Anna Walsh
I want to welcome you to our webinar today, The Future of Real Estate Intelligence: How to Get Faster and Better Underwriting While Using Your Own Model. I'm so excited, and I know our speakers are so excited.
I'm thrilled to introduce Erika Banach and Fred Canney. They are here today to talk to you about what the future looks like in this industry. My name is Anna and I'm with Archer and I'm going to be your host today. Now, I'm going to turn it over and start with Erika. Can you tell me a little bit about yourself and tell the audience too?
Erika Banach
Yes, thank you so much for having me. I'm excited to be here today. And thanks everyone for tuning in. A little bit about me: I started my career in real estate working for Extell Development Company, which is one of the most active real estate developers in New York. I got to wear a lot of different hats, which allowed me to really understand real estate and finance. I am a graduate from the University of Scranton, and I received my MBA from the Zicklin School of Business at Baruch College.
A little over three years ago, I joined MMCC, the finance division of Marcus and Millichap, and I was brought on to help drive the capital markets growth strategy. Just to take a little step back, for anyone that doesn't know who Marcus and Millichap is, we are a leading commercial real estate brokerage that specializes in investment sales, financing research and other advisory services, loan sales and auction (to give an example). We have over 80 offices throughout the US and Canada and I am based out of the Manhattan office. MCC really requires me to hone in on the goals of our advisors and the corporate team. We spend a lot of time gathering feedback and using this feedback to improve our platform. We've made a lot of headway in the last few years developing our CRM and some new business tool development.
But really, and I think this is what we're here to talk about today, is how we're always exploring new tech solutions and working to create more efficiencies throughout all of the things that we do. That's what really brings us to Archer. I want to just give a shout out to Richard Matricaria who is Marcus and Millichap's Western region CEO, who really saw the value in bringing Archer in and put together a strategic partnership with them, which allows us access to their generative AI platform. So this has been a really fun initiative that I've been working closely with Richard on and rolling it out to our Capital markets advisors. So I'll pause there and allow that segue to go over to Fred.
Fred Canney
That's great. Thanks, Erika, and thanks for joining us today. Everyone's really excited to be here and excited to be here with Erika and Anna. So talking about the future. Where and where are we heading and what are we trying to deal with? Real estate. So my background is a little different to what you might find in real estate. I actually started my career with a company called Newmont Mining, which is the world's largest gold mining company. I spent my career doing everything from being an intern at the start to running financial planning and analysis for Asia Pacific. That revamped how we made investment decisions for $10 investment pipeline and learned how to use data and analytics and smart analytics to make good M&A decisions.
Then, I ran corporate development for North America, our business development for North America and then I was the acting CFO for North America, where I oversaw the whole finance suite, the investment committee, accounting, budgeting, forecasting, capital expenditures, and all that type of analysis.
So how did I get into real estate? I I graduated from University of Denver with a finance degree, but in 2018 went back to get my Masters, went to Wharton school out of San Francisco. There, in their Executive MBA program and and during that program met my co-founder Thomas Foley, who really brings a commercial real estate brokerage and fintech background, and he had this idea to create revolutionary technology for the investment manager. And with my finance background and his understanding in real estate, we really jumped in to start building Archer.
We started building technology, looking at how we analyze markets better and how we identify every deal we might be able to go after and then automate the underwriting of every single deal in the nation. And how do you then capture all the data that's being generated so that you can make the best investment decisions possible? When we started down that path, building that technology for the last five years, we've been really building that. And lo and behold, 2 1/2 years in, we pivoted to a SAS platform and now I get to work with folks like Erika and her team and our wonderful customers across the board to just build as a revolutionary technology for real estate largely around that acquisitions and investment analysis. So excited to be here to talk about that!
Anna Walsh
Awesome. Well, thank thank you both. I'm gonna keep it going. Get right into the content for those on the call today. Again, thank you for attending. If you have questions, you can drop them in the Q&A box. We're gonna get to as many as we can at the end. And with that, I'm gonna turn it over to Erika and Fred. Fred, you want to get us started?
Fred Canney
Perfect. Yeah. So the problem that we experienced at the very beginning and have seen repeatedly is what brings us here to talk about underwriting in commercial real estate, and largely we'll focus on multifamily, student housing and affordable housing and that type of underwriting. There's so many data points, and this graphic at the bottom gives a slight indication of all the work that goes into evaluating a property. You open up your Excel and type in data to parsing and understanding rent rolls and T twelves that are finding the right comps to analyze. It just takes a really long time for every deal. And 90% to 98% of the deals that you look at are dead ends. They don't progress into an actual deal. And so there's a lot of wasted effort that's going into underwriting that because your tools are separated. Everyone has to do this work in Excel cause the deals are unique and underwriting is unique based on the firm you're in. There's just a lot of energy going into underwriting and it takes a tremendous amount of time. Erika, do you want to talk about what you've experienced and what the underwriting challenges have been for MCC?
Erika Banach
I can talk about it in a few different ways. The Capital markets team has about 220 employees. About 115 of those are originators at various levels of their career. We have our more senior agent originators that really focus on being out there with the clients and they may have a team that uses an analyst to really help hone in on all the deal terms and really to make them more efficient is to make the analysts more efficient, right?
We have been able to utilize the platform [Archer] to save a lot of time on getting back to the client as fast as possible. And really that's what we want to be doing. And we're lucky that we have a really great bench of analysts and people in our execution. Roles have migrated over to origination roles and because they're just starting off, they're starting to create their own models and really try to hone in on a different business model where they're not just sitting at a desk and churning out deals they need to. It's really a huge benefit for the people that are in that transition period where they can turn around. They're underwriting much more efficiently and quickly. So we've seen a lot of success there as well.
Fred Canney
Yeah, that's great. We want to. And if you want to move to the next slide here, one of the things that we started to do with Archer, we talked with different customers. Erika's team, the broader MCC team and our investment clients. We really started with the Archer model, which is an Excel based underwriting model focused on multifamily. The fastest underwriting groups that we've talked to can do a back of the envelope analysis in maybe 30 minutes. A full underwrite, including all the steps outlined here, usually takes 6 to 8 hours or longer. That's the normal thing and we experienced this too when we were doing our underwriting. As much as we were trying to capture data and automate, it's a very long process.
So we started looking at the uniqueness of the models. All the individual steps that go into underwriting (and we've got some of them laid out here): setting up your model, parsing and interpreting rent roles and T twelves and understanding the analytics behind those documents, gathering your rent comps, you might go to various websites to gather rent, comp, data expense benchmark, and you might go talk to your asset manager and have them send you examples, or you go dig through old Excel files to find comps into your benchmarking… and on and on and on.
All of these steps are distributed. And they take time, and that time is just time put into deals that often are dead ends. And so we started with the Archer model and we started automating all of this, building up a comp database, building up the property level information to identify the right benchmarks for you, building up mileage rates for every county in the nation, collecting debt quotes and pulling in daily interest rates so that you could build up a debt quote that was fairly accurate. With just a machine.
And the problem we're running into is while the Archer model could do this, folks like Marcus and Millichap Erika’s team, other clients of ours have all these other stakeholders, right? You've got your IC decision makers, you have your BOV templates, you like to put out, you've got your loan sizing templates that you might use with various clients. And so while we could achieve this automation in eight hours down into about 15 minutes within the Archer model. The big innovation we were able to launch working alongside customers like Erika and and others at MMCC was to really do this within your own model. So I'll pause there. I wanted to talk more about how that that happens, but Erika, I'd love to pass it over to you to talk about your experience.
Erika Banach
When we first started rolling out the core Archer product, a lot of what we did was listen to feedback. And the team is really apt to make changes, make modifications, and be very honest, so that was really helpful, especially when we were talking to some of the individuals that were really much more focused on doing very complex structured financing transactions. And they needed much more flexibility. The models that they have created and spent countless hours cultivating and perfecting for specific deals and for specific clients there was a lot of resistance on… they didn’t want to just throw that away. And just work in something that doesn't give them everything they need. I want my model and I want to be able to use that. That's what my clients are used to seeing and that's what the end product is. And I don't want to move away from that. It was really interesting because the Archer team took it on as a new challenge and something that we can work together to create.
Fred Canney
Yes, totally. To talk a little bit about how this works, especially for groups that may not be familiar with Archer or the type of technology that we're building, Archer comes out of the foundation of an acquisition suite. Acquisitions teams, that's probably 70% of our customer base, brokers, debt brokers like Erika’s team lenders, property managers, we now have solutions that work with all these different groups. But the genesis of the company was that we started with deal sourcing and market strategy: where do you want to invest, what deals are available to you aggregating all of that data together to make it a really simple headquarters for your team of what's available for me to pursue?
We worked on the underwriting here on the left hand side, the underwriting and automating parsing. And so a key to all of this is the storage of data from all your underwrites. The storing of T twelves that get parsed and roles that get parsed, storing of sale comps, and then all of Archer's automation layered on where we can now take just about any Excel model from any of our customers. We tie it into the Archer system, and now you've got the ability to go into the web app, find a property, click under, right. And in about a minute have your documents parsed, your model is spun up, it's pre loaded with the property information we put in mileage rates, put in growth assumptions. Your GP LP splits if that's critical for your company. And we populate roughly 90% of the model for you. And we have about 50% of our customers or 60% of our customers that use the Archer model that we had achieved all this automation with and then about 40% of our customers that BYOM (or bring their own model) to the platform now and we expect that's just going to continue growing. Because it's just such a unique way to live in the environment that our customers need to be in.
And so we're really excited about the BYOM concept. The last piece I just want to highlight is that we've also developed a way which I think is so critical to the repeatability. To achieve this, compounded insights at the end and that is really every file that's run: every parsing that's completed by the team, every underwrite that's completed by the team, gets stored back into the Archer cloud. It's private to the team. So for example, if your Florida team, I think we talked about one of these examples, your New Jersey team has a client who wants to do a deal down in Florida and they haven't done any deals in Florida. But your Florida team has done a ton and has all these comps. This compounding data insights should allow your New York team to see the comps and analysis and work that your Florida team has done to provide a better experience for your customers.
Erika Banach
I just want to interject there because what's unique about our team of originators and is really that they have been very transparent with one another. We have, I mentioned our CRM before, but there's a lot of information sharing that's already happening. So this is just adding to that layer of transparency and providing enhanced services for the client. So everybody wins. When people are putting more and more information in [Archer], it becomes more and more powerful.
Fred Canney
And I want to make sure that I'm clear so that nobody walks away with the wrong understanding. All the data that you put in is private to your team. So whenever we work with our different customers, the underwriting that's completed, the way that you're looking at deals, what price you might pay, right, all of those critical very proprietary pieces of information. As well as your proprietary model that you bring in, connect with Archer lives and breathes with an Archer for your team, but we don't extract those insights and share them with anybody else. Your comps aren't shared or anything like that. It all stays private.
Erika Banach
Yeah, that was a big, big question that everyone had. So it was easy to just get in front of it and make sure that we answered it early on.
Fred Canney
Totally.
Anna Walsh
I think that's a perfect segue, Fred, into it showing off what it looks like, so maybe I can stop sharing and you can share your screen and show our audience a quick demo of Archer.
Fred Canney
[Fred demos Archer, please watch video above or contact us directly if you’d like to see more]
Erika Banach
As everyone just witnessed, it is so fast. I was joking with Fred, even if it took two minutes, I think we'd still be OK. But just you think about it, as you know, we have a lot of teams and people working together. You can sit there and input and make modifications as you go.
And running the underwrite, you get information constantly provided to you over the course of a week. You can be constantly adding that to the product and you're just getting more and more data in there and you're getting better and more refined. People love efficiency. It's not always the easiest thing to get people to jump right on board to, but when you start seeing the benefit of the speed and the accuracy, it's a much easier sell to our advisors. Fred did a great job of explaining the better underwriting. But I do see a question and it kind of relates to that. And I'll ask Fred to elaborate. But with the parsing, it's not just [Microsoft] Excel, it's you're also able to parse. Yes, and I know you said in the questions “can you take a PM?”. I'll turn that over to Fred, but I do know you can take a PDF and get pretty close… I think it's 99% close to, or 95%, or am I overselling it?
Fred Canney
We can definitely handle PDFs and Excel files, and today we parse and extract rent rolls and T12s. We're looking at other documents, RO statements, HUD, rent statements, things like that, that are fairly standardized. And our success rate on Excel files is 96 or 98% around where there's no human intervention it can extract. Yeah, it's really close and PDF's a little bit lower, more like the 80% mark. And so Archer actually backs it up with a client success team that are all real estate folks.
Erika Banach
That was close. And to that point, I want to pay a compliment where a compliment is due. When we first started working with Archer a few months ago, they didn't have that capacity and now they do. So it's so quick. Our feedback [to Archer] is directly applied to improvements that directly benefit us, so it's really been quite helpful and our team feels that. So when they say no, I can't use this because I only get my rent rolls and PDF and that's all that [Archer] provides, that feedback has now become a solution. It’s terrific.
Fred Canney
That's been actually one of the most fun things in this role, coming especially out of a corporate role into a startup, is that we get that feedback and we get to take action on it. We get to build collectively with our customers about the biggest challenge they are facing and how we actually help bring our knowledge of data structuring and automation and smart analytics and good use of this information (filling in gaps) to actually help you achieve what you're trying to do. And groups like yours, Erika especially, you've been tremendously helpful in providing that feedback: the good, the bad, the ugly, which then helps us actually make a better platform for everybody.
Erika Banach
Yeah. And on the other note, talking about smarter comps and quotes, I think that's been a huge benefit, not just for our team to collectively understand that and receive the compounding benefit of putting information in and getting more information out, but how that impacts our clients. The more information we have, the better we know the market. The better we know the market, the better we know your neighbors. The better we can assist you in getting to your goals, so that is huge. That's a huge advantage for us and we're very happy to lean into that and take advantage of it. The best we can.
Fred Canney
Great. Yeah.
Erika Banach
I won't tell you all our secrets, though. I see here that there are the seller propensity factors and there are some real great benefits. But you know our team is going to take everything they can in order to really target clients and make sure that we're delivering on everything we set out to do and I'll let Fred talk a little bit about it, but I'm not going to. I'm not going to give away the secret sauce.
Fred Canney
I was going to say if you want to know more, reach out, happy to talk to you about how we bring seller propensity, help people find refinance opportunities or owners that might sell soon, and and how you can use predictive underwriting to really unlock some of those conversations collectively with your Marcus Millichap broker.
Erika Banach
Exactly. Thank you. That was a good plug. I appreciate that. Now, I will say that we're talking about an underwriting tool, but it's so much more than that. And I think about how I'm helping people adopt the technology. Maybe underwriting is too far of a reach. They say that “I don't want to learn something new”. But when we start giving them more information on how they can use the technology in different ways, you're giving them a little treat. You're enticing them. They're going to start using it, and that means they're going to start logging on. That means they're going to start working in it slowly, and that's going to lead to better adoption. So really that's been really helpful to have. All the other bells and whistles, along with us wanting people to really utilize the underwriting product.
Fred Canney
No, that's super helpful I think. And if you want to move to the next slide, that's a great segue. Erika, I think your comments really speak to this. We're trying to meet our customers where they're at. Whether that's working in Excel or where they're at on a technology adoption curve and understanding what their flow looks like, so that you can take advantage of real estate intelligence or the future of real estate. Before I dive into this slide, Max had a question here about plans integrated underwriting for developers doing build up. We do have developer clients today, and largely where we can be helpful is we can help build up a view of stabilization. So whether you're looking for stabilized comps expectations of operating expenses upon stabilization, for your lender, say as you're going to justify your budget, we can actually put in any address put in the hypothetical configuration. We don't currently have tools to help with what your construction cost is going to look like. How long is that ramp up gonna be? But the Archer model can be tailored to handle lease up situations and from the moment certificate of occupancy forward. We do have customers using it for that.
Erika Banach
But give them a few months. Who knows? I know one of our early adopters who is utilizing the BYOM, and it's already putting in a construction model, so they're testing it now and seeing how it works. They are pleasantly surprised as to how far they can get with a limited amount of work put in.
Fred Canney
That's awesome. So when we think about the future of real estate, ideally we want to be part of that journey for you or with you as we're bringing this technology to the market. One of the things that I think a lot about, and where we started five years ago, is this data consolidation, aggregation, standardization. It's not sexy, it's not fun, but it's super necessary. This is the dawn of AI, and every day there's another company pumping popping up with an AI application, right? And really, AI starts with some with good data, with good standardized clean training sets and data, and real estate today is pretty terrible with that. A lot of industries aren't very good at this, but real estate data is notoriously sparse.
You've got non disclosure states. We've got gaps in data providers. And so one of the things that Archer has really focused on is data aggregation, standardizing, filling gaps. And capturing the data that you're creating in Excel and making it usable for you and for so many of our customers. That's where your underwriting is going to happen. That's where your loan sizing is going to be. And for the most part, Excel is like a one way street. You do all this work in Excel, and you store it in a file structure somewhere. And next week next month, two years down the line, if you need to refer to that, you have to reference somebody's brain that goes, oh, yeah, we worked on that neighborhood deal two years ago. Let's go find that property and I'll go find the files. And it's manually finding that data.
But we're not getting the same data in real estate [as other industries] because so much of the data is locked up in Excel. So for us, it really started in how do we gather that information, get it out of Excel, store it on AWS or Azure, standardize it, and make it reusable for the user. And that's the start. Once you have that, which is what we've been building, then you can start going through automating the workflow. So that's the 8 hour underwriting down to a one minute process. And then the analyst or the associate or the broker or the originator can spend their [extra] minutes for complex deals, really analyzing. Or you got the data at your fingertips and you can actually make sense of it. And so while we're automating the busy work.
And where we're heading is as you do your work, as you parse documents, as you look at deals, we don't want any of those to be dead ends or lost work, but we know that you need to get to a no decision on 98% of the things you look at and get to the yes decision on the 2%. So how do we actually get you to focus on the 2% that matter? And that's by automating the workflow so you can get to that no faster and don't lose any of that information or insights because all of that's proprietary data for you.
And with that proprietary data captured, we can start to do some really cool AI things. We think generating IC memos, interpreting the emails from brokers as they come in, so you're not monitoring all the email blasts, but instead it's all monitored for you. That's the type of stuff where we really see technology driving what we would call real estate intelligence, which is this idea that everything you do is building on itself. And you're getting smarter and faster and better with your decision making, saying no to the right things, saying yes to the right things and spending your time matters. So Erika, I wanted to give you a chance also to talk through what you see as the future.
Erika Banach
I do want to note what you said. I think it's important that when you have a gap in the data, that you know why, and you know that that's where you're gonna spend your time. Like Fred was talking about the construction information, we're not interpreting what we anticipate the construction cost of a huge high rise in New York City. But that's the gap you have to fill out. So we have amazing teams that can. Are able to do that and put that together. But understanding the why behind the gap and the how behind the gap really is helpful and in projecting forward, I think if it was just an empty square, it would be a little alarming. But really there's information as to why that information is not populated, which I think is really helpful in the process and moving the model forward.
Fred Canney
Yeah, I think that's really important. So go ahead.
Erika Banach
Where? Yeah. Where do I see it going? I don't know. I have a laundry list of things that I want you guys to start tackling. But, we'll get there step by step. I think for me and my focus right now is really getting people comfortable with the product and making sure that they have what they need in order to be successful. We are making sure that we're meeting everybody where they are currently and getting them what they need and that's a learning process. It's been really eye opening for me personally to understand. What are the resistance points and how do we work around them? Are those things that we can't go through? And so I'm excited. We are just getting started. And the bring your own model recently [rolled out], so I'll be curious as to hear the feedback and we'll go from there, but it's a constant evolution, which is really I think quite exciting.
Fred Canney
Very exciting. I think something I'd love to throw out there is advice for the folks on the phone or for other folks that might watch this later. As commercial real estate professionals, how do you take advantage of technology or where do you go next? One of the things that I would suggest is to start incorporating some of these tools into your process, right? Obviously we would love to support with Archer, so give us a call.
We're happy to show how to learn more about what you're doing and is there a way that we can fit and help. But the other thing is, while you're thinking about innovating, be open to innovation. And have high expectations of your technology, something that we've seen a lot and heard from our customers is a lot of tech has come into real estate, not from real estate and with the promise that it's going to solve what real estate couldn't.
And instead at Archer, we're really approaching it from our investments background. We have a real estate background. Our customer success team, our real estate folks with years of brokerage acquisitions, asset management, collectively they're over 50 years of experience in real estate. We're really trying to build with our customers to create solutions that actually help as opposed to coming with an individual point solution and a promise that we can do it better. And then the other thing that I would really suggest is as you look at your underwriting, think about all the different parts that go into that. And look for areas to standardize, automate and really streamline so that you can get to the no decision faster.
And we've heard repeatedly it's gotten worse and worse. Recently, I heard for 40 deals you look at to 1 you might bid on. And you go in, analyzing 40 deals and say we're really good and it's two hours per analysis. We're taking two weeks of analyst time to put in one dead end when we should be spending that 80 hours on the 11 properties we truly want to pursue. And that's where I think we can really be helpful is to help you get to those nos faster. Store all your data so you get better and hone focus where it actually matters. So. You might think of Archer as a virtual deal assistant. Serve an analyst right alongside you, trying to help you do your work faster. But that's really what we're here to do with the bring your own model and the Archer model as well.
Erika Banach
Yeah. And I hear this sentiment a lot in conferences and when I'm out. AI is a tool. But in order to apply it effectively, you really have to take a step back and understand what your strategy is. And if your strategy is leaning towards underwriting, you want to make a plug with and use AI in that underwriting channel. And then work to apply the appropriate tools to those processes and not the other way around. I know I'm inundated constantly with the newest thing and people from the field are bringing me lots of fun things to look at and it's great. But if it doesn't align with where we're going and where we're trying to get to, it's really not as helpful as you think it is. So get your strategy set and then start applying the appropriate tools to that.
[Erika and Fred answer audience questions]
Anna Walsh
Excellent. Thanks so much. And I think that wraps up the questions. Thank you all so much for attending. As Fred has alluded to, this is being recorded and we'll distribute later, if you'd like to share with any of your colleagues or take a look again at some of what you've learned and if you want to incorporate some of this into your own workflow, definitely reach out to Archer, or both Fred and Erika are on LinkedIn if you have any questions for them directly. Thank you, Erika, and thank you Fred, so much for being here today. This is so great. And with that, I'm going to hop out. Thank you everyone.