A prompt is the plain-English instruction you type into an AI tool to tell it what you want. “Summarize the rent and term details in this lease.” “Draft a letter of intent for a 12,000-square-foot industrial space.” “Rewrite this listing to lead with the loading-dock count.” That is a prompt: a request, written the way you would brief a capable new assistant. There is no code and no menu of features. And here is the part most firms miss — the quality of what you get back is decided almost entirely by the quality of what you put in. The tool is the same for everyone; the prompt is not. Which is why, of all the things a commercial real estate firm could learn about AI, writing a good prompt is the single skill that pays off first, fastest, and across every task on the desk. This piece explains what a prompt is and how to write one that works.
Before the mechanics, a word on why this matters more than the tool you pick. Most small firms that try AI and give up do not have a bad tool — they have a bad prompt. They typed four words, got a generic paragraph, and concluded the technology was oversold. The good news in that story is that the fixable part is free. The same subscription, driven by a better instruction, produces work you would actually send. That is the whole argument of the CRE AI training playbook and a large part of why a lean shop can out-operate a much larger competitor: the advantage is a learnable skill, not a bigger budget.
What a Prompt Actually Is
A prompt is the text you send to a large language model — the technology inside ChatGPT, Claude, Gemini, and Microsoft Copilot. If you have not met that term, it is covered plainly in our explainer on what generative AI actually is; the short version is that these tools work with language, and your prompt is how you hand them a language task.
The model reads your prompt and generates a response that continues from it. That is the mechanism. Give it “Draft a cold email to an industrial tenant whose lease expires next year” and it produces a plausible email. Give it “Summarize the key lease terms below” followed by a pasted lease, and it produces a summary. The prompt is both your instruction and your input — the request and, often, the material the request operates on.
The reason this is easy to underrate is that it looks like search. You type into a box; something comes back. But a search engine retrieves a page that already exists, and the quality of the result barely changes if you phrase the query well or badly. A prompt is different: you are not retrieving, you are directing. A vague direction produces vague work. A precise direction produces precise work. The box looks the same. What you do with it is a skill.
Why the Prompt Is the Skill, Not the Tool
Firms shopping for AI ask the wrong first question. They ask which tool, when the answer that changes their results is how to ask. The general-purpose tools are close enough in capability that, for the language work a brokerage does every day, the difference between them is small next to the difference between a good prompt and a poor one on any of them.
Consider what that means for a firm deciding where to spend. You can compare subscriptions for weeks and still get mediocre output, because the bottleneck was never the model — it was the four-word request. Or you can spend an afternoon teaching the team to write a proper prompt and lift the quality of every draft on whatever tool you already have. One is a purchasing decision; the other is a training decision, and it is the one that moves the needle. That is why the smart sequence is training before tooling: the skill compounds across every task, while the tool is close to a commodity.
The compounding is the point. A broker who can write a good prompt writes better LOIs, lease summaries, market paragraphs, prospecting email, and investor updates — all from one underlying ability. Buy a specialized tool and you have improved one task. Teach the prompt and you have improved every task where the work is language, which in commercial real estate is most of them.
The Anatomy of a Good Prompt
A good prompt is not a magic phrase and it is not long for the sake of it. It has four parts, and once you see them you cannot unsee them. Think of it as briefing a sharp assistant who is fast and literal but knows nothing about your specific deal until you tell them.
Role — who you want the AI to be. “You are a commercial real estate analyst.” “Act as a leasing broker writing to a prospective industrial tenant.” Naming the role sets the register and the assumptions. It is the difference between a generic paragraph and one that reads like your profession wrote it.
Context — what it needs to know. The deal facts, the audience, the constraints. “This is a 10-year lease with three months of free rent and a 3% annual escalator.” “The recipient is a logistics company we have not worked with before.” An AI tool knows nothing about your situation that you have not put in the prompt. Missing context is the single most common reason output comes back generic.
Task — the specific thing to produce. Not “help with this lease” but “list the base rent, term, renewal options, CAM structure, and assignment clause in a table.” Precise verbs and a precise deliverable. Vague tasks get vague results, every time.
Format — the shape you want back. “In a five-row table.” “As three short bullet points.” “In two paragraphs, under 150 words, in a professional tone.” Specifying the format saves you the reformatting and gets output you can use as-is.
Role, Context, Task, Format. You do not need all four on every prompt — a quick rewrite needs less than a full market write-up — but when output disappoints, the fix is almost always a missing one of these four. Add the part you left out and try again. That habit of adjusting and re-asking is itself part of the skill; the first prompt is a draft of the request, not a one-shot.
A Weak Prompt and a Strong One, Side by Side
Nothing teaches this faster than seeing the same job done two ways. Take a task every brokerage runs weekly: summarizing a lease.
The weak prompt:
“Summarize this lease.”
Paste the lease under it and you will get a summary — a reasonable one, even. But it decides for you what matters, wanders across clauses you do not care about, skips the ones you do, and hands you prose you have to re-read to find the rent. You spend as long fixing it as you saved.
The strong prompt:
“You are a commercial real estate analyst. Summarize the lease below for an acquisitions review. Return a table with these rows: tenant, premises size, base rent and rent schedule, lease term and commencement, renewal options, CAM/operating-expense structure, assignment and subletting rights, and default provisions. Flag anything unusual in a short note under the table. If a term is not stated in the lease, write ‘not specified’ rather than guessing.”
Same tool, same lease, ninety seconds more thought. The second version pins the role, states the audience and purpose, names exactly the eight terms you need, fixes the format as a table, and — the underrated line — tells the tool what to do when a fact is missing so it does not invent one. What comes back is something you can drop into a deal file after a verification pass, not a paragraph you have to rebuild.
The gap between those two outputs is not a better tool. It is the same tool, driven by someone who has learned the skill. Run the exercise once on your own next lease and the lesson sticks better than any explanation.
Where a Better Prompt Will Not Save You
Honesty about the limits is what separates useful training from hype. A stronger prompt improves how the tool handles language. It does not turn the tool into something it is not, and there are three places where no prompt rescues you.
Arithmetic and exact figures. These tools predict plausible text, not verified calculation. Ask one to compute effective rent across a term with free months and escalators and it may return a confident, wrong number regardless of how well you phrase the request. Keep the math in your spreadsheet and use the prompt for the words around it.
Verified fact. A prompt cannot make the tool stop occasionally stating something false with full confidence — an invented figure, a misremembered statute, a fabricated citation. That behavior, called hallucination, is a property of how the technology works, not a phrasing problem. The rule holds no matter how good your prompt is: the AI writes the draft, a human owns the facts.
Judgment. Whether the deal is good, whether to push on a term, what a client actually needs — a prompt cannot outsource that, and you would not want it to. The skill makes the routine drafting faster so your judgment goes where it belongs.
Knowing where the prompt stops working is part of knowing how to use it. A firm that trains the skill also trains the limits, which is why fluency is a discipline and not a trick.
The Prompt Is Also Where Your Data Goes
There is one more reason the prompt deserves your attention, and it is the one a careful principal raises first. The prompt is not only your instruction — it is the doorway your firm’s information walks through to reach the tool. When you paste a lease, an offering memorandum, or a rent roll into the box, that content is now in the prompt, and the question of what may go there is a real one.
This is why learning to prompt well and learning what to put in a prompt are the same lesson taught from two sides. A fluent user does not just write a sharp instruction; they classify the material first — public and internal facts go in freely, NDA-bound or personal data gets redacted or kept out, and the whole team works from one paid business tier where the vendor states it does not train on your input by default. Getting the instruction right and getting the input handling right are two halves of one skill.
So prompt fluency is not only about output quality. It is the moment where good habits about confidential deal data either exist or do not — train the skill properly and you are training both at once.
How a Small Firm Builds the Skill
The encouraging part is how small the lift is. This is not a certification or a new hire. It is a repeatable structure — Role, Context, Task, Format — applied to the two or three documents your firm produces most, practiced until it is automatic. A team of ten can reach genuine fluency in an afternoon of guided practice plus a couple of weeks of doing real work with it.
The mistake to avoid is the one most firms make: buy the subscription, hand it to the team, and hope. A powerful tool given to people who have not learned to drive it produces a few cautious experiments and quiet disuse — the exact pattern behind “we tried AI and it did not do much.” What a firm looks like once the whole team, not one enthusiast, has the skill is described in the anatomy of an AI-fluent brokerage: drafts move faster, the quality is consistent, and nobody is afraid of the blank box.
The shape is simple: teach one structure, practice it on real leases, LOIs, and market write-ups, pair it with a simple rule about what data goes in, then let the habit spread by use. The ordered version of that program is the CRE AI training playbook, which starts here — with the skill — rather than with a shopping list.
Where to Start
You do not need to become technical to write a good prompt, and you do not need to guess which of your firm’s tasks are the right ones to practice on. A free AI-readiness assessment is a short working session that looks at your actual mix of brokerage, management, and acquisitions work, finds the handful of document-heavy tasks where a well-written prompt pays off fastest, and shows your team the four-part structure on your own documents. Book a free AI-readiness assessment and you will leave with the one skill in hand and a plan to spread it — matched to your deals, not a generic tutorial you work through alone.
Frequently Asked Questions
What is a prompt in AI, in simple terms?
A prompt is the plain-English instruction you type into an AI tool like ChatGPT, Claude, Gemini, or Microsoft Copilot to tell it what you want. It can be a request (“draft a cold email to this tenant”), a task plus material (“summarize the lease below”), or both. There is no special syntax — you write the way you would brief a capable assistant. The tool reads your prompt and generates a response that continues from it, so the clearer and more specific your prompt, the more usable the result.
Why is writing a prompt called a skill?
Because the same tool produces very different results depending on how you ask. A vague, four-word request returns generic output; a precise request that names the role, context, task, and format returns work you can nearly use as-is. Unlike a search box, where phrasing barely changes the result, a prompt directs the tool rather than retrieving a fixed answer — so the phrasing is most of the outcome. That gap between a weak prompt and a strong one, on identical software, is why prompting is a learnable skill worth training.
What makes a good prompt?
Four parts: Role (who you want the AI to be — “a commercial real estate analyst”), Context (the deal facts and audience it needs to know), Task (the specific deliverable — “list these eight lease terms in a table”), and Format (the shape you want back). You rarely need all four on a quick job, but when output disappoints, the fix is almost always a missing one of these. Add it and re-ask. Adjusting and trying again is part of the skill, not a sign you did it wrong.
Why is prompting the first AI skill a real estate firm should learn?
Because it compounds across every language task and costs nothing to improve. A broker who can write a good prompt produces better LOIs, lease summaries, market write-ups, prospecting emails, and investor updates — all from one underlying ability, on whatever tool the firm already has. Buying a specialized tool improves one task; teaching the prompt improves every task where the work is language, which in commercial real estate is most of them. It is a training decision that moves results more than the tool-selection decision most firms fixate on.
Can a better prompt fix every AI problem?
No, and knowing the limits is part of the skill. A better prompt improves how the tool handles language. It does not make the tool reliable at arithmetic — keep effective-rent math in your spreadsheet. It does not stop the tool from occasionally stating something false with confidence, so a human still verifies every figure, legal term, and market claim. And it cannot replace judgment on whether a deal is good or a term is worth pushing. Prompting makes the routine drafting faster; it does not outsource the thinking.
How is a prompt different from a Google search?
A search engine retrieves an existing page, and phrasing the query well or badly barely changes what you get. A prompt directs a tool to generate something new — a draft, a summary, a rewrite — so the specificity of your instruction is most of the result. You are not looking something up; you are briefing a fast, literal assistant that knows only what you put in the prompt. That is why “summarize this lease” and a detailed, structured request produce such different output from the same tool.
Do I need technical skills or coding to write prompts?
No. A prompt is written in ordinary English, and the skill is closer to briefing a new hire than to programming — say who you want the tool to be, give it the facts, state the specific task, and specify the format. That low barrier is precisely why a small firm with no IT department can adopt these tools. The investment that pays off is a couple of hours teaching the team a simple structure and practicing it on real documents, not hiring anyone technical.
Is it safe to put lease or deal information into a prompt?
It can be, with the right handling, because the prompt is where your data enters the tool. Two things matter: whether the vendor trains on your input (the business and enterprise tiers of the major AI products state they do not, by default) and whether your NDA permits sharing the material at all. The safe practice is to put the team on one paid business tier and classify each document before pasting — public and internal facts go in freely, NDA-bound or personal data is redacted or kept out. Learning to prompt well and learning what to put in a prompt are the same lesson.
How long does it take a small firm to learn to prompt well?
Faster than most expect. The core structure — Role, Context, Task, Format — takes an afternoon of guided practice to grasp, and a couple of weeks of doing real work with it to make automatic across a team of ten. The barrier is not difficulty; it is the common mistake of buying a subscription and handing it over without any training, which produces a few experiments and then disuse. A short, structured push on your actual documents turns curiosity into a skill the whole team shares.
Arthur Wandzel
