Leverage the full potential of your Multifamily Real Estate investments with insights from Elizabeth Braman, CEO of RevolutionRE.
As our special guest, Elizabeth unravels the complexities of AI and data standardization in apartment investing, offering strategies for multifamily businesses looking to enhance their efficiency. We delve into the ways companies can align their data strategies with their overarching goals, transforming the way we think about property investment and management.
Navigating the often-treacherous terrain of data integration, our conversation with Elizabeth Braman focuses on creating a seamless and scalable system that can stand the test of time.
She highlights the importance of crafting KPIs that truly reflect a company’s performance and the critical nature of involving the entire team in the data journey.
We also tackle the real-world challenges that crop up when trying to wrangle various data sources into a single, coherent entity, particularly in the Multifamily Real Estate industry where standardization can feel like a distant dream.
Rounding out our deep-dive into the data-centric world of Multifamily Real Estate, Elizabeth sheds light on the game-changing impact of her innovative approaches to data standardization and the role AI plays in revolutionizing the multifamily industry.
We explore how historical data reveals more than just past performances.
Connect with Elizabeth Braman
Thank you for tuning in to today’s episode. If you found value in our conversation, please subscribe to our podcast on your favorite platform and leave us a rating and review.
Your feedback not only helps us improve but also helps others find us. And if today’s episode sparked a thought or provided a new insight, consider sharing it with a friend who might also benefit.
Together, we can grow our community and continue to learn and innovate. Thanks for listening, and until next time.
About the Multifamily Innovation® Council:
The Multifamily Innovation® Council is the executive level membership organization that makes a difference in your bottom line, drives a better experience for your employees, and allows you an experience that keeps demand strong for your company. The council is uniquely positioned to focus on the intersection of Leadership, Technology, AI, and Innovation.
The Multifamily Innovation® Council is for Multifamily Business leaders who want to unlock value inside their organization so they can create better experiences and drive profitability inside their company.
To learn more or to join, visit https://multifamilyinnovation.com/council.
For more information and to engage with leaders shaping the future of multifamily innovation, visit https://multifamilyinnovation.com/.
Connect:
Multifamily Innovation® Council: https://multifamilyinnovation.com/council/
Multifamily Innovation® & AI Summit: https://multifamilyinnovation.com/
Patrick Antrim: https://www.linkedin.com/in/patrickantrim/
Speaker 1: Welcome back.
00:00:00
Today's guest is Elizabeth Brayman.
00:00:03
She is the founder and CEO of Revolution RE.
00:00:06
That's a SaaS company providing data solutions to apartment
00:00:10
owners.
00:00:10
They do this for managers and really any industry-related
00:00:15
company as more and more data is coming together.
00:00:18
Prior to forming Revolution RE, elizabeth served as the chief
00:00:23
production officer at Realty Mogul.
00:00:25
Forming Revolution RE, elizabeth served as the chief production
00:00:27
officer at Realty Mogul.
00:00:28
This was an online platform raising capital for real estate
00:00:32
companies through private placements and A-plus REIT
00:00:33
offerings.
00:00:33
Prior to that, she was a chief production officer at ReadyCap
00:00:36
Commercial, a small balance sheet commercial real estate
00:00:39
loan originator funding loans from private REIT offerings that
00:00:44
has since gone public.
00:00:45
Now Elizabeth is a certified commercial investment member
00:00:50
this is the CCIM designation and a licensed attorney in
00:00:53
Washington DC, in the Commonwealth of Virginia.
00:00:55
She holds a Bachelor of Arts degree from the American
00:00:59
University and a Master of Business Administration and JD
00:01:03
from the George Washington University.
00:01:05
Today, what I'm doing is I'm bringing in Elizabeth to share
00:01:08
her knowledge on how AI and data standardization are key to
00:01:13
unlocking new efficiencies and opportunities in apartment
00:01:15
investing.
00:01:16
Elizabeth, welcome to the Multifamily AI Podcast.
00:01:20
Speaker 2: Awesome.
00:01:21
Thank you, patrick, it's great to be here.
00:01:23
Speaker 1: Yeah, so much background and expertise in what
00:01:27
you do from a legal aspect.
00:01:29
I love that when we talk data, you're not here to give public
00:01:32
and legal advice, and nor am I, but it's interesting.
00:01:35
That type of background, serving at the leadership role
00:01:39
of a company solving data, is compelling, obviously One of the
00:01:43
questions I'm thinking about for leaders, business leaders.
00:01:46
So on this podcast I'm trying to help owners and operators
00:01:51
that may be from sort of non-technical roles understand
00:01:54
the value that can be unlocked leveraging data in their
00:01:57
business, like they leverage debt as a business tool, and
00:02:01
would love to know what you think those questions should be
00:02:05
as people think about their data strategy.
00:02:09
Speaker 2: Great question and I think it really obviously
00:02:12
depends on the organization.
00:02:13
I think the first thing I recommend is really taking a
00:02:19
look in the mirror and saying all right, what is a data
00:02:23
strategy?
00:02:24
What does that mean to our organization?
00:02:26
Because it can mean a lot of different things.
00:02:29
And what really drives the data strategy should be your
00:02:34
business goals and objectives.
00:02:36
What are you looking to accomplish with data?
00:02:39
What do you think you can accomplish with data?
00:02:41
And then look at things like how does it impact governance
00:02:47
and data security and all of those things.
00:02:50
But really starting with your goals, your objectives defining.
00:02:56
I'm a big fan of smart framework .
00:02:58
Is it specific, measurable, achievable, relevant and
00:03:02
time-bound?
00:03:02
And I go through this exercise with our clients as well,
00:03:07
because when they're working with us on developing their data
00:03:11
strategy, it's really important that it's defined, that they
00:03:16
know who their key stakeholders are, that they get buy-in from
00:03:22
their key stakeholders and that they're measuring what success
00:03:27
means to them.
00:03:28
Because data strategy is one of those things where, if it's too
00:03:32
amorphous, it's too vague, there's really no success metric
00:03:36
that an organization has.
00:03:39
That's when a lot of time, effort and energy is put into
00:03:43
things that eventually get abandoned because there's no end
00:03:47
point to it and people can't look and see why are we doing
00:03:51
this, which is important because you need your team to really
00:03:55
participate in the process of implementing a strategy.
00:03:59
It can't be done by a vendor or in a silo.
00:04:03
One person can't take it on all by themselves if it's an
00:04:07
organizational data strategy.
00:04:10
Speaker 1: And getting into how to actually define this.
00:04:13
Many business leaders you mentioned we think about
00:04:16
dashboards, we think about the data that we've always had to
00:04:20
make P&L decisions, knowing what's going to move the
00:04:25
investment needle in the business.
00:04:26
So a lot of the business leaders are more aware of their
00:04:30
data than they probably want to admit, because it's really it's
00:04:34
decision making right.
00:04:36
But when we introduce AI and different, we're talking about
00:04:42
standardization challenges and things like that.
00:04:44
When it comes to solving those problems, oftentimes a leader
00:04:50
will pass it off Okay, that's IT or that's somebody else's job
00:04:55
internally.
00:04:56
I'm curious, with the acceleration of technology and
00:05:00
talent, are these things companies should be doing inside
00:05:05
their organization or relying on outside parties to guide them
00:05:10
through right?
00:05:11
And then what I'm curious about is what's the leader's role in
00:05:17
understanding that data strategy ?
00:05:20
Is it something they just hand off to IT or their technology
00:05:24
team?
00:05:25
Maybe they don't have one?
00:05:27
Speaker 2: Yeah, I was going to say goodness, no, I would hope
00:05:29
not.
00:05:29
You don't want to give the poor IT a data strategy rollout
00:05:34
without a lot of input from the business stakeholders.
00:05:38
Those two things can't be done in a vacuum.
00:05:42
I think if an organization does have an IT group and a lot of
00:05:46
companies don't big real estate companies that have I shouldn't
00:05:50
say big the larger real estate companies obviously have
00:05:55
internal teams, but there are companies that have billions in
00:05:58
AUM that don't have a large enough team to take this on
00:06:04
independently.
00:06:05
So it depends on the size of the organization.
00:06:07
But it's either done as something in tandem with the IT
00:06:12
team or with external vendors.
00:06:15
But picking the right external vendors can be great or it can
00:06:19
be not great.
00:06:21
Getting good references, but also who you connect with, how
00:06:24
do you meld in terms of your work?
00:06:29
Because it's going to be a process and there is no quick
00:06:34
fix.
00:06:34
You can't bring a vendor in and just wave a magic wand and say,
00:06:38
okay, data, that's not a thing.
00:06:41
Standardization obviously can help the process, but there's
00:06:46
still a requirement of every real estate company to
00:06:51
participate in the process of and I decide whether it's a
00:06:55
long-term relationship that you want to build from there and
00:07:07
look at this kind of as a long-term strategy.
00:07:11
You can't spin up a complete integration with every single
00:07:18
data source and data plan and have it done in a week.
00:07:22
That's having reasonable expectations is critical.
00:07:26
Getting your vendor or your IT team to buy into those goals,
00:07:32
expectations, timelines, all of that is really critical to
00:07:38
having a successful data strategy rollout.
00:07:40
But I think it's really up to the company to work with
00:07:44
external or internal.
00:07:45
There's obviously benefits to having someone with completely
00:07:50
fresh eyes look at your organization and help guide that
00:07:53
process, and it's also important for internal folks who
00:07:57
really know all the nitty gritty of how data has been
00:08:02
captured and stored.
00:08:03
So a little bit of both, I'd say, or a lot of both, depending
00:08:07
on how big your data strategy is and your goals are.
00:08:12
Speaker 1: Yeah, and I love the idea of the expectations and a
00:08:15
lot of times when we're evaluating even technology, I
00:08:18
guess you can group data into a almost like a product right
00:08:22
Productization of round, I think you've called it.
00:08:23
Is it's standardization as a service right?
00:08:24
Productization of round, I think you've called it.
00:08:25
What is it?
00:08:25
Standardization as a service right?
00:08:27
But, more importantly, in terms of those expectations,
00:08:30
sometimes leaders are looking okay.
00:08:32
So when I have a problem, what's the problem?
00:08:34
This technology or this strategy is solving and in many
00:08:38
cases there may, depending on the leader's role and how they
00:08:42
are up to speed on where things are in the business.
00:08:45
We may not know what we don't know.
00:08:47
So how do you come into an organization and manage those
00:08:51
expectations and demonstrate the value of getting this
00:08:55
standardization correct?
00:08:56
Because it seems like a lot of people want to do it, but
00:08:58
there's some people run into some challenges doing it.
00:09:02
Speaker 2: Yeah, for sure.
00:09:03
I think I always laugh when someone asks me is it done?
00:09:07
It's never done.
00:09:09
It's a process, having a really well-structured process, and
00:09:16
that's why I use the SMART framework and kind of work with
00:09:19
teams when we're engaged in the onboarding process to really
00:09:24
define everything they want to accomplish and then trying to
00:09:28
prioritize and work through timelines for having small wins
00:09:34
so that you're iterating and building it out over time.
00:09:37
You have to see some measurable impact when you're looking at
00:09:43
data and so, depending on how complicated your tech stack is,
00:09:50
how many data sources and data sets you're looking to bring in,
00:09:54
how different the data structure is to start, and it's
00:10:00
why people look for scalable solutions to this stuff so that
00:10:04
they can find a way to make a meaningful impact right out of
00:10:10
the gate.
00:10:11
Just even just getting all your data in one place is an
00:10:15
accomplishment.
00:10:18
Right formats, having daily access to property data from
00:10:25
multiple systems that's a big win to be able to get that done.
00:10:29
Then you look at okay, how do we refine our KPIs?
00:10:36
What kind of metrics do we want to start tracking?
00:10:39
What different teams within the organization want to be
00:10:43
involved?
00:10:43
What different teams within the organization want to be
00:10:46
involved?
00:10:46
Is it all marketing or is there the leasing team, the
00:10:52
management boots on the ground?
00:10:53
Who is this really going to impact and how do we get them
00:10:55
from step one to step two throughout this journey of data,
00:11:00
but there is no done.
00:11:01
Of data but there is no done.
00:11:02
It's always just a continuous, never-ending product process,
00:11:06
which is why you have to stay on top of it and create some
00:11:10
business roles internally.
00:11:11
But it's a commitment.
00:11:13
There's definitely a return, which is also why you want to
00:11:16
have measurable.
00:11:17
See how you're getting a good return on rolling out a strategy
00:11:23
.
00:11:23
Yeah, I like the small wins.
00:11:26
Speaker 1: I like what you said about the small wins and making
00:11:28
it measurable when data is coming from so many different
00:11:31
places.
00:11:31
What are the challenges of getting this?
00:11:33
If it's an accomplishment to get your data in one place right
00:11:36
?
00:11:37
Speaker 2: Yeah.
00:11:37
Speaker 1: Are there examples that even other industries are
00:11:40
doing that we could learn from?
00:11:43
Speaker 2: Yeah, it's.
00:11:44
The big difference between real estate and other industries is
00:11:47
it's not the most open framework .
00:11:50
I think.
00:11:50
If you look at banking, you can add your credentials into a lot
00:11:54
of systems and all of a sudden you've got a direct connection
00:11:57
with your bank.
00:11:58
Or marketing, if you want to get all of your Google Analytics
00:12:02
in one place, open APIs.
00:12:03
That's not quite the case in multifamily, and so we're
00:12:09
extracting data from core systems.
00:12:12
Sometimes we have one, two, three different methods of doing
00:12:15
it.
00:12:16
It's challenging because you've got many stakeholders involved,
00:12:20
whether it's the management company that's generating and
00:12:23
storing the data, ownership groups who own some of the data
00:12:28
to different property management systems that all have different
00:12:31
data structures.
00:12:33
So extracting the data is the step.
00:12:36
One is getting it into a common and consistent format.
00:12:40
That's the transform.
00:12:41
So ETL is the extract.
00:12:45
The transform and then loading it into a place where you can
00:12:49
now use it for BI is just a front-end visualization of data
00:12:56
that's in a structure that you can use.
00:12:57
But if it's coming from multiple systems and it's in
00:13:00
multiple structures and it's in multiple structures, having a BI
00:13:08
front end and trying to create multiple reports from those
00:13:10
source systems is a lot of work and it's why a lot of
00:13:13
organizations are challenged with their reporting if they are
00:13:17
trying to pull it all together and generate portfolio-wide
00:13:21
analysis.
00:13:24
Speaker 1: And obviously having a standardization of data plays
00:13:26
a huge role in that process, right?
00:13:28
Are people skipping steps in this?
00:13:32
Do you think right now?
00:13:34
Speaker 2: Yeah, I think if you're doing a bespoke approach
00:13:39
and it's reasonable, it makes sense.
00:13:41
If you have one system of record and you set it up to feed data
00:13:46
into a front-end visualization and then you add a second one,
00:13:50
now you're doing the same process twice side by side.
00:13:54
But it doesn't really mesh unless you take a step back and
00:13:59
get all that data in one format.
00:14:01
Because these systems are not consistent, there isn't an
00:14:06
industry-wide this is our standard for how data is
00:14:10
structured and because there isn't one way of doing it, you
00:14:16
are limited to either having to standardize it internally and
00:14:20
create a data model, and sure not a lot of real estate
00:14:23
companies want to create a data model internally.
00:14:27
It's expensive, time-consuming, requires specialized knowledge
00:14:32
and it's just messy.
00:14:32
So that's why we built what we built specifically for the
00:14:38
industry, because it didn't seem like there was a solution that
00:14:43
was out there that would provide that same aggregation of data
00:14:48
for people to be able to then use it for AI applications, for
00:14:54
BI applications, for all the cool technological use cases,
00:14:59
because the data just wasn't in a format that made it easy to
00:15:04
use the data just wasn't in a format that made it easy to use.
00:15:11
Speaker 1: And when it's in this format, when you mentioned the
00:15:12
ETL extract, transform and load, are there things that or
00:15:13
questions we should be asking around?
00:15:14
Making sure that the quality of that process gets the like, how
00:15:20
do you ensure that process?
00:15:22
How do you make sure that ETL process gets the right data in
00:15:26
the right places?
00:15:27
Speaker 2: Yeah, that's a good question.
00:15:28
What do they say?
00:15:29
Garbage in, garbage out, so we can clean up some of the data
00:15:34
that comes into our system.
00:15:36
You can't fill in empty holes.
00:15:40
If there's big blanks in data, there's no way to create it.
00:15:44
Ai is also helping to clean and create methods of cleaning data
00:15:53
, and a lot of these models don't need perfect data.
00:15:59
They need a lot of directionally correct data to
00:16:02
generate really interesting results and to be impactful.
00:16:06
Speaker 1: This could just also be like just as simple as a
00:16:09
state record right, A field where it's maybe open text in
00:16:14
one database and the other it's abbreviation.
00:16:18
Speaker 2: Absolutely.
00:16:19
Our resident name is last name, comma, first first name and
00:16:23
somewhere else it's a first name field and a last name field.
00:16:26
It's all across the board, getting the data in just a
00:16:31
format that you can query, and everyone who's used Excel with a
00:16:35
sort function knows that like it needs to be a little bit
00:16:39
clean to be able to use effectively.
00:16:42
Some of the AI solutions are now starting to fill in the
00:16:46
blanks of helping to clean things up.
00:16:49
So we in our tool if it's Google Analytics, google with
00:16:54
three O's, we can combine all those and essentially for ad
00:16:59
sources, we're doing some cleanup on data like that, where
00:17:02
we're just making it more consistent when at the property
00:17:06
level it's being input slightly differently.
00:17:09
Our financial model, which takes millions of rows of data,
00:17:14
different categories every company has what?
00:17:16
Four, five, 600 rows of financial data and mapping it to
00:17:22
a more tight, cleaner set of data points so that you can do
00:17:28
benchmarking.
00:17:28
We call it our common model.
00:17:30
There's huge value in having even a little bit of cleanup of
00:17:36
the data so that you can do more comparative analysis and market
00:17:41
analysis.
00:17:44
Speaker 1: What can go wrong in that process, or what should we
00:17:46
avoid doing?
00:17:47
Maybe someone's trying to do this in-house, or oh, it's a
00:17:53
good question.
00:17:53
Speaker 2: There's tons of things to avoid, but I'd say you
00:17:57
want to.
00:17:58
There's a lot of monitoring and tracking that your data feeds
00:18:01
are actually active and they aren't stopping for whatever
00:18:05
reason.
00:18:06
You might have a night where a system just can't handle.
00:18:11
It times out.
00:18:12
It's too much data coming in.
00:18:15
Speaker 1: In those types of timeouts.
00:18:17
Does that stuff get queued or is it a restart?
00:18:21
Speaker 2: Yeah, that's a great question.
00:18:22
It depends.
00:18:23
So sometimes it just like stops and you have to go and you get
00:18:27
notification that you have to start over.
00:18:30
Then there's sometimes you can just pick up where you left off.
00:18:35
And if it's a timeout, there's also time zone issues, which are
00:18:39
really strange.
00:18:40
So if you're doing a nightly and you have some data that's
00:18:43
coming in, but for the change in time zones things will happen
00:18:48
before 12 o'clock and then sometimes after, which makes it
00:18:51
a different day.
00:18:51
So you have to make sure that you time all of your data feeds
00:18:56
so that they're not impacted by daylight savings or, if you're
00:19:03
in a different country, different things like that.
00:19:07
So there's definitely ways that the extract process can fail.
00:19:11
Then there's ways that the data can be off.
00:19:17
A lot of people like to do.
00:19:18
Timing issues can happen.
00:19:20
So managers will come in on Monday after a long weekend and
00:19:26
they input their move outs from over the weekend and then you've
00:19:31
got different timing issues that directionally, can make
00:19:35
data not match from one system to the next.
00:19:40
So that goes back to expectations.
00:19:44
Speaker 1: Yeah, exactly, and I'm also like from all of this
00:19:48
work, it's the business decisions, almost like the
00:19:51
construction process is very it can be messy, yeah, and it is.
00:19:56
Speaker 2: You're building a house a lot of dust, a lot of
00:19:59
unknown things, which is why having a good blueprint ahead of
00:20:02
time keeps you from having structural problems.
00:20:06
If you set it up wrong or build a house wrong and then have to
00:20:10
go back and try and figure out like oops, we didn't put the
00:20:15
plumbing in, or the electricity is in the wrong place, Like
00:20:19
those are the things that you wish you had done.
00:20:21
They say derf to do it right the first time.
00:20:24
Sometimes time makes it hard, because if you've been working
00:20:29
on this for a very long time, things change and you have to go
00:20:33
back and revisit some of those structural things.
00:20:36
But that's technology.
00:20:38
It's always an iterative process.
00:20:41
Speaker 1: When you're in these conversations with executives
00:20:43
today and you're working through this, what are they leaning
00:20:45
into and what are they pushing?
00:20:47
Not only say pushing back, but more cautious on?
00:20:50
Or if we don't understand something, it's hard to make a
00:20:55
decision about it right, and so there's going to be a lot more
00:20:58
education, I think, around this and unlocking the value.
00:21:00
Because you mentioned the different stakeholders,
00:21:03
collaboration between teams and I always think about the
00:21:08
dashboard is like a lot of the technology that we're using is
00:21:15
making the assumption that we're going to be in an office
00:21:18
looking at a computer, using it.
00:21:19
So if you have this centralized data standardized and it's gone
00:21:25
through the ETL process, then that can be unlocked in mobile
00:21:30
like all over right.
00:21:31
You can take that wherever it needs to be.
00:21:35
Speaker 2: Yeah.
00:21:36
Speaker 1: Yeah.
00:21:37
Speaker 2: That's the load portion.
00:21:38
You can load it into a number of different applications or
00:21:43
uses.
00:21:43
But I think a lot of people are rightfully so leaning in on
00:21:47
corporate data governance, wanting more visibility into
00:21:51
tracking their data feeds and making sure that they're
00:21:56
reporting on the right information, especially if
00:21:59
they're a public company or have REITs with reporting
00:22:04
requirements.
00:22:05
All of those things are critical to have a handle on how
00:22:10
the data is coming in.
00:22:11
Speaker 1: Tell me more about that data governance.
00:22:12
Speaker 2: Explain that, what you mean by that in, so explain
00:22:18
that what you mean by that.
00:22:19
A lot of companies, at least large corporations, are looking
00:22:21
at how they secure and store their data, how they're
00:22:25
maintaining a handle on the accuracy and also the security
00:22:32
issues related to data.
00:22:34
The multifamily industry we're dealing with residents and
00:22:38
residents' housing, and housing is obviously a highly regulated
00:22:43
industry.
00:22:43
The payments and activities of residents, their data is PII, so
00:22:52
the storage of their data can be highly regulated.
00:22:56
There's a lot of different.
00:22:58
So when we look at corporate governance, it's really how does
00:23:07
a company maintain the highest integrity and policies and
00:23:10
documentation and legal structure around how they are
00:23:14
storing, using, maintaining data ?
00:23:17
So it's some effort to be done internally for big companies to
00:23:22
make sure that they're putting a plan in place and documenting
00:23:26
it appropriately.
00:23:28
Speaker 1: And do you offer?
00:23:28
Is your program a data store or is it wherever they want?
00:23:33
Is it going to other places, or how does that work?
00:23:37
Speaker 2: It can be yeah, so some companies look to have
00:23:42
their data fed into a data warehouse.
00:23:45
Some companies want us to maintain and store their data
00:23:52
for them.
00:23:52
We have SOC 2 compliance, so all of our data is encrypted in
00:23:56
motion at rest.
00:23:58
Very few people have access to the data and all of our policies
00:24:03
and procedures are available in our SOC 2 reports, which you
00:24:07
can get on our website.
00:24:08
But you have to sign an NDA because our auditors require
00:24:12
that.
00:24:12
But the whole process is not an insignificant amount of effort,
00:24:16
but it's really important and something that I always
00:24:19
encourage people to talk to their vendors about, because it
00:24:24
is incumbent upon the real estate companies to make sure
00:24:27
that they're the ones that could be liable for any data breaches
00:24:32
if they're not maintaining good practices, if they're being not
00:24:39
careful with how they store and maintain their data.
00:24:44
Speaker 1: And leaning back into AI, how is AI impacted?
00:24:47
It seems like a lot of these transformations and
00:24:51
standardizations are getting easier.
00:24:53
There still needs to be the governance side of things.
00:24:57
It's one thing to turn something over to AI, but I'm
00:25:00
under the belief that people begin and end the process, so
00:25:04
maybe it's taking out some level of review.
00:25:07
But how is AI impacting the standardization process for your
00:25:11
perspective?
00:25:12
Speaker 2: Well, it's a great question and there's obviously
00:25:15
lots of different applications for AI.
00:25:18
The AI is thriving.
00:25:21
The need for standardization because AI needs like data in a
00:25:26
structure that it doesn't have in its knowledge base.
00:25:35
It kind of lies and gives you answers that you're a little
00:25:39
confused with because it's like what is it talking about?
00:25:42
It makes stuff up and that's because it doesn't have the
00:25:46
right data to answer the questions.
00:25:50
It tries to fill in the blanks.
00:25:52
It's being helpful, but not really the blanks.
00:25:58
It's being helpful, but not really.
00:25:59
So there's certain things that AI is driving in terms of the
00:26:00
need for standardization, and then there's, equally, ai is
00:26:04
providing standardization with more ways to make data clean,
00:26:12
usable, accessible, so they feed each other in certain ways, and
00:26:17
how much faster and better the technology is at letting us make
00:26:24
informed models using available data.
00:26:29
So it's requiring less and less as long as the models are
00:26:35
trained and maintained.
00:26:37
But yeah, I know it's interesting.
00:26:39
They feed each other in some ways back and forth.
00:26:42
But there's so many AI applications and real estate has
00:26:46
been pretty lagging in how these applications are being
00:26:53
utilized.
00:26:54
I think there was a good study McKinsey put out and if anyone's
00:26:57
interested, feel free to email me.
00:26:59
I'll share it, but it was talking about how the data is
00:27:02
really the shape of data.
00:27:04
The lack of a foundational data model can really hinder
00:27:08
people's ability to use data in a lot of these AI applications.
00:27:14
Speaker 1: And these large language models are moving so
00:27:17
fast and it's so expensive to do that.
00:27:19
It's interesting, like when you compare it with your private
00:27:22
company data in an environment where it's not training the
00:27:26
large language model.
00:27:27
But it goes back to what we're talking about here, which is, if
00:27:30
you don't have the standardization of the data or
00:27:33
in the data in the right place, you can't retrieve it and then
00:27:37
get better responses when you're using AI applications inside
00:27:42
workflows.
00:27:42
And so I always wonder, as you're moving companies' data
00:27:46
into a single source of truth and also standardizing it, what
00:27:52
are you seeing the challenges there?
00:27:55
Is it they have different systems?
00:27:58
Maybe they're a third party operator and they don't just
00:28:01
have one system?
00:28:03
They're trying to manage multiple systems and I imagine
00:28:06
training people and all of that comes into play.
00:28:08
But in terms of the technical, what are those challenges in
00:28:12
bringing all these things together?
00:28:15
Speaker 2: Yeah, that's exactly right.
00:28:16
It's ownership groups who work with multiple property
00:28:19
management companies.
00:28:20
It's property management companies that have multiple
00:28:24
ownership groups that require them to use different core
00:28:28
systems and or are buying property management companies.
00:28:33
So sometimes it's just a company buys another company,
00:28:37
and it's just the nature of real estate.
00:28:39
It's transactional in nature.
00:28:41
So when you have a sale of one property to another buyer, are
00:28:48
they going to use the same system, are they going to use a
00:28:53
new system, and so you've got a lack of kind of historic data
00:28:59
and then you're onboarding it into a new system.
00:29:01
So anytime you have people manually adding data into
00:29:05
systems, that also add some room for error, we might say and
00:29:11
there's just, they're inconsistent.
00:29:12
They're not.
00:29:13
These systems weren't made to be exactly the same and have the
00:29:19
same definitions and have the same data structure.
00:29:22
They just they weren't set up that way.
00:29:24
And there's not that many systems out there.
00:29:29
It's not.
00:29:29
The percentage of the real estate market that's
00:29:32
professionally managed is that it's consolidated, but those
00:29:38
companies aren't using a ton of different core systems and those
00:29:42
core systems really are capturing all the financial data
00:29:45
, all the operational data, and then you have all this ecosystem
00:29:49
of prop tech companies that kind of have opened in the last
00:29:55
however many years, really taken off, and so there's lots of
00:29:58
stuff, but that's just adding more data to the problem.
00:30:01
Where is it being stored?
00:30:03
Where the core systems aren't storing that data, and these new
00:30:09
PropTech solutions aren't storing all the underlying
00:30:13
property data.
00:30:13
So you're just creating additional data silos.
00:30:17
You're exacerbating the foundational issues that exist
00:30:21
with new technology.
00:30:23
So it makes your head explode.
00:30:26
Speaker 1: Yeah, it can.
00:30:27
When you think about these companies, companies', entities
00:30:31
stay around longer than properties in the investment
00:30:34
cycle, are shorter term, right, and so we're asking people to
00:30:38
think longer term about things when they're in a short term
00:30:41
investment cycle, right.
00:30:42
So that's always been a challenge.
00:30:45
But if you think about the entities themselves and also the
00:30:48
data, you may have company data and then you have property data
00:30:51
and in the way that all of that is organized, I'm just curious
00:30:58
if and it's probably true, this was my belief in that when a
00:31:02
developer builds an apartment building, the utilities and the
00:31:07
city infrastructure is value that a new buyer doesn't have to
00:31:11
come in and do and do.
00:31:13
In other words, because somebody took the risk and developed the
00:31:17
plumbing, the electrical, the utility, all the things that are
00:31:25
required under the ability to go vertical, the new buyer
00:31:26
obviously gets that right, but nobody really values it on the
00:31:29
sale of an asset, right?
00:31:31
Yes, but when we talk about, if you just think about the
00:31:35
listeners, just think about all the different entities you have
00:31:37
you may have your company entity and then a holding entity and
00:31:40
then each property has an entity .
00:31:41
Is that a way to think through data in terms of if we get
00:31:46
property data at a specific location that later, if there's
00:31:50
a disposition or a consolidation or merger, that there's
00:31:54
enterprise value that can be realized on the exit by making
00:32:02
investments in getting this clean data for the new buyer.
00:32:06
Speaker 2: Yeah, I totally think so, but it's really up to
00:32:10
sellers to view that as an asset or buyers to ask for the
00:32:17
historic data of the seller.
00:32:20
They may or may not feel comfortable doing that.
00:32:22
Speaker 1: You're lucky to get a PDF today, I know right.
00:32:24
Speaker 2: Are they going to?
00:32:25
I guess, if the market gets tight enough, you could put that
00:32:29
into purchase contracts, that you want two years of historic
00:32:34
data or access to historic data files.
00:32:37
I think it'd be really hard to get a seller to agree to it, but
00:32:41
it would be great for a buyer.
00:32:42
Speaker 1: Let me just ask though that's the business kind
00:32:45
to me again what would that mean if you had that on a new
00:32:48
acquisition?
00:32:51
Speaker 2: If you had it on a new acquisition, you could
00:32:53
probably track the trends of your performance a lot better.
00:32:56
So if you're trying to and I think this is one of the best
00:33:00
use cases of data is to show your investors that your
00:33:05
operational strategy brings alpha, that you have somehow
00:33:10
managed to create opportunity in the asset by doing X, y and Z,
00:33:18
that's hard to prove when you're starting with ground zero and
00:33:22
there's nothing to compare it to .
00:33:23
If you had the historic data, you could show what's been doing
00:33:27
this for the past 12, 24, 36 months and even adjusting for
00:33:32
market conditions.
00:33:32
And even adjusting for market conditions.
00:33:35
This is how much value we've been able to add.
00:33:38
This is how much that we're putting into the property.
00:33:47
That's creating a bump in NOI and this is the multiple that we
00:33:50
, as a data-driven company people like to say we're a
00:33:53
data-driven company to prove it, to actually show that what
00:33:56
you've done has created this major impact.
00:33:59
I think that's super powerful and it's hard to raise capital
00:34:02
right now.
00:34:03
That's a tremendous use case.
00:34:07
Speaker 1: We're talking about the apartment car facts
00:34:09
basically.
00:34:10
Speaker 2: Yeah, some people have been called it what's like
00:34:13
the for hotels, the report that kind of shows the market value
00:34:18
for that.
00:34:18
Speaker 1: yeah, sure, but it'd be interesting.
00:34:20
You buy a property and this is the data from the time that we
00:34:23
owned it, but imagine, as it transferred ownership over three
00:34:30
decades, that it would be more valuable in terms of
00:34:34
underwriting, repairs, maintenance and remaining life.
00:34:36
There's so many other.
00:34:37
Where are we in all?
00:34:39
Speaker 2: of this.
00:34:39
We can do that with financials.
00:34:40
We can do that with the financials.
00:34:41
So if you have a property that you're looking to acquire and
00:34:45
you have the trailing 12 months and you can get maybe the past
00:34:49
couple of years, we can show from a financial perspective how
00:34:54
you can compare from a financial perspective.
00:35:00
How you can compare because we can convert it into a common,
00:35:02
consistent format with your current performance, your
00:35:03
current tracking.
00:35:04
So being able to take your pro forma track to that track, to
00:35:09
budget, but tracking to historic performance.
00:35:11
It's a really interesting use case and, for sure, something
00:35:15
that I encourage people to do, if not with our platform, but
00:35:19
with their own internally see how a property was performing
00:35:25
and how your efforts have changed the story at that asset.
00:35:32
That asset is really compelling when you're going out to market
00:35:35
and trying to get people to invest in you.
00:35:36
They say past performance is not an indicator of future
00:35:43
results.
00:35:43
But when you can show comparative how this property
00:35:44
was doing compared to the market and then how we're doing
00:35:46
compared to the market, it just tells a really interesting story
00:35:50
of performance.
00:35:54
Speaker 1: Yeah, that's interesting and I know that
00:35:56
there's so much more that we can cover.
00:35:58
On all of this, I go back to where we started, which was
00:36:01
getting those small wins.
00:36:02
I've been calling it crawl, walk, run right.
00:36:05
So it's maybe somebody listening, has access to data,
00:36:08
obviously, but maybe doesn't have a defined strategy in a
00:36:12
market where that alpha beating the returns in other ways than
00:36:18
just passing on rent increases if we can find ways to innovate,
00:36:22
the business having that data is huge in that decision-making
00:36:27
process.
00:36:27
Do you ever find people get overwhelmed with this because
00:36:30
they are not data scientists and there's not an abundance of
00:36:35
people to help with this and relying on the tools you have
00:36:40
available to you?
00:36:41
And there's just gaps, that they don't see, that they're missing
00:36:45
, and we can go down the whole unstructured data conversation
00:36:48
and getting that into a structured Like that's a whole
00:36:50
nother, like phase two probably, of unlocking value.
00:36:54
But are you feeling or do you see that people get overwhelmed
00:36:58
with this conversation and how do they prioritize making the
00:37:02
next move and making sure it's going to drive actual
00:37:05
effectiveness in the measured results?
00:37:08
Speaker 2: That's a really good question, I think.
00:37:10
I say don't try and boil the ocean.
00:37:12
Pick a couple of discrete KPIs that every company has.
00:37:16
The thing that they think is this is the thing that we are
00:37:19
really good at.
00:37:23
Speaker 1: Where would?
00:37:23
Speaker 2: be a good place to start.
00:37:25
Look at reducing days on market or days to if you are turning a
00:37:30
unit.
00:37:30
If you can turn a unit in two days less than you have been
00:37:38
Huge impact on a property.
00:37:39
So what we do is we set up, you track your goal, you put in
00:37:49
alerts, you have a start date and an end date and you look at
00:37:51
what is the overall impact and that's something that's truly
00:37:54
measurable.
00:37:54
If you're dropping two days from the days that it takes to turn
00:37:59
a unit, that reduces your days on market overall, adding two
00:38:05
more days of rent to each unit that you're turning.
00:38:08
That is something that you can show adds to your NOI and adds
00:38:14
to the value of your asset.
00:38:15
Same thing with looking at reducing an expense, your
00:38:19
marketing expense, if you're able to drop it.
00:38:23
Or I look at customer acquisition, your lifetime value
00:38:27
, looking at that as a ratio across your portfolio at each
00:38:31
property and setting a target for I'm going to reduce my cost
00:38:37
to acquire residents but I'm going to increase the lifetime
00:38:41
value of that resident.
00:38:43
Those are things that you can really show to investors who are
00:38:47
being really thoughtful, not just to transactionally turn
00:38:51
these tenants but actually to increase the value of the
00:38:56
housing experience for them.
00:38:58
Renewal rates, increasing those that's a huge way to increase
00:39:06
your overall.
00:39:07
If you're looking at your trade-outs, these types of
00:39:10
things, sitting down and picking , just start with three, start
00:39:14
with five, but looking at, if I did this and then was able to
00:39:19
accomplish it, what does that do to my property and portfolio
00:39:23
and how do I communicate that to my key stakeholders, whether
00:39:27
it's a property management company wanting to tell their
00:39:30
ownership groups we've added $213 to the property value
00:39:36
this quarter by doing X, y and Z and we help to automate the
00:39:41
process of communicating that, of showing that value.
00:39:44
That has tremendous ROI.
00:39:48
Speaker 1: when you look at things that way, yeah, and do
00:39:51
you find that people want to do something like that?
00:39:54
Look at those.
00:39:55
They have tools that aim to deliver some element of that,
00:40:00
but it's in that environment, right?
00:40:01
Especially if you're in multiple systems.
00:40:03
That's where it becomes tricky to understand that stuff.
00:40:06
And do you find that when you get these small wins under their
00:40:11
ability to measure something that truly makes a financial
00:40:15
difference in the business that they lean into and get more
00:40:18
motivated into investing more and more into the data strategy?
00:40:22
Speaker 2: Oh, absolutely, it's almost.
00:40:24
You want to gamify it in a lot of ways Get your managers
00:40:28
involved or your analysts involved in creating these goals
00:40:34
and tracking to them, because it's not one property, it's the
00:40:39
entire portfolio.
00:40:40
So, each person doing their part and making these small
00:40:49
incremental changes, they can have a huge overall impact to
00:40:50
the value of the portfolio just by being more thoughtful about
00:40:55
how you look at your data and how you track to performance in
00:41:01
small ways.
00:41:01
Overall, that adds.
00:41:06
It adds up.
00:41:07
It all adds up.
00:41:08
Speaker 1: Yeah, it sure does I always think about.
00:41:10
When you mentioned the slip days of time to turn, especially
00:41:14
in renting.
00:41:15
It's like empty airline seats that we just can never recapture
00:41:19
.
00:41:20
And that's that alpha, as you mentioned, that you can offer a
00:41:23
definitive or differentiating value in running your business
00:41:26
through this certain management company.
00:41:28
Listen, this has been great, Elizabeth.
00:41:30
I really appreciate you coming on.
00:41:31
I know that I would love to have you back for more on this.
00:41:34
For those of you listening, if you want to know more about AI
00:41:37
and data for better apartment investing, reach out to us and
00:41:41
let us know what are your questions that you have around
00:41:43
this, specifically to your portfolio, to your use case, and
00:41:46
just go to multifamilyaipodcastcom.
00:41:49
You can click there and you can even send us a message, but
00:41:52
reach out to us, let us know and go find that information.
00:41:55
Probably Elizabeth has it top of mind, but would love to visit
00:41:59
back with you, Elizabeth, as we continue down this journey of
00:42:02
extracting the knowledge ETL extracting the knowledge,
00:42:06
transforming it in a way that non-technical business leaders
00:42:10
understand it and then loading it into their business.
00:42:12
That was a pun on the ETL.
00:42:13
Speaker 2: I love it.
00:42:13
No, that's so great.
00:42:15
Speaker 1: Yeah, look, you're doing great.
00:42:17
We love that you're making an impact in this industry and we'd
00:42:21
love to talk to you more about this.
00:42:22
But until then, if you want to learn more about Elizabeth and
00:42:25
some of the projects she's working on, go to Multifamily AI
00:42:28
Podcast.
00:42:29
In the show notes you can click links and get to her LinkedIn
00:42:32
and all the other great places to connect with her.
00:42:34
Until then, wishing you guys the best and we'll see you on
00:42:37
the