How to use Grok Bot to manage a plumbing company's content planner
Phoenix plumbing owner Jon ONeill hired a Grok Bot named Mary as marketing manager to own the website content planner, review posts and manage the marketer. On day one it caught HTML errors.
Suggested prompt
Suggested prompt written by Agents In Use, not the source author. Replace bracketed inputs and review tool permissions first.
You own our website content planner in [sheet]. For each weekday article, check the keyword, title, draft and live URL, review the page for errors and send me a short list of fixes for the marketer.
Reach people exploring AI agents
See placements →Evidence at a glance
Sourced account- Source poster / publisher
- Jon ONeill (@HouseHackerJon)
- Original account
- Read the source ↗
- Published
- Source link checked
- Reported result
- Jon says Mary caught HTML errors on day one and that replacing the agency will save $2,000 a month. He gives no traffic figures.
- Human involvement
- Jon manages Mary and sets direction. The marketer and agency still do the production work described.
A linked account supports attribution. It does not mean we independently reproduced its result.
Report an issue with this case
Open correction emailOpens your email app with the case link included. Nothing is sent automatically.
How the agent did it
- He created a Grok Bot named Mary as marketing manager.
- Mary took over the website content planner of dates, keywords, titles, drafts and URLs.
- She reviews each post and manages the part-time marketer.
@HouseHackerJon, who owns a plumbing company in Phoenix, wrote a thread about using Grok Bot to bring in customers and, in his words, save $2,000 a month.
He explains that a part-time marketer ran a content planner for the company website, with a date, keyword, title, draft and live URL for weekday articles. He was supposed to review all of it and kept falling behind.
So he "hired" a Grok Bot named Mary as marketing manager. Mary owns the content planner, reviews the posts and catches mistakes he might miss late at night. He manages Mary, and Mary manages the marketer. On day one, he says, Mary caught HTML errors in the code for some of the site's graphs, and the agency apologised for missing them. In the thread he describes the end state as the bot writing the articles and Grok building the pages.
He opens with a customer who called at 7:30am about tree roots in her line and said she found the company on Google through its website and reviews. He does not claim that call came from Mary's work.
In replies he says the setup fits on a $200 a month plan and that the saving comes from replacing the marketing agency, not his office manager. He gives no traffic or ranking figures.
Who it's for: Trades businesses that publish website content but cannot keep up with reviewing it.
Prompt
Suggested prompt (written by Agents In Use, not the author's)
Source: X · Jon ONeill (@HouseHackerJon)
Frequently asked questions
What does the Mary bot do?
Jon says Mary owns the content planner, reviews the posts and catches mistakes, and manages the part-time marketer.
Where does the $2,000 a month saving come from?
In replies he says it comes from replacing the marketing agency. His office manager stays full time.
More details
Remaining steps
- On day one she caught HTML errors in the site's graph code.
About this account
Mary reviews content and directs the marketer without Jon checking each article himself.
Platform: X. Attribution: https://x.com/HouseHackerJon/status/2088305236003926468. Read the original post text, author and publication date through the X API on 2026-10-19. Found via the grokbot.dev use-cases page; every fact here comes from the original post, not grokbot.dev's summary. Prompt: Suggested (written by Agents In Use, not the author's). Details come from the author's own thread replies, read through the X API.
Credit: Jon ONeill (@HouseHackerJon) — X
Source published
Added to Agents In Use
Source checked
Tools and topics
Grok Bot
Content marketing · SEO · Trades
Editorial completeness: 100/100. Measures documented fields, not agent performance or verified outcomes.