AI Systems Instructor • Real Estate Technologist
Quick Answer: Export your comparable sales data, load your Context Card into Claude, and use the HOME Framework to generate a CMA narrative that explains pricing strategy in plain English. Claude's large context window handles complex comp data better than other AI tools.
The numbers in a CMA speak for themselves, but the narrative around those numbers is what wins the listing. Most agents hand sellers a stack of comps and hope the data does the talking. This guide shows you how to use Claude to write the narrative portion of your CMA—the market analysis, the pricing rationale, and the strategy summary—so sellers understand not just what their home is worth, but why your pricing strategy will get them the best result.
Tools Needed
Claude Pro ($20/mo), your Context Card, MLS comparable data
Pull your comparable sales from MLS: 3-5 recently sold properties, 2-3 active listings, and 1-2 pending sales. For each comp, note the address, sale price, price per square foot, days on market, and any condition adjustments. Claude needs structured data to produce structured analysis.
Tip: Include one comp that sold above asking and one that sat on market. The contrast strengthens your pricing narrative.
Paste your Context Card into Claude's project or conversation context. For CMA work, add a supplemental section about your pricing philosophy, your local market expertise, and how you present data to sellers. Claude's 200K context window means you can include extensive background without hitting limits.
Tip: Add your brokerage's CMA template structure to your Context Card so Claude formats the narrative to match your existing presentation.
Hero: You are a market analyst for residential real estate in [your market]. Outcome: Write a 300-word CMA narrative covering market conditions, comp analysis, and recommended list price with rationale. Materials: Paste all comp data plus subject property details. Execute: Professional but accessible tone, use specific numbers, explain adjustments in plain English, end with a clear pricing recommendation.
Tip: Ask Claude to present pricing as a range rather than a single number. Sellers respond better to a strategic range with reasoning than a single take-it-or-leave-it figure.
After the main narrative, prompt Claude for: a one-paragraph market conditions summary, a pricing strategy timeline (what to expect in weeks 1-4), and 3 bullet points for why now is the right time to list. Use the OODA Loop to check each section: does it support your pricing recommendation? Does it address the seller's likely objections?
Tip: Generate a 'Seller FAQ' section with answers to the 5 most common objections: 'My Zestimate says more,' 'My neighbor sold for more,' 'I'm not in a rush,' etc.
[Context Card loaded] Hero: You are a residential real estate market analyst in Williamson County, Tennessee. Outcome: Write a 250-word CMA narrative for a listing presentation. Include market context, comp analysis, and a pricing recommendation with rationale. Materials: Subject: 456 Maple Creek Dr, Franklin TN - 4BR/3BA, 3,200 sq ft, built 2015, updated kitchen 2023 Sold Comps: - 123 Oak Run: $680K, 3,100 sf, $219/sf, 8 DOM, similar condition - 789 Cedar Ln: $705K, 3,400 sf, $207/sf, 22 DOM, original kitchen - 321 Birch Way: $725K, 3,300 sf, $220/sf, 5 DOM, fully updated Active: - 555 Elm St: Listed $699K, 3,150 sf, 14 DOM, similar updates Execute: Professional but conversational. Use specific numbers. Explain price-per-square-foot adjustments. Recommend a strategic list price range. No jargon the seller wouldn't understand.
Williamson County's market is rewarding updated homes in the $650K-$750K range with fast sales—the three most recent comparables in your immediate area averaged just 12 days on market. Your home at 456 Maple Creek Drive has meaningful advantages over the competition. At 3,200 square feet with a 2023 kitchen update, you're positioned between the Cedar Lane sale ($207/sf with an original kitchen) and the Birch Way sale ($220/sf, fully updated). Your kitchen renovation closes that gap. Adjusting for your updated finishes and comparable square footage, the data supports a price range of $690,000-$710,000. Here's why: - At $690K ($216/sf), you're competitively priced to generate multiple showings in the first week and potentially attract competing offers. - At $710K ($222/sf), you're at the top of the comparable range but justified by your kitchen update and overall condition. The active listing on Elm Street at $699K gives us a real-time benchmark. They've been on market 14 days with similar updates—pricing at or slightly below them positions you for faster activity. My recommendation: List at $699,000. It's aggressive enough to drive immediate traffic, defensible with the comp data, and positions you for the strongest negotiating position in the first two weeks.
Use Claude instead of ChatGPT for CMAs. Claude's larger context window handles more comp data, and its writing tends to be more analytical and less formulaic—exactly what you need for data-driven narratives.
Generate a one-page 'Market Snapshot' summary alongside your full CMA. Sellers often share this with their spouse, and a concise summary travels better than a 20-page report.
Ask Claude to anticipate seller objections and draft responses. 'What if the seller says their Zestimate is $50K higher?' gives you ready-made answers for the listing appointment.
Include seasonal context in your prompts. 'Spring market in Nashville typically sees 15% more buyer activity' adds credibility to your timing recommendation.
Pasting raw MLS data without organizing it first
Fix: Structure your comp data clearly: address, price, size, price/sf, DOM, and condition notes. Claude produces better analysis from organized inputs.
Asking for a single price recommendation instead of a strategic range
Fix: Always request a price range with rationale for each end. It gives sellers a sense of control and shows strategic thinking.
Using AI-generated CMA narratives without verifying the numbers
Fix: Claude can misinterpret or miscalculate data. Always cross-check every number, percentage, and price-per-square-foot figure against your source data.
Learn the Frameworks
Stop guessing with AI. Join The Architect workshop to master the frameworks behind every guide on this site.