draft.
3. Review, correct figures and adjust the tone.
4. Generate the executive summary and the cover email.
5. Deliver to the client.

Many firms connect this workflow to their CRM: when an opportunity closes, the sales team exports the client data and uses it as the context of the prompt. That improves consistency between what is in the CRM and what is sent, and also the customer experience at every touchpoint.

As an illustration: a small consultancy with ten proposal requests a month might spend three hours writing each one. With structured prompts and a template library, that time can drop to 30 minutes per proposal, freeing the rest for the commercial side. The result does not come from the tool alone, but from having good prompts and a defined process.

Message personalization and a prompt library for recurring documents

Personalizing the message to the recipient raises the chance of a reply. Writing to a finance director is not the same as writing to a marketing manager. Adjust the prompt to the point in the sales process:

• First contact: mention something specific about the prospect, such as an earlier conversation on LinkedIn.
• Sending the proposal: connect the email to the previous meeting.
• Follow-up: recall the value without repeating the whole proposal.

Saving your prompts as templates is the step that turns a one-off test into lasting savings. Group the documents you repeat (standard proposal, executive summary, cover email, follow-up after seven days, success story post) and create one prompt for each. A realistic starting point is to save about 30 prompts: with them you cover most of your weekly writing tasks, and the return on AI depends less on the tool and more on your method.

AI limits and data security in client documents

Artificial intelligence speeds up the draft, but it does not replace your judgment. Before sending any AI-generated document, check:

• The figures: AI can generate incorrect data. Validate every number against your real source.
• Contractual commitments: deadlines, scope and responsibilities need human review.
• The tone with the client: you know the nuance of a business relationship, the machine does not.

Frequently asked questions about AI prompts for professional services

These are the questions that come up most often when you start using AI prompts with ChatGPT, Claude, Copilot or Gemini in daily work.

What are the best professional AI prompts for professional services?

The best professional prompts always include role, context, task, format and limits. The most useful ones cover commercial proposals, executive summaries and client emails. Generic prompts return empty text; prompts that describe your client and your sector produce drafts you can review and send in far less time.

How do I write a prompt to draft a commercial proposal?

Write it in five blocks: role (consultant or account manager), context (client, sector, problem and project data), task (which document you want), format (structure, length and tone) and limits. Add your meeting notes as material. The more specific the context, the less generic the draft.

What is a prompt and why does the verb you choose matter?

A prompt is the instruction you give an AI tool to generate a text. The verb steers the result: “summarize” produces a short text, “write” a full one, “rewrite” starts from something existing and “analyze” returns conclusions. A vague verb returns a vague result.

Which AI is best for professional services firms: ChatGPT, Copilot, Claude or Gemini?

It depends on your work environment. ChatGPT is versatile for proposals and analysis, Claude suits long formal documents, Gemini fits Google Workspace and Copilot fits Microsoft 365. Try two options and keep the one that fits your workflow best.

Can I use these prompts in ChatGPT, Claude, Gemini and Copilot?

Yes. The structure of role, context, task, format and limits works in all four tools. Only the way you provide the material changes: attach or paste in ChatGPT and Claude, refer to the open file in Copilot, name the Drive document in Gemini.

Can I use AI to write executive summaries without reviewing them?

It is not advisable. AI speeds up the draft, but it can generate incorrect figures or alter nuances. In professional services an executive summary is a commitment to the client: always check data, deadlines and tone before sending it.

Is it safe to enter client data into ChatGPT or other AI tools?

It depends on the tool and on your internal policy. Before pasting contracts or sensitive data, check the tool’s data handling policy and your organization’s rules. A safe practice is to anonymize: use markers such as [CLIENT] and add sensitive details only in the final version.

How much time do proposal prompts really save?

It depends on the document and how well you review it. In many cases the draft goes from hours to minutes and your time goes to validating figures and adjusting the commitment. The saving is real when you have a prompt library and a defined process, not when you improvise every time.

Your next step with AI prompts for professional services

Pick one of the prompts in the library, change what is in brackets and try it on an anonymized case. Then change a single field and compare. With structured prompts, you start by reviewing a draft instead of staring at a blank page: the professional judgment stays yours, but the mechanical part speeds up.

If you want to take this beyond one-off writing and apply AI to your team’s processes, structured training helps you do it with a method. Founderz AI training for business teams is built to apply artificial intelligence to proposals, emails and real workflows.