AI for consultants: 12 prompts for client work

I'm a consultant starting work with a client in [industry, e.g. UK social housing]. Give me a briefing: how organisations in this sector make and spend money, the five biggest pressures on them right now, the jargon I'll hear with plain-English meanings, and the questions a smart outsider should ask in week one. Tell me which points I should verify, because your information may be out of date.

Write an interview guide for a 45-minute discovery interview with a [role, e.g. head of customer service] at [type of organisation]. We're looking at [the problem area, described generally]. Include an opening that puts them at ease, ten open questions that move from their day-to-day to the root causes, follow-up probes for each, and a closing question that surfaces what I didn't think to ask.

Here's a general description of how a process works at a client: [describe the steps in your own words, with no names or confidential figures]. Turn it into a numbered process map as a table: step, who does it, tool used, rough time, and what can go wrong. Then mark the steps where delays, rework or handovers are most likely.

Here is a process as a list of steps: [paste your anonymised steps]. For each step, say whether it's a good fit for AI help, simple automation, or should stay with a person, and why. Be cautious: flag any step where errors would be costly or where personal data is involved. Finish with the two changes likely to give the most benefit for the least risk.

I've finished discovery interviews. These are the themes I heard, in my own words: [list them, anonymised]. Group them into four or five findings. For each, write a one-line headline that states the "so what", the evidence I should cite, and the question a sceptical executive would ask.

My recommendation to a client is: [describe it generally]. Act as a sceptical finance director. Give me the eight hardest questions you'd ask, the weakest point in my logic, and what evidence would change your mind.

A client needs to choose between these options: [list them]. Their priorities are [e.g. cost, speed, risk, staff impact]. Build a comparison table scoring each option against each priority, with a one-line reason for each score. Then state what would have to be true for each option to be the right choice.

Draft a proposal outline for a consulting project. The client is [type of organisation]. Their problem: [in a sentence]. What I'd do: [approach, in a few lines]. Timescale: [weeks]. Write: a summary of their situation in their words, objectives, approach in phases with what they get at the end of each, what I need from them, and assumptions. Leave fees as a placeholder.

Outline a [10]-slide deck for [audience, e.g. an executive team]. The one thing I need them to decide is: [decision]. My main findings are: [list]. Give each slide a headline that is a full sentence making a point, the one chart or evidence it needs, and a note on what to say. Put the recommendation on slide 2.

Write a one-page executive summary. Situation: [two lines]. What we found: [three points]. What we recommend: [the recommendation]. What it costs and returns: [in general terms]. What we need decided: [the ask]. Plain English, no jargon, under 250 words.

Here are my notes from a client meeting, with names replaced by roles: [paste]. List the decisions made, the actions with an owner and a date for each, the open questions, and anything that sounded like a risk. Then draft a short follow-up email to the client confirming them.

Write a weekly status update for a client project. Done this week: [list]. Planned next week: [list]. Risks or blockers: [list]. Decisions needed from the client: [list]. Keep it under 150 words, lead with anything they need to act on, and use a calm, factual tone.