Most agents don’t have a tool problem. They have a tool overload problem. A CRM, a transaction platform, an email inbox, a Slack workspace, a content calendar, a newsletter platform. Every one of them needs logging into, updating, and checking, and every one of them competes with the hours an agent should spend in front of clients.
The shift that’s starting to change that is simple: stop adding apps you have to open, and start building bots that do the work in the background and only come to you for approval.
In a recent RealtyHack walkthrough, two Austin brokers pulled back the curtain on exactly how they do that. Ryan Rodenbeck, CEO-Broker at Spyglass Realty and host of the RealtyHack Podcast, showed the bot stack he uses to run a brokerage. Erika Albert, an Austin broker, showed the bots she built for her own production, including a transaction coordinator she set up the same day.
Both are using the same platform, GrokBot. Here’s what each of them built, what it does, and how to decide where to start.
“GrokBot is probably the best AI tool I have seen for real estate agents. I have tested almost every single AI tool there is, and I think it’s the most efficient, useful, and easy to set up and use.” — Ryan Rodenbeck
Table of Contents
What an AI Bot Stack Looks Like for a CEO-Broker
Ryan’s setup is built around one idea: a CEO-Broker spends most of the day triaging information, not producing it. His bots are designed to read, sort, and draft so he only has to decide.
Inbox Watch: surfacing what matters from hundreds of emails
Ryan called this one his favorite. Inbox Watch goes through his email and his Slack messages and surfaces only the items that need attention. In the demo it flagged an invoice for a pool his family is building, the kind of item that gets buried in a full inbox.
The key detail is what happens next. He can tell the bot to draft a response, so the workflow becomes review and approve rather than read, think, and type. He’s no longer checking email constantly, and he’s not wading through the hundreds of messages that arrive every day.
SEO desk, press release desk, and idea bot
Three of Ryan’s bots handle the marketing back office:
- SEO desk: reports which articles need to go up on the site
- Press release desk: drafted and sent a press release when Spyglass launched its insurance company
- Idea bot: generates content ideas tied to SEO
Each one has a single job. That narrow focus is a recurring theme in both setups, and it’s a big reason the output stays useful.
The AI receptionist and the master bot that routes everything
Ryan also runs an AI receptionist and assistant named Shasta, plus a main GrokBot that acts as the front door. He can drop anything into the main bot and it routes the task to whichever specialist bot should handle it, so he doesn’t have to remember which bot does what.
Cutting a podcast into Reels automatically
One of the most practical demos was content repurposing. Ryan dropped a YouTube link to a podcast recorded the week before, and GrokBot cut it into multiple short clips for Reels, pulled directly from the episode.
That’s the repurposing workflow most agents know they should be doing and rarely have time for. If you’re building a content engine around long-form video, the same principles apply to marketing your real estate business through social media, with the bot handling the cutting instead of you.
For brokers weighing how much of this to build versus hire, it also reframes the question behind running a team versus staying solo. A well-built bot stack covers a lot of what used to require another body in the office.
What an AI Bot Stack Looks Like for a Producing Agent
Ryan’s view is the broker’s chair. Erika’s is the production desk, where the work is tied directly to live deals. Her bots are built around the transaction itself.
Transaction coordinator bot: timelines, key dates, and daily checks
Erika had only set up her TC bot that same day, and it was already doing real work. Once a deal is executed, it builds the timelines she uses for her deals, whether she’s representing a buyer or a seller. It then runs a daily check for tasks and emails tied to the transaction and makes sure everything reaches the right people.
“Basically it’s going to do everything that a transaction coordinator would do. When a deal gets executed, it sets up all of those timelines that I have for my deals.” — Erika Albert
On the demo deal, the bot:
- Read the contract and filled in the key dates
- Identified the lender and the other key parties
- Found the inspection email and added a calendar alert for it
- Emailed the sellers a copy of the executed contract
- Confirmed the title company had the contract
- Marked the sellers as pending in Follow Up Boss
That’s the checklist of small tasks that quietly consumes an hour after every contract. It’s also where deals tend to slip when an agent is busy, which is why it’s a strong first candidate for automation.
Listing bot: showing feedback and weekly seller reports
Erika’s listing bot has one job: manage listings once they go live. It watches her email for showing notifications and reaches out for feedback after every showing. It collects that feedback and delivers it to the seller on a weekly basis.
It also coordinates with her marketing bot to know what went out that week, then drafts a weekly email to each seller covering the full property package sent, the views the listing received, and where it was featured. It keeps an eye on listing expiration dates too.
“Here’s everything you need to know about what we did for your property this week.” — Erika Albert, describing the weekly seller email
Seller communication is one of the most common sources of friction on a listing. Consistent weekly reporting answers the question sellers are always asking, which is what you’re actually doing for them.
Buyer deal matcher connected to Follow Up Boss
Erika’s deal matcher goes into Follow Up Boss, learns who her buyers are and what they want, and recommends properties that fit. From there, it drafts an email for her to approve that explains why each property matches what the buyer asked for.
She noted this one is tailored to her specific use case, but the pattern applies broadly: a bot that reads your CRM and does the matching work. If you’re evaluating how a CRM fits into this kind of setup, RealtyHack’s guide to the real estate CRM is a useful starting point.
The rest of the roster: coach, admin, and STR rules watch
Erika’s stack includes a few more specialists worth noting:
- Coach bot: suggests better ways to use GrokBot in her business
- Inbox triage: alerts her twice a day about items that truly need attention
- Admin bot: handles website maintenance, listing sync, and publishing blog posts
- STR rules watch: tracks short-term rental regulations she needs to stay current on, and could be pointed at any topic an agent needs to monitor
How One Blog Post Becomes a Month of Content
The most interesting part of Erika’s setup is how her bots hand work to each other. It’s less a collection of tools than a small production team.
Marketing bot, SEO/AEO bot, and video editor working together
Her marketing bot builds a comprehensive marketing plan that includes blog posts. Each blog post turns into a YouTube video, and each one also turns into a Reel and a Story. The marketing bot coordinates with her SEO and AEO bot so the content stays relevant to what people are searching for.
Once a post is approved, it goes to her video editor bot. That bot has access to some of her previous photos, sometimes asks her for specific ones, and uses them as B-roll to produce YouTube and Instagram cuts. Those go to the admin bot, which publishes on a schedule. Captions sometimes need to be added afterward, depending on the platform.
Where the human approves
This is the part that makes the system trustworthy. Erika doesn’t hand the keys over. She approves the content calendar, then approves the individual posts. The post arrives as a Google Doc she can edit, and on the day of the demo she made changes to a post about condo prices before it moved on.
“I am just approving these content calendars, and then I’m approving the individual posts.” — Erika Albert
Newsletter bot linked to Mailchimp
Her newsletter bot drafts monthly newsletters. It knows to pull the latest stats and what type of content to include, sends her a few ideas, and once she approves one, it executes through her Mailchimp account. If you’re building out your own email program, the fundamentals of real estate email marketing still apply. The bot just removes the drafting grind.
The Full Bot Roster at a Glance
| Bot | What it does |
|---|---|
| Inbox Watch | Surfaces important emails and Slack messages, drafts replies |
| SEO desk and idea bot | Flags articles to publish and generates content ideas |
| Press release desk | Drafts and sends press releases |
| Shasta (AI receptionist) | Handles reception and assistant tasks |
| Transaction coordinator bot | Builds timelines, tracks key dates, runs daily task checks |
| Listing bot | Collects showing feedback and sends weekly seller reports |
| Marketing bot and video editor | Plans content and turns posts into video, Reels, and Stories |
| Newsletter bot | Drafts monthly newsletters and sends through Mailchimp |
| Deal matcher | Matches buyers in Follow Up Boss to properties and drafts emails |
What to Automate First
Eighteen bots is not where either of them started, and it’s not where you should start either. A few principles come through clearly from the walkthrough.
Start with the repetitive admin that eats your day
Look for the tasks you do the same way every time: sorting email, setting up a transaction, drafting a weekly update. Those are the easiest to hand off because the rules are already in your head. The transaction coordinator bot and inbox triage are natural first picks for most agents.
Keep a human approval step on anything client-facing
Both setups follow the same pattern: the bot drafts, the human approves. Inbox Watch drafts replies. The newsletter bot sends ideas before it executes. The content calendar gets approved before any post gets made. That approval step is what keeps the output on brand and error-free.
Give each bot one job
Notice how narrow the bots are. A listing bot manages listings. A newsletter bot writes newsletters. Specialists stay focused, and a main bot routes between them, which is easier to manage than one assistant trying to do everything.
Use a coach bot to find new use cases
Erika’s coach bot exists to suggest better ways to use the platform in her business. If you’re not sure where automation fits, that’s a smart way to surface ideas instead of guessing.
Pick one bot and give it a week
The takeaway from this walkthrough isn’t that every agent needs a stack of eighteen bots. It’s that the work most agents dread, the inbox, the post-contract checklist, the weekly seller update, the content grind, is now work a well-configured bot can draft while you stay focused on clients.
Here’s the short version to carry forward:
- Pick one repetitive task. Inbox triage or a transaction coordinator are the strongest starting points.
- Keep yourself as the approver. Let the bot draft and let your judgment make the final call.
- Connect it to the tools you already use. Follow Up Boss, Mailchimp, and your calendar are where the real value shows up.
- Add specialists over time. Expand only after the first bot is saving you real hours.
The agents who build these systems early will spend their time on what AI can’t do: building relationships, negotiating, and guiding clients through big decisions. For more on building the systems behind a growing business, explore agent growth resources at RealtyHack.
To listen to more of our podcast episodes, visit The #RealtyHack Podcast Page. The #RealtyHack Podcast is also available to listen to on Spotify, Apple Music, and your other favorite podcast directories.
Sign up for my newsletter and get updated on the next #RealtyHack!