Analyzing and negotiating
3 AI Prompts I Use to Price, Find, and Close Real Estate Deals
The three AI prompts I actually use: reverse-engineering price from DSCR, finding submarkets I'd never heard of, and catching errors my lawyer missed.
I've bought hundreds and hundreds of multifamily rentals plus hospitality and RV parks, and I want to share what I think is the most usable skill in real estate right now. AI is obviously changing a lot of the landscape. You can get super deep into this with custom agents and all the other ways people are competing. But if you're not doing the simple things I'm about to show you when you negotiate your deals, you are missing out on most of what AI actually has to offer today.
These are three prompts. They take minutes. If you follow them when you're analyzing and finding deals, you're going to close more real estate, and these skills are only going to get more important as time goes on. If you're not doing this in 2026, you're already behind. The tools will change, but checking their output against real deal information remains important.
Prompt 1: Reverse-Engineer the Price From DSCR
The most important piece for me is the ability to instantly re-evaluate the pricing of a deal without touching a calculator.
If you're a bank, the most important metric is debt service coverage ratio: DSCR. Banks like to see a debt service coverage ratio of 1.25 or above. What I love about real estate is that everything is named roughly what it actually does. Debt service coverage ratio is the amount you have to pay your debt, and then continue to pay your bills above and beyond that.
If my total debt, principal plus interest, costs $10,000 a month, and my property brings in $15,000 a month of net operating income after expenses, I have a 1.5 DSCR. I can pay myself 50% of what I'm paying in mortgage. That's a very stable spot. Banks want to see you have enough to cover your obligations and pay your debt, with 25% or more of those debt payments available in surplus to pay yourself and your investors.
Here's where AI comes in. You can put in your net operating income directly, or (if the broker gave you an offering memorandum) you can download the OM, drop it in ChatGPT or any other AI, and say "summarize this for me." It will spit out the net operating income. As long as those are actual numbers, you can now underwrite the deal with one question:
"At what price is the debt service coverage ratio 1.25 or higher?"
It will instantly run all the math and tell you the price a bank will likely value the property at and lend on. It'll ask you a couple of questions first: at what loan to value, and what interest rate are we assuming. You can literally answer "75% loan to value and at market interest," and it will do the research and spit out a number.
This does not replace knowing how to do the math. But the ability to take your assumptions and edit them instantly with no thought whatsoever is amazing. For myself, I like to negotiate around a 1.3 DSCR. When I believe I have the true net operating income, I plug it in and ask: based on this, and these debt assumptions, at what price?
Using the Same Prompt for Creative Finance
You can flip this for creative finance, and this is where it gets fun.
Say your seller is married to a million-dollar price. They will not move off it. Instead of fighting about price, ask:
"At what interest rate of interest-only debt would I need to negotiate to get a debt service coverage ratio of 1.3 or higher?"
If you think about it, it makes sense. We're taking our debt service. If the debt service is cheaper, it increases our debt service coverage ratio. If the price is lower, we're paying less debt, the payments are lower, and again our debt service coverage ratio goes up. So you can solve for either the interest rate or the price. Either way, you're playing your sellers off the bank.
That makes the negotiation call very easy. You call the broker or owner and say: I analyzed this the way a bank is going to look at it, they're probably only going to lend here, and I want to come in at 75% loan to value, so this is my price.
It's really hard to argue with that. If leverage is the big constraint for you ("I don't want to lever less than this") then that is your maximum price, and that's realistically what a bank is going to look at and how every other investor is going to analyze the deal. It's a fantastic way to negotiate price, and AI does it for you.
Prompt 2: Find the Submarkets You've Never Heard Of
Zooming out further: choosing markets. There are hundreds and thousands of markets all over the country that work fantastic. The best market is typically your backyard. But sometimes you're in a really rough market, or a very competitive one, and you need to expand your horizons.
What I like to do when I choose an area is pick five or six submarkets inside it. When I made the decision to move to Dallas, Texas, I drew a two-and-a-half-hour radius with AI around Dallas and asked:
"Within a two-and-a-half-hour drive of Dallas, what are major high-growth cities outside of the main DFW metroplex?"
It spit out Abilene, Stephenville, Granbury, Weatherford, Tyler, Sherman and Denison, and Wichita Falls. I went through all of those markets and asked AI about their population growth. It pulls from sites I love, like World Population Review, among others. If you're using ChatGPT there's a link at the bottom you can click to see where it got its sources, so if you want to dive deep you can actually fact check the research. But essentially you get demographic information instantly.
What am I looking for? Job stability and population growth. Put all your parameters in, let it recommend markets, then hop into LoopNet or Crexi and start looking through listings there.
Here's what it did for me. I did not know all of those markets when I was looking at that area. If I had been on a map just going through DFW, I might have missed some of them or never heard of them at all. Being from Washington state, I had never heard of a Wichita Falls: sorry if you're from Wichita Falls, I just didn't know that was a place. Before researching Dallas and the surrounding markets, I also didn't know what a Stephenville was.
Today I own 186 units in Stephenville. It came up as a potential high-growth market because of the job diversity, the population growth, and the fact that it's the fastest-growing D1 college in the United States. All of those combined make for a very, very strong market that I have made a lot of money in. Leasing is going fantastic, my employees there are fantastic, my tenants are happy, my buildings are full, and rent has gone up in a stable and sustainable way where I can keep excellent tenants and continue to increase the income of my properties. All because of a simple search.
Stephenville is the second-largest market I own in. The largest? Abilene, Texas. That was a little farther than I thought I was going to invest: it's at the very peak of that two-and-a-half-hour radius. It came onto my radar through the same search, and 225 units in that market two years later came from these exact same skills.
So spell out your buy box, your perfect parameters, and look around the markets you're considering. See if you're missing submarkets or cities in easy driving distance from where you want to invest that you didn't even know existed.
Prompt 3: Have AI Stress-Test Your Legal Documents
This last one is a no-brainer. AI has caught so many errors in documents created by legal professionals.
To be clear: ChatGPT does not replace having a lawyer. Neither does Claude, nor any of the other AIs. You should still have a great real estate lawyer. That said, they will often miss things.
I use this prompt every time we do an operating agreement or a purchase and sale agreement. I have it search for weaknesses:
"Read this document and look for potential problems or contradictory language in this document."
Add anything else you're specifically worried about. It will read the entire document and recommend how to fix the weaknesses it finds.
I actually caught contradictory tax language this way. My lawyer drafted it. I reviewed it. It was about an 84-page operating agreement, and we both missed the same thing: a legacy clause from an old contract that found its way into the template for our current contract. It was not applicable. It shouldn't have been there. Would it have caused legal problems? Maybe. And if it did, I definitely would have lost, because it assigned taxes incorrectly to the wrong investors.
I had a great lawyer. I'm good at reading these. AI found it in seconds and potentially saved me hundreds of thousands of dollars.
The most likely situation is that you never have a conflict and everyone does what you verbally agreed on. But if your contract in writing is incorrect and everyone signs it and your lawyer gives you his blessing, too many people assume it's going to be perfect. This is something AI can check instantly.
Key Takeaways
- Ask AI at what price a deal hits a 1.25 DSCR (I negotiate at 1.3) to get the number a bank will actually lend on, and use it as your maximum price in the negotiation.
- For creative finance, hold the seller's price constant and solve for the interest-only rate that produces a 1.3 DSCR instead.
- Draw a driving radius around your target metro and ask AI for high-growth cities outside the main metroplex, then screen them on job stability and population growth before going to LoopNet or Crexi.
- That exact search put Stephenville (186 units) and Abilene (225 units) on my radar: two markets I'd never heard of.
- Run every operating agreement and purchase and sale agreement through AI looking for contradictory language. It caught an 84-page error my lawyer and I both missed.
- None of this replaces knowing the math or having a lawyer. It replaces the hours between having the question and having the answer.
You can use this workflow to summarize a Crexi offering memorandum, test a price, draft an offer, and identify questions in an operating agreement. AI can speed up the first pass. The numbers, source documents, and final agreement still need your review and the appropriate professional input.
You can write the letter of intent there too. I go into ChatGPT (I have a custom one called Osgood GPT, but it's powered by ChatGPT) and say: I'm writing a letter of intent for this property at this address that we just analyzed, please fill it out with everything I put above in this chat. It writes it and asks questions for anything that's missing.
So apply all of it. Negotiate your pricing or your terms, or at least fact check them. Get the legal review. Find markets you didn't even know existed, almost instantly. If you apply this today, you'll buy more real estate tomorrow.
Watch the full video above for the walkthrough of each prompt as I run it. If you want to go deeper, you can learn about my mentorship at mentorship overview, and there's a free course on getting started in multifamily at multifamilystrategy.com/get-free-training. And if you want a community of investors doing the same thing (thousands of them in one place with free calculators and educational resources) look for Multifamily Strategy on Skool.
Read the episode transcript
0:00 Hello and welcome back to Multif Family 0:02 Strategy or if it's your first time, 0:04 welcome to Multif Family Strategy. I'm 0:05 Christian, your channel host. I've 0:06 bought hundreds and hundreds of multif 0:08 family rentals and hospitality and RV 0:11 parks. I'm going to be sharing with you 0:12 what I think is the most usable skill. 0:13 Now, obviously, AI is changing a lot of 0:15 the landscape. We can get super deep 0:17 into this and custom AI agents and other 0:20 ways that people are competing, but if 0:21 you are not doing the simple things 0:22 we're going to share in this video right 0:24 now, negotiating your deals, you are 0:26 missing out on so much of what AI has to 0:29 offer. So, if you follow these simple 0:30 tips and these simple prompts in your AI 0:33 when you are analyzing and finding 0:35 deals, you're going to close more real 0:36 estate. And these skills will become 0:38 more and more important as time goes on. 0:40 But if you're not doing this right now, 0:41 2026, you're missing out. If you're 0:44 watching this video in the future, this 0:45 is doubly important. So, here we go. 0:49 Diving in. First of all, the most 0:51 important piece for me is the ability to 0:54 instantly without using any calculators, 0:56 re-evaluate the pricing of a deal. Now, 0:59 if you're a bank, the most important 1:00 metric to you is debt service coverage 1:02 ratio or DSCR. Now, banks like to see a 1:05 debt service coverage ratio of 1.25 or 1:07 above. And if you're watching this right 1:08 now, you're like, "What the heck is 1:09 that?" What I love about real estate is 1:11 everything is named roughly what it's 1:14 actually doing. So, debt service 1:16 coverage ratio, this is the amount that 1:18 you have to pay your debt and then 1:19 continue to pay your bills above and 1:20 beyond that. If my total debt, principal 1:23 plus interest, cost $10,000 a month, and 1:26 my property brings in $15,000 a month of 1:29 net operating income after expenses, I 1:31 have a 1.5 DSCR. I can pay myself 50% of 1:36 what I'm paying in mortgage. And that's 1:37 a very stable spot. Banks want to see 1:40 that you have enough to cover your 1:41 obligations and pay your debt with 25% 1:44 or more of those debt payments available 1:46 in surplus to pay yourself and to pay 1:49 your investors. That is a fantastic 1:51 metric. Now, what you can do in AI is 1:54 you can simply put in your net operating 1:56 income, put in those incomes and 1:57 expenses or if they put them in an 1:59 offering memorandum, you can literally 2:01 go into chat GPT or any other AI, 2:04 download the offering memorandum and 2:06 say, "Hey, summarize this for me." It 2:08 will spit out a net operating company. 2:10 As long as those are actual numbers, you 2:12 may now use that to just underwrite the 2:14 deal as hey, at what price is the debt 2:18 service coverage ratio or DSCR 1.25 or 2:22 higher? And it will instantly run all 2:24 the math and it will tell you this is 2:26 the price that a bank will likely value 2:27 the property at and will likely lend on. 2:29 Now, it will ask you a few questions 2:30 such as at what loan to value and what 2:33 interest rates are we assuming. But you 2:34 can literally say something like, "Hey, 2:35 75% loan to value and at market 2:39 interest." And it will do all the 2:40 research for you and it will spit out a 2:42 number. Now, this is really helpful if 2:44 you are trying to negotiate on price or 2:46 verify your assumptions. You can get a 2:48 first look. This does not replace 2:50 knowing how to do math, but the ability 2:53 to instantly take your assumptions and 2:55 edit them with no thought whatsoever is 2:58 absolutely amazing. Now, for myself, I 3:00 like to negotiate around a 1.3 DSCR. So, 3:03 I'll do exactly that. When I believe I 3:04 have the true net operating income, I'll 3:06 plug that in and say, "Hey, based on 3:08 this and these debt assumptions, at what 3:10 price?" Now, if you want to use this for 3:11 creative finance, you can do the same 3:12 thing. And you can say, hey, if the 3:14 price is this, so say your seller is 3:16 married on a million dollar price, that 3:18 is they're they're married to that 3:19 price. You can say, hey, at what 3:21 interest rate of interestonly debt would 3:23 I need to negotiate to get a debt 3:25 service coverage ratio of 1.3 or higher 3:29 and it will simply do that. Now, if you 3:31 think about it, it makes sense, right? 3:33 Because we're taking our debt service. 3:34 If the debt service is cheaper, it 3:36 increases our debt service coverage 3:38 ratio. If the price is lower, we're 3:41 paying less debt. Again, the payments 3:43 are lower. We increase debt service 3:45 coverage ratio. So, you can do it with 3:47 either interest rate or you can do it 3:49 with the price. But either way, you can 3:52 play your sellers off the bank. Now, you 3:54 can make that phone call if you're 3:55 negotiating price. Very easy. Hey, 3:58 analyze this the way that a bank's going 4:00 to be looking at this. They're probably 4:01 only going to lend here. Now, I want to 4:03 come in at 75% loan to value. So, this 4:06 is my price. I'm a broker or an owner. 4:08 Really hard to argue that because if 4:10 that's the big thing for you, hey, I 4:11 don't want to lever less than this. 4:14 Well, then this is my maximum price. And 4:16 that's realistically what a bank's going 4:17 to look at. And that's realistically how 4:20 every other investor is going to analyze 4:22 the deal. Fantastic way to negotiate 4:24 price. AI can do it for you. Now, 4:26 zooming out even further, choosing 4:28 markets. What do we need to choose a 4:30 great market? Because there's hundreds 4:32 and thousands of markets all over the 4:34 country that work fantastic. The best 4:36 market typically is your backyard, but 4:38 sometimes you're in a really rough 4:39 market or a very very challenging 4:41 competitive market and you need to 4:42 expand your horizons. What I like to do 4:45 is when I choose an area, I want to 4:47 choose five or six subm markets in that 4:49 area. So when I made the decision to 4:51 move to Dallas, Texas, I drew a 2 and a 4:54 half hour radius with AI around Dallas. 4:57 I said, "Hey, within a 2 and a half hour 4:59 drive of Dallas, what are major high 5:01 growth cities that are outside of the 5:03 major like the main DFW metroplex?" It 5:06 spit out cities like Abalene, Texas, 5:09 Stevenville, Texas, Granberry, Texas, 5:11 Weatherford, Texas, Tyler, Texas, 5:14 Sherman and Dennis, Texas, Witchah 5:17 Falls. I went through all of these 5:19 different markets and I asked AI about 5:21 their population growth and it pulls 5:22 from sites that I love like World 5:23 Population Review and a bunch of other 5:25 ones. Now, if you're using Chad GBT, 5:27 there'll actually be a link in the 5:28 bottom where you can actually click it 5:30 and you can look at where it got its 5:31 sources from. So, if you really want to 5:33 dive deep into it, you can actually fact 5:35 check the research, but essentially, 5:37 it's going to give you demographic 5:38 information instantly. What am I looking 5:41 for? I'm looking for job stability and 5:43 population growth. You can put all of 5:45 your parameters in. It will recommend 5:47 markets to you. And now you can hop into 5:49 Loopnet or Crexy and actually start 5:50 looking through listings there. what 5:52 this did for me and what I believe it's 5:54 going to do for you. I did not know all 5:56 of those markets when I was looking at 5:57 that area. If I was on a map just going 5:59 through DFW, I might have missed some of 6:02 those or just never heard of them. I'll 6:04 be completely honest, being from 6:05 Washington State, I've never heard of a 6:07 Witchah Falls. Sorry if you're from 6:08 Witchah Falls. I just didn't know that 6:10 was a place. I also, before researching 6:14 Dallas, Texas, and the surrounding 6:15 markets, I didn't know what a 6:17 Stevenville was. Today, I own 186 units 6:20 there. And it's because I was 6:21 recommended this as a potential high 6:23 growth market because of the job 6:25 diversity, population growth, and the 6:28 fact that it is the fastest growing D1 6:30 college in the United States. All of 6:32 these combined make for a very very very 6:34 strong market that I have made a lot of 6:36 money in. Leasing is going fantastic. My 6:38 employees there are fantastic. My 6:41 tenants are happy. My buildings are 6:42 full. rent has gone up in a stable and 6:45 sustainable way where I can keep 6:47 excellent tenants and continue to 6:49 increase the income of my properties all 6:51 because of a simple search in chat GBT. 6:54 So that worked for me. Do what works for 6:56 you. But you can literally spell out 6:58 your buy box, your perfect parameters 7:00 and look around the markets you're 7:01 looking in. See if you are missing 7:03 submarkets or cities in an easy driving 7:06 distance from where you want to invest 7:08 that you didn't even know existed. That 7:10 happened to me in Stevenville. And 7:11 again, it's the second largest market I 7:13 own in. Largest market I own in, guess 7:15 what it is? Abalene, Texas. Little 7:17 farther than I thought I was going to 7:18 invest, but it's the very peak of that 2 7:20 and a half hour radius. It came on my 7:22 radar 225 units of that market 2 years 7:25 later using these exact same skills. 7:28 Last, and this is just a no-brainer, it 7:30 has caught it being chat has caught so 7:33 many errors in documents that created by 7:35 legal professionals. Now, Chat GBT does 7:38 not replace having a lawyer. Nor does 7:41 Claude AI powered by Anthropic or any of 7:43 the other AIs. You should still have a 7:45 great real estate lawyer. That being 7:47 said, often they will miss things. This 7:49 prompt I use every time we do an 7:50 operating agreement, a purchase and sale 7:52 agreement. I have it search for 7:54 weaknesses. I say, "Hey, read this 7:56 document and look for potential problems 7:59 or contradictory language in this 8:02 document." And if you have anything else 8:04 you're looking for, you prompt it there. 8:06 It will then read the entire document 8:07 and it will recommend how to fix certain 8:09 weaknesses. Now, I have actually found 8:11 in certain documents we had 8:12 contradictory tax language. My lawyer 8:14 drafted it. I reviewed it. It was about 8:16 an 84page 8:18 operating agreement. We both missed the 8:20 same thing. They had a legacy clause for 8:23 an old contract that we did that found 8:26 its way into our template for our 8:28 current contract. It was not applicable. 8:30 It shouldn't have been there. Would it 8:32 have caused legal problems? Maybe. If it 8:34 did, I definitely would have lost. It 8:37 would have assigned taxes incorrectly to 8:39 the wrong investors. I had a great 8:40 lawyer. I'm good at reading them. AI 8:43 found it in seconds and potentially 8:45 saved me hundreds of thousands of 8:47 dollars. Again, the most likely 8:49 situation is you don't have a conflict 8:50 and your investors, you know, we all do 8:52 what we verbally agreed on. However, if 8:54 your contract in writing is incorrect 8:56 and everyone signs it and your lawyer 8:58 gives you your blessing, too many people 9:00 assume, hey, this is going to be 9:01 perfect. This is something that AI can 9:03 do instantly. But in the course of this 9:06 video, you could have found a listing on 9:08 Crexy. You could have taken that OM 9:11 offering memorandum, plugged it into 9:13 GPT, summarized it, found a price that 9:16 works, written the offer, and you could 9:18 have already reviewed your entire OA. It 9:21 takes minutes to do all of these things. 9:24 You can also submit an offer, an LOI. 9:26 Literally, I go into chat GPT and I have 9:29 a custom one called Osgood GPT, but it's 9:31 powered by chat. Plug into chat GPT. I'm 9:34 writing a letter of intent for this 9:35 property at this address that we just 9:37 analyzed. Please fill it out with 9:38 everything that I just put above in this 9:40 chat and it will write it out and it 9:42 will ask you the questions for anything 9:43 that's missing. It's absolutely 9:45 incredible, but you can be saving a 9:48 incredible amount of time with these 9:49 simple props. So, apply all of that. 9:52 Negotiate your pricing or your terms or 9:54 at least fact checkck it on chat GPT or 9:57 your AI of choice. Get the legal review. 10:00 Find markets you didn't even know 10:02 existed almost instantly. If you apply 10:04 this today, you'll buy more real estate 10:05 tomorrow. If you want to join a 10:07 community of investors who are doing the 10:08 same thing, thousands and thousands of 10:09 investors in one place, all looking at 10:11 multif family with some of the same 10:12 tools like free calculators, educational 10:14 resources, and community all in one 10:16 place. Check out skol.com/multifamily 10:20 strategy. link is below or just look for 10:22 the multif family strategy community on 10:24 skolschool.com. 10:27 We'll see you there.
Put these ideas to work.
Get support from Christian and the coaching team with your next multifamily deal. See how the mentorship works or start your application.
Apply Now
