The problem with generic prompts
If you type "write an investment committee memo for this deal" into an AI tool and paste some numbers, you will get something back. It will be structured. It will use the right headings. And it will be obviously, visibly generic — the kind of output that tells your manager you used AI without thinking about it.
The analysts who get noticed for using AI well are not the ones who use it most. They are the ones who give it enough context, structure, and specificity that the output reads like it came from someone who knows the deal. That requires a different kind of prompt.
This article explains the framework. At the end, there is a link to a prompt playbook built specifically for real estate investment and asset management work — 36 templates you can copy, fill in, and run today.
What a good real estate prompt actually looks like
A well-structured prompt for professional real estate work has four components:
- Role: Tell the AI who it is. "You are a real estate investment analyst" produces better output than no instruction at all.
- Context: Give it the specifics — asset class, location, strategy, key numbers. The more precise the input, the more precise the output.
- Task: State exactly what you want, including format, length, and tone. "Write a 150-word downside scenario narrative in the style of an experienced fund manager presenting risk objectively, not alarmingly" is far better than "write about the risks."
- Constraints: Tell it what not to do. "No bullet points — flowing prose only" or "do not overstate the sustainability credentials" prevent the most common AI failure modes in professional documents.
Here is the difference in practice. A weak prompt:
A strong prompt:
The second prompt takes thirty seconds longer to fill in. The output is a genuine first draft rather than a structural skeleton you still have to write from scratch.
The four task areas where AI saves the most time
Based on where junior analysts and associates spend the most hours, these are the four areas where well-constructed AI prompts deliver the clearest productivity gain.
Investment analysis and underwriting. IM summaries, assumption sense-checks, downside scenario narratives, IC memo drafts, comparable evidence summaries, due diligence question lists, debt structure write-ups, negotiation talking points, and return attribution narratives. These tasks are structurally repetitive — the same sections appear in every deal — which makes them ideal for template-based prompting.
Asset management and reporting. Quarterly AM reports, tenancy schedule summaries, investor update letters, board-level commentary, lease event action plans, capex programme summaries, hold/sell recommendation memos, void management notes, and annual business plan sections. This is where most of the recurring writing load sits in an AM role, and where a strong first draft saves the most calendar time.
Market research and data gathering. Submarket overviews, sector outlooks, news-to-brief synthesis, macro briefing notes, occupier demand analyses, competitor benchmarking, data table narratives, location assessments, and executive summaries. The specific challenge here is turning scattered inputs — news articles, data points, agent reports — into a coherent, confident analytical view. Good prompts structure that synthesis.
ESG and sustainability tasks. ESG framework explainers, EPC implication assessments, sustainability sections for IC papers, net zero narratives, ESG due diligence question lists, TCFD climate risk summaries, and annual report ESG sections. ESG literacy is now a baseline expectation in institutional real estate, not a specialism. Prompts in this area help you be conversant and credible without needing to be a compliance expert.
What AI cannot do — and why that matters
Two caveats that belong in any honest article about AI and professional work.
First, AI produces a first draft, not a finished document. The deal-specific intelligence — the local market knowledge built from calls and site visits, the nuance in the tenant relationship, the judgement about whether this is actually a good deal — that is yours to add. An AI-assisted analyst who edits intelligently will always outperform one who pastes unchecked output.
Second, AI market data has a knowledge cutoff and should never be treated as a live source. Any yield, rent, vacancy rate, or macro statistic that comes from an AI prompt must be verified against CoStar, MSCI, JLL Research, or equivalent before appearing in any client-facing document. The prompt gives you the structure; the numbers are your responsibility.
A practical starting point
If you want to apply this approach immediately, the most useful single prompt to start with is the IC memo template above. Run it on your current deal or a recent transaction you know well. Compare the output against what you would have written from scratch. The gap — in time saved and in structural quality — is where the value sits.
For a complete set of tested, copy-paste ready templates across all four areas, the playbook below contains 36 prompts specifically written for early-career real estate professionals — each one with a when-to-use context note, the template with placeholders, and a practical tip on getting the most from it.