AI Automation

The AI automation stack I use in property management

AI automation workflows for property management

Being an "AI-powered" VA does not mean running everything through one chatbot. It means knowing which tool is good at what, and building them into the daily work so the repetitive stuff stops piling up. Here is the setup I actually use, and what I reach for each one.

Claude, ChatGPT, and Gemini, and what I use each for

I keep all three around because they are good at different things:

Property listings and content open on a laptop
I keep the AI inside the tools I already use, so it is right there in the work.

The four jobs I automate

1. Intake and summarization

Inbound inquiries, applications, and maintenance reports arrive as unstructured text. AI turns them into clean, consistent records, so a maintenance request becomes a dispatch-ready ticket and a leasing inquiry becomes a qualified lead with the key facts pulled out.

2. Drafting communication

Tenant replies, follow-ups, owner updates, listing descriptions, all drafted in seconds and tuned to the right tone. I review and send; AI removes the blank-page time.

3. Reporting and breakdowns

Raw activity, occupancy trends, leasing pipeline, maintenance turnaround, delinquency, gets turned into a clear weekly breakdown an owner can read in a minute. This used to be the most tedious part of the week.

Reviewing a weekly property report on a laptop
Raw activity turned into a clean weekly summary an owner can skim.

4. SOPs and reusable templates

The biggest win here is consistency. I build prompt templates and SOPs so every reply, report, and workflow comes out to the same standard, whether it is me doing it or a teammate. New work starts from something that already works, not a blank page.

It is not about one clever prompt. It is about a setup you can reuse, so the whole operation gets faster in the same way.

Browser extensions: AI where the work is

I use browser extensions so the AI sits inside the tools I already work in, AppFolio, the CRM, email, instead of copy-pasting between tabs. That is what separates using AI now and then from actually building it into the work.

Key takeaways

  • Use the right model for the job: Claude for drafting/SOPs, ChatGPT for fast content, Gemini for research.
  • Automate the four language-heavy jobs: intake, drafting, reporting, and SOPs.
  • Reusable templates beat clever one-off prompts, consistency is the real win.
  • Put AI inside the tools you already use via extensions, not in a separate tab.
Jay Mark Calaor
About the author

Jay Mark Calaor

AI-fluent property management coordinator and VA, certified in Claude and AI fluency. I design AI into leasing, maintenance, and operations across 800+ units. See how I run operations →

Work with me

Put AI to work in your operation

I help property teams design AI into the workflow, not bolt it on. Let's map what to automate first.