An Austin Realtor Kept 6 of 28 Claude Workflows, Including a CRM Sweep That Turned Up a Deal
In two months an Austin agent built about 28 Claude tools and kept 6, including a Follow Up Boss lead sweep and a YouTube pipeline. The sweep flagged a buried lead that turned into a deal; slides alone are at least 5x faster.
Who this is about: Real estate agent (William Zhang, Austin, Texas; about $6.5M volume in 2025; founder of Real Estate AI Society)
What it takes
- Cost
- Free planClaude: Free plan. Pro is $20 a month billed monthly ($17 a month billed yearly), and Max starts at $100 a month. Team seats are $25 a month billed monthly.Price checked Oct 24, 2026 on claude.com. Prices change; check before you buy.
- Setup time
- A few weeksOur estimate from the source, for one business.
- Skill needed
- Some tech skillsComfortable installing an app, pasting an API key or running a command.
Explain it simply
A real estate agent who gets most of his business from YouTube set up an AI assistant that he greets each morning. It reports his video stats and new leads, scores his contacts and drafts follow-up texts, plans showing routes from the MLS, and makes scripts, thumbnails, slides and social posts. He still sends the messages and clicks the buttons himself.
What was done
Morning briefing: "good morning" pulls YouTube stats via the Google API, checks competitor channels, sweeps Follow Up Boss and ends with "today's one thing".
CRM sweep: scores every lead A-D using the Keller Williams model he customised, drafts a follow-up in the lead's language and creates a task with a due time.
MLS routing: with browser access it logs into Unlock MLS, finds the homes clients send and preps a 10-15 showing route; teaching it his MLS took about an hour.
YouTube flywheel: topic research, hook, script, thumbnail concept and slides; plus branded image posts, carousels and reels rendered locally.
A sync pushes new Follow Up Boss leads into Real Geeks so they get listing emails from his own site.
A separate Scottsdale realtor, Molly Rodham, saved listing descriptions she liked as a Claude skill that copies their rhythm and tone. She says first drafts need far less rewriting than plain Claude or ChatGPT, and she gives no time-saved figure.
How it went
One lead flagged by the CRM sweep was buried under newer deals; he reached out and it "turned into a deal". Slides are at least 5x faster, and a batch of thumbnails takes about ten minutes. He calls the CRM sweep and morning briefing the highest-ROI starting points.
- Promotional: he founded Real Estate AI Society, promotes its curriculum and rankings, offers the skills as a newsletter sign-up incentive, and says the social engine powers his own Real Agent Clone tool.
- Honest cons: setup takes real time, he still has to act in each platform, and skills built for one MLS or CRM don't transfer.
- The first briefing tracked every task and burned him out; he rebuilt it around one priority a day.
- Molly's Instagram reel is a separate grade D example. She runs coaching for agents with a Claude module and a paid webinar, and the reel ends with "comment ai for my Claude guide."
Prompt
Run my CRM sweep.
1. Pull every lead, note, activity and deal from [CRM] updated in the last [90] days.
2. Score each lead A, B, C or D using this rubric: [paste your lead-scoring model].
3. For every A and B lead with no contact in [7] days, draft a short follow-up in their language that references our last conversation.
4. Create a task for each with a due time today.
Show me the list before creating tasks. Don't send any messages.Setup notes
- Give it your own lead-scoring rubric (he started from the Keller Williams model) rather than letting it invent one.
- Start with the CRM sweep and morning briefing; he found MLS routing hard to reuse because every MLS differs.
Business fit
Who this helps, and how:
Content-driven real estate agents: get a daily briefing and drafted follow-ups for leads you'd otherwise miss.
Local & serviceTeam leads and brokers: build an MLS or CRM skill once and share it with every agent.
Agencies
Related
- Evidence grade BA Kia Sales Manager Gets 2–3 Hours a Day Back With Claude and Browser AutomationStrong evidence, using Claude, Playwright and Supabase
- Evidence grade DA Loan Officer Built a Self-Updating Mortgage CRM With Grok Bot That Feeds Itself From the LOSLight evidence, using Grok Bot, Zapier and Arive (loan origination system)
- Evidence grade CA 70-Year-Old Manufacturer Cut RFQ Turnaround Nearly in Half With Claude CodeVendor or secondhand, using Claude Code
- Evidence grade DA Claude Agent Takes Instagram DM Orders for a 7-Location Sushi ChainLight evidence, using Claude, Meta API (Instagram DMs) and The restaurant's CRM
- AI for sales prospecting examplesDepartment hub: AI for sales, 34 examples
Similar tools
Other tools for this job, and how they differ. Chosen by our editors, never paid for.
- ChatGPT
OpenAI's assistant with custom GPTs; can hold a lead-scoring rubric and draft follow-ups from a CRM export.
- Gemini
Google's assistant with custom Gems and Gmail, Docs and Sheets links, used here to research and score leads.
See it in use: A Consultant Turns a Company Name Into a Scored Lead in Under 20 Minutes
- Grok Bot
Runs a morning briefing as a routine on its own cloud computer with a browser, asking before it sends anything.
Prompt: Morning readout: what's new, what matters, what to do
“Strong evidence” (grade B) means: The person who did the work describes the setup and results in detail. How we grade. Spotted a mistake? Suggest a correction.