Using AI to Be More Efficient: What are you doing?

Question

Hey All,

I was talking with my business partner (we own 8 doors together and I own another 9 on my own). All are out of state. Some of the challenges we are seeing is in different markets there can be different property managers, different issues going on, etc. I'm curious on how you might automated things to avoid missteps or things going through the cracks.

For example:

You get an owner statement that you load up and AI can call things out like "it looks like you had another HVAC issue based on a past Owner Statement" or "The rent came in but not the reimbursement for utilities based on the leases stored."

I'm curious what others are doing as I build this out for myself.

Thanks!

Ashish Acharya · Answer

Rick, with 17 doors across different markets, I think AI becomes most useful as a second set of eyes on the portfolio, not as a replacement for the PM or bookkeeper.

A few things I’d automate are exactly what you mentioned: compare owner statements month over month, flag duplicate or unusual vendor charges, catch missing utility reimbursements, identify lease expirations and rent increases coming up, and compare maintenance spending by property. I’d also have it flag anything coded differently from prior months so you can investigate before year-end.

From the tax side, AI can help surface issues, but I would not let it make the final call on whether something is a repair or capital improvement. If you replace an HVAC system, roof, flooring, or another major component, tracking that separately can also help your CPA evaluate a partial disposition instead of continuing to depreciate something that is no longer there.

And with 17 rental doors, I’d review cost segregation across the portfolio if you haven’t already. It may accelerate depreciation on qualifying components, but the important question is whether those additional losses are actually usable under your passive-loss situation.

The biggest opportunity to me is using AI to tell you what needs human attention, rather than trying to automate every decision.

Happy to connect!

Garret Rumbea · Answer

Hey Rick, also based out of SoCal and investing out of state in larger multifamily. I just built a dashboard to help me find discrepancies in regards to leasing, occupancy, and marketing that gives me good visibility into what our PM is saying and tracking on their end vs what actually is happening. Anyways, let's connect some time since we're close by!

Jim Johnson · Answer

I’d use AI less for making property management decisions and more for catching things I might otherwise miss across multiple properties. A few areas where I think it can be genuinely useful:

• Compare owner statements month to month and flag unusual expenses
• Spot repeat repairs on the same system or property
• Check charges against lease terms, like utilities or other reimbursements
• Track maintenance trends and identify when repairing something repeatedly may no longer make sense
• Flag missing rent, deposits, or expected income
• Summarize what changed across the portfolio so I’m reviewing exceptions instead of every line item

The part I’d be careful with is giving AI too much autonomy. I’d want it saying, “You should look at this,” rather than making the call itself. With different PMs and markets involved, that kind of second set of eyes could be really useful.

Russell Cabus · Answer

Hi @Rick Albert! AI and automation can guide property management decisions by flagging unusual expenses, repeat repairs, missing payments, and unresolved maintenance issues. The key is keeping your data in one place and verifying each alert before taking action.

I hope this is helpful!

Giuseppe Cavucci · Answer

This is the exact blind spot with out of state ownership, you get the symptom (another HVAC issue) with no context on whether it was actually predictable or something worth pushing back on with the PM.

What's helped me think about this: every major system in a property has a known expected lifespan (roof, HVAC, water heater, appliances). If you track install dates against those numbers, an "HVAC issue" on a 4 year old unit reads very differently than the same note on an 18 year old unit. One is bad luck or a bad PM, the other is just physics catching up. Without that baseline, every owner statement note looks equally alarming or equally dismissible, and you can't tell which.

Doesn't solve the leasing/occupancy side Garret mentioned, but for the maintenance-specific noise, having that age-vs-lifespan reference point is what lets you ask the right follow-up question instead of just trusting whatever the PM tells you.

Eduardo Cavasotti · Answer

Rick, with 17 out of state doors, I would separate alerts from authority. Let the system flag the mismatch, repeat HVAC note, missing utility reimbursement, or statement change, but keep the owner as the one approving what touches the books. The hard part is not the prompt. It is normalizing every PM statement, lease, repair note, and deposit into the same shape so the alert is tied to a real record.

Shiloh Lundahl · Answer

Ways that I am using AI to be more efficient is in design and in advertising. I can have ChatGPT paint the house several different colors to determine which color I like the best. I can also take pictures of the house during the repairs and I can remove anything that is kotnsupposed to be there in the pictures and I can add staging and list it while it's still getting finished so it saves me on time and money from not needing to get professional pictures.

Rick Albert · Answer

Yea my thought would be call outs, not so much decision. AI lacks that (for now) given most people don't memorialize decisions and therefore wouldn't be able to give the best advice.

I'm also thinking since Claude or Gemini can input into spreadsheets, maybe it can also handle some high level accounting too and then give an end of year report for taxes. Saves everyone time and money.

Rick Albert · Answer

Absolutely!

Using AI to Be More Efficient: What are you doing?

Using AI to Be More Efficient: What are you doing?

Most Popular Reply

11 Replies