Real Estate Listing Description Automation MLS copy, posts and emails from the listing data, for review
A new listing becomes an MLS description, social posts and a buyer email in ten minutes — fair-housing checked, in your voice.
- Installed for you · one-time$249
- Run for you$197/month
- Custom build · this process$2,000–3,500 one-time
- Starts whenNew listing event (CRM / MLS feed / form)
Per agent or small team, up to 30 listings a month. Brokerage-wide deployment with per-agent voices and MLS-specific integrations is priced per seat as a custom build. Custom builds start at $2,000 site-wide; this band is for this process and its integrations.
Real Estate Listing Description Automation is an automation Agentmelt installs in your own tools. The free template is a file you import yourself; the kit ($49) adds the setup guide and prompts. Installed for you costs $249 one-time and is live within two working days of access, with a recorded walkthrough and 30 days of fixes. Run for you costs $197/month, with setup and maintenance included. A custom build of this process is $2,000–3,500 one-time; custom builds start at $2,000 site-wide. No call is needed for any option. Written by Max Zeshut; last revised 2026-09-21.
Reference stack: Claude + n8n. Node names are on each step; the file is under the template.
Founder at Agentmelt
Reviews every build before it goes live; the workflow escalates to a person, never past one.
What is real estate listing description?
Listing description automation takes what an agent already has — the listing data, the photos, the neighbourhood — and drafts every marketing asset the listing needs: the MLS description inside the character limit and the fair-housing rules, the social posts for each platform, the email to the buyer list and a virtual-tour script, in the agent's voice, and puts them in front of the agent for a two-minute review before anything is published. An AI step describes the photos and writes; a rules step checks compliance and length. About three hours saved per listing, and no listing goes out with a description written at midnight.
- Writing a listing description takes an hour and reads like every other listing.
- Fair-housing wording gets checked after the listing is live, if at all.
- The social posts, the email and the tour script never get written because the description took the hour.
- Every agent in the brokerage writes in a different voice, or does not write at all.
- The AI writing tool produced a description that praised a room the photos do not show.
If two of these are yours, this is the process to look at first. Free audit — or read on.
What it looks like once it runs.
Every listing needs the same set of marketing assets and every agent writes them from scratch or not at all. The workflow takes the listing data and photos, drafts the MLS description within character limits and fair-housing rules, produces the social posts, the email to the buyer list and a virtual-tour script, and puts them in front of the agent for a two-minute review. About three hours saved per listing, every listing.
- 11:00 A listing is added with its photos. Each photo is described: quartz island, west-facing deck, finished basement.
- 11:03 The MLS description is drafted within the character limit, leading with the strongest feature; the social captions and buyer email follow.
- 11:05 The fair-housing wordlist and the brokerage disclaimer are checked.
- 11:06 You get one message with every asset; approve, edit or regenerate one from your phone.
- On approval The description goes to the listing, the posts are scheduled, the email is queued.
Agents write listing descriptions at 11pm, reuse the same three adjectives, forget the buyer-list email, and post to social days after the listing goes live. Marketing quality depends on how busy the week was.
Within an hour of a listing entering the CRM or MLS, every asset is drafted, consistent and compliant. The agent reviews and approves on their phone. Listings are marketed on day one, and the brokerage has a single standard across agents.
Trigger, then 8 steps.
Seven of the eight steps run without anyone. Step 06 waits for a person.
New listing event (CRM / MLS feed / form)
Receive the listing
Address, price, beds/baths, square footage, lot, year built, features, HOA, school district, agent notes and photo URLs.
Describe the photos
A vision-capable model describes each photo (kitchen with quartz island, west-facing deck, finished basement) so the copy references what buyers will actually see.
Draft the MLS description
Within the MLS character limit, in the agent's voice, leading with the strongest feature, with a fair-housing compliance check for protected-class language.
Draft the derivatives
Instagram and Facebook captions, a 'just listed' story script, a buyer-list email with the three best photos, and a 60-second virtual-tour narration script.
Compliance and brand check
Fair-housing wordlist, brokerage disclaimer, required MLS phrases, and price/feature consistency against the source data.
Review and approve
The agent gets one message with every asset; approve, edit or regenerate a single asset from the phone.
Publish
MLS description to the CRM/MLS field, social posts scheduled, the email queued in the ESP to the buyer segment matching price band and area.
Track
Every listing's assets, approval time and (from the CRM) showings and days on market, so the brokerage can compare marketed vs unmarketed listings.
Within an hour of a listing entering the CRM or MLS, every asset is drafted, consistent and compliant
Within an hour of a listing entering the CRM or MLS, every asset is drafted, consistent and compliant.
One decision, at step 06. Seven steps ran without you.
Each pattern page explains it in plain words and lists every automation that uses it.
The template is built on n8n (how the n8n AI Agent node works); the same steps run on Make, Zapier or plain code. The tool is chosen to fit your accounts, you own the result, and the template comes in all three formats.
A typical first month — a solo residential agent with four to six new listings a month, on Follow Up Boss, Mailchimp and Buffer.
Not a client report
The listing description gets written at 11pm with the same three adjectives, the buyer-list email never goes out, and the social posts appear three days after the listing is live. Description, captions, email and a tour script take about three hours a listing — when there is time, which is what decides how well each listing is marketed.
The first month- Day 1The first listing enters Follow Up Boss. Within the hour every asset is drafted; the photos are described first — vaulted ceilings, a covered porch facing the hills, a mudroom off the garage — so the copy is about this house, not houses.
- Week 1The agent reviews on their phone. The fair-housing check flags 'perfect for a young family' in the first draft, and a flag cannot be approved without an edit. Two facts are checked against the listing data; nothing is invented.
- Week 2The voice is adjusted from three of the agent's own past descriptions. The buyer-list email goes to matching leads on the day the listing goes live — the one asset that reliably produces a showing from an existing lead.
- Week 3The second and third listings go through with light edits only. Posts publish through Buffer after approval; the brokerage disclaimer and the MLS phrases are inserted, not left to chance.
- Week 4Tracking shows which assets produced showings. The agent stops writing descriptions at 11pm.
| Metric | Before | After a month |
|---|---|---|
| Time per listing on description, captions, email and script | ≈3 h | ≈20 min of review and edits |
| Listings marketed on the day they go live | some | all of them |
| Buyer-list email sent | rarely | every listing, on day one |
| Assets published without the agent's approval | — | none, by design |
Three hours a listing is the blueprint's figure; the day-one effect is the point. Your number is the first-showing date on the listings after the install against the ones before it — the CRM has both.
This is a worked example, not a client's results: the company is a composite of a stated size, and the figures are the ranges this blueprint's own metrics make typical at that size. A real engagement replaces it, with permission, once an install has run a month.
Data it touches
- CRM or MLS listing data (Follow Up Boss, kvCORE, MLS feed)
- Listing photos
- Agent voice sample and brokerage rules
- Buyer list segments (ESP)
- Social scheduler
Guardrails
- Fair-housing language check runs on every asset before review; flagged copy cannot be approved without an edit.
- Facts in the copy are checked against the listing data — no invented square footage or school ratings.
- The agent approves every asset; nothing publishes automatically.
- Brokerage disclaimers and MLS-required phrases are inserted, not left to the model.
Three hours per listing, and the day-one problem
The time saving is real — description, captions, email and a tour script take a working agent about three hours — but the bigger effect is timing. Listings marketed on the day they go live get their first showings sooner, and the buyer-list email that usually never gets sent is the one asset that reliably produces a showing from an existing lead. The workflow makes day-one marketing the default instead of the exception.
Compliance is built in, not reviewed later
Fair-housing rules prohibit language that expresses preference for or against protected classes, and the ways to violate them accidentally are many ('perfect for a young family', 'walking distance to church'). The compliance step checks every asset against a maintained wordlist and the brokerage's rules before the agent sees it, and flags cannot be approved without an edit. That is a stronger control than most brokerages have for human-written copy.
Photos make the copy specific
Listing data says '3 bed, 2 bath, 1,850 sq ft'. The photos say 'vaulted ceilings, a covered porch facing the mountains, a mudroom off the garage'. Describing the photos first and feeding those descriptions to the copy step is the difference between generic and specific — and it is what makes agents accept the drafts with light edits instead of rewriting them.
What it connects to.
| Tool | Role in this workflow |
|---|---|
| n8n | Intake, generation pipeline, approval, publishing |
| Claude | Photo descriptions and all copy |
| Follow Up Boss / kvCORE | Listing source and CRM |
| Mailchimp / Constant Contact | Buyer-list email |
| Buffer / Meta API | Social publishing |
Want this running without building it?
$197/month
We set it up in your tools, run it on our servershost it on our n8n cloud (or deploy it to your own n8n, Make or Zapier), watch it, and fix it when something changes. Setup included, cancel monthly; the automation stays yours if you leave.
$2,000–3,500 one-time
Your systems, your rules, your edge cases. A one-off build delivered into your accounts to keepon Claude and n8n, delivered into your accounts to keep, with documentation and a recorded walkthrough.
Per agent or small team, up to 30 listings a month. Brokerage-wide deployment with per-agent voices and MLS-specific integrations is priced per seat as a custom build.
Also sold one-time, installed in your own accounts, as part of the real-estate agents package and the copywriters package — do it yourself $49 · installed for you $249 · done for you $490.
When this becomes a custom build.
The subscription covers the standard version — $197/month, setup and maintenance included. Three things move it to a custom build ($2,000–3,500 one-time), delivered into your accounts to keep. If one of them is you, say so on the audit and the plan prices both.
Brokerage-wide deployment with a voice per agent and per-seat pricing.
MLS-specific integrations that push the description and photos directly, rather than to a document.
Above 30 listings a month, or several languages.
Running the standard version first is the usual path: the exceptions it logs in the first month are the scope of the custom build, with your own numbers. How custom builds work.
Frequently asked questions.
Does it post to the MLS directly?
It writes the description to your CRM or MLS field where the integration allows; for MLSs without an API the agent copies the approved description. Photos are never uploaded automatically.
Can each agent have their own voice?
Yes — a voice sample per agent is part of setup. For brokerages, per-agent voices and a shared compliance layer are configured together.
What about rentals or commercial listings?
The same workflow with a different asset set and rules; commercial listings need a custom build for the different data fields and marketing formats.
Can AI write MLS listing descriptions that comply with fair-housing rules?
It can draft within them when the rules are part of the workflow: the description is written from the property's facts and the photos, then checked by a rules step against the fair-housing word list and the MLS character limit before the agent sees it. Words about who should live there never appear; words about the property do. The agent's review is the last check, and it takes two minutes.
Does listing description automation post to the MLS directly?
The standard workflow produces the description and the assets in a document and a review board; the agent pastes the description into the MLS, because most MLSs gate direct posting behind vendor programmes. Brokerages with an MLS integration can have the workflow push directly as a custom build.
Can each agent have their own voice?
Yes — a voice guide per agent: phrases, banned words, three descriptions they were proud of. The workflow reads the guide on every draft, so a brokerage-wide install produces descriptions that sound like each agent rather than like the brokerage. Per-agent voices across a brokerage are priced per seat.
What this blueprint is built on.
Vendor documentation the steps rely on, and the rules the guardrails are written against.
- U.S. Department of Justice — The Fair Housing Act www.justice.gov/crt/fair-housing-act-1
- Follow Up Boss — API documentation docs.followupboss.com/
- Meta — Graph API developers.facebook.com/docs/graph-api/