Prompt Expander for Real Estate Brokers
Type a short request the way you'd ask a colleague. Get a structured, paste-ready prompt for ChatGPT, Claude, or Perplexity that produces real broker-grade output.
Your expanded prompt will appear here.
Pick a task type, describe what you want, and click Expand.
Powered by Groq running Llama 3.3 70B. Each expansion is a real LLM call. The prompt it produces is engineered to instruct ChatGPT, Claude, or any other tool to return a structured table, source URLs, and a 12-month freshness window.
Station CRM is the CRM for commercial real estate brokers who actually use AI. Market intelligence, deal pipeline, and a built-in chief of staff that knows your book. See how it works →
What is a prompt expander for real estate?
A prompt expander takes a short, conversational request from a real estate broker and rewrites it as a complete prompt for ChatGPT, Claude, or Perplexity. A broker types something like "find me F&B tenants likely expanding in SoHo" and gets back a structured prompt that establishes the broker's role, names the corridor and asset class, requests a ranked list with reasoning, and constrains the output to avoid generic AI fluff. Station CRM's prompt expander is built for commercial real estate workflows: prospecting, market research, outreach, due diligence, OM writing, and lease review. It runs on Groq for sub-five-second response times and is free with no signup. Most brokers prompt AI tools the same way they Google search, with three or four words. The output reflects that. A purpose-built prompt expander closes the gap between how brokers think and how AI tools want to be asked.
Brokers who get the most out of ChatGPT and Claude already understand this. They prompt with role, context, asset class, output format, and constraints. The expander does that work automatically so brokers can stay in their own vocabulary and still get high quality output.
How brokers actually use it
Broker types: "find tenants likely expanding in SoHo this year"
Expanded prompt: "You are a NYC retail leasing broker covering SoHo. List 8 to 12 retail brands or restaurant groups that have credible signals of expansion in NYC over the next 12 months and would plausibly take 1,500 to 4,000 SF in SoHo. For each, include: store count and footprint trend over the last 24 months, recent funding or news indicating expansion, who handles real estate decisions, and one specific reason SoHo fits their thesis. Format as a ranked list. Cite sources. Skip any brand that has only generic press, no concrete expansion signal."
The expanded version produces a useful answer the first time. The original version produces a generic list that requires three or four follow-up prompts to become useful.
What the seven task presets do
Prospecting generates prompts that ask the AI to surface specific signals (closings, expansion announcements, ownership changes, 1031 windows) and rank the output with reasoning, instead of returning a generic list of names.
Market research structures the prompt to request sourced figures, time windows, geographic boundaries, and what's changing. Output comes back as a brief with sections for rent ranges, recent comps, tenant categories, and risk factors.
Outreach drafts cold and warm broker emails with the right hook, call to action, and word constraint. No generic openers, no "I hope this email finds you well."
Due diligence turns a vague "what should I check" into a checklist for ownership history, zoning, lease comps, tenant credit, neighborhood trajectory, and red flags.
OM and marketing copy structures property description prose for the right audience: institutional buyer, owner-user, tenant rep, depending on what the broker says.
Lease and LOI review asks the AI to identify economic terms, escalations, obligations, options, and unusual provisions, then output a structured summary with flagged issues at the end.
General is the catch-all for anything CRE that doesn't fit the other six.
Why generic AI tools fall short for CRE
Generic prompt tools assume you're writing marketing copy or blog posts. The conventions of a good real estate prompt are different. A broker prompt has to specify market and submarket, asset class, deal size, who the broker represents (tenant rep vs. landlord rep), and what output format the answer should take. Without those, ChatGPT will produce a confident, generic answer that mixes asset classes, ignores the local market, and skips the constraints the broker actually cares about.
This expander encodes those CRE-specific conventions. It's not a wrapper around ChatGPT, it's a prompt engineering layer that makes ChatGPT and Claude useful for the work brokers actually do.