The Collect Phase in Commercial Real Estate: Building Value Through Smart Data Collection

Introduction to the Collect Phase in CRE

Drew: Welcome back to Peak Property Performance Podcast. I am Drew Hall and we've got Bill Douglas with us here.

Bill: Hey, everybody. Drew, how you been?

Drew: Not too bad. Not too bad. Thanks. It's nice outside. The birds are singing.

Bill: Hey, it doesn't matter. Nobody's listening.

Drew: Dude. Well, I used to go, oh, he says that all the time and one day I was like, oh, you're right. Nobody's listening. So nobody wants to hear complaining. So that's what I was going to say.

Bill: Yeah, it's like an implied. I'm asking you, but I only want one side of the answer, please. I mean, it was my dad's way of saying, shut up, son. Skip the formalities.

Understanding the Importance of Data Collection

Drew: Yeah, that's awesome. All right. Well, so let's see, in our series right now, we're going through the five C's and we've touched so far on the first to clarify and connect. And so we're cruising on through to the final three, which will be today. We'll talk about collect and then later we'll go into coordinate and control on future episodes. So yeah, today is collect.

Bill: So in the spirit of, you know, starting with kind of a thematic sentence around the C of collect, let's think of it this way. You want to capture the right data from the right places so that you can stop guessing and start optimizing again. That's pretty loaded. There's a lot going on there.

Drew: There's a lot going on. Yeah. Let's talk in and around all that. I'm going to throw some extra spin on that and say, collect, you have to collect the right people from your organization to help you make this happen. Like you have the right people or maybe you need to add one, maybe you don't, but it's going to take more than just you.

Bill: Of course, you are going to spearhead or you're going to be an influencer, no matter what your position is in the organization, that is the commercial real estate owner, but it takes a team. So by collect, we're talking about the team and the data. But yeah, so I got my team out there for the purposes of this call. Let's talk more specifically about the data.

Connecting Physical and Virtual Systems

Drew: We talked about how easy and how difficult it can be to connect the physical and virtual systems. Talk a little bit about actually collecting the data. What does that mean, Drew?

Bill: Yeah. Yeah. Well, just as a quick reminder too, we've come through the process of clarifying and connecting. So now we're thinking about looking at a system, a particular system, maybe it's some building automation system or something that, oh my gosh, look at this, all of a sudden I've discovered this thing. Like I know it exists. I know that it's connectable. It is now connected. It is now participating actively in some kind of flow of data, like we talked about in the last episode.

Drew: So, hmm, let's get in here and find out what is this data? Like what can we possibly do with this data? Is it all something that's leverageable for the collective good, right, of the campus and maybe even further, but just for the campus, let's say, because not always is every piece of data going to reap masterful benefits, right? But there's going to be some key, some critical data that will, that absolutely will.

Bill: So, you know, depending on what that system is, it's going to, at least at the beginning, there's going to be some obvious data points that are very, very helpful and important in terms of knowing what you can do with that data. Like, oh my gosh, I didn't even realize we were measuring that. It never shows up on a dashboard. What we can do with it is actually the next C in coordinate.

Drew: So yeah, I like to encourage, I should say we, we like to encourage people to collect as much data as they can from a system because it's a lot easier to weed it out than it is to go back and ask for more.

Bill: Oh, absolutely. Yeah. Without getting all of it, then somebody, maybe you're the person in this organization needs to figure out what is the data like? Is it structured, semi-structured, unstructured? And then before that, we as a collective team have to figure out where are we actually going to store this? Are we going to store it jointly with a vendor? Are we going to use a cloud provider? Are we going to use our own hardware?

Strategies for Data Storage and Management

Drew: If it's our own hardware, is there a business continuity plan in place because this data is so valuable, right? If it's a cloud provider, who's going to manage it? And then what do we plan on doing within the future? Because that'll teach us how to structure it. Machine learning in the future, when we get to other things, doesn't need to have fully structured data. It really doesn't.

Bill: And are we going to put this in different data silos and one data lake and databases? We're not going to dive into those weeds, but these are questions that we do go over in the book and we do give you guidance on. You can definitely dive in the weeds on that, whichever one of the team members needs to, it's in there. But collect as much as possible because it's easier to not use than it is to, oops, I wish I had that.

Drew: Yeah, absolutely. So, I mean, that's the big barrier that we've crossed now, I would say, is before this point, before the collect, it was only a, what if, like, ooh, what if we can actually get this data? Let's do it. And so now we're at that phase of, hmm, check this out. The data is here. The data is flowing. And you're right. Deciding where to put that data is critically important so that you can get at it and start to do the last two C's.

Bill: Yep. Well, it's a big shift from the traditional, we're integrated, because integrated has systems talking amongst themselves. In the collect model, they're still going to, if they're integrated, they're still going to be integrated. We advocate strongly that the facility or the entity that owns the property of their portfolio collect the data themselves. Let these systems be integrated, but they should be able to pump data to you, to your data repository.

Drew: I don't want to define that repository. I'm not a data scientist. We're going to call it a data lake most often, because that is the easiest, broadest definition. But we mean your repository. Again, whether somebody manages it for you, you manage it. It's a moot point. It's your data. There's something in the book, own your damn data. So it is your data, own it, but collect it here. Put it somewhere so that it becomes usable.

Bill: And we get a lot of people that say, well, how fast can I use it? The answer is, we don't really know. We don't know how much is the data, how deep is it, how broad is it? We typically say, you know, after you collect data from multiple systems for six months, then we can start running queries on it to find out what you can do. A lot of this is not an exact science. A lot of this is hypotheses and then, whoa, look at that. This is awesome.

Drew: Sometimes you run it and there's not there, not much there. So you go on your next hypothesis, but without the data collected in your repository, you can't make that next step. So focus on getting this data so that later on you can do coordinating. Having this data drives the asset value of your facility and your portfolio.

Enhancing Asset Value with Data Insights

Bill: When you trade a portfolio, it's proven. If you have a X year history, say it's five year history of data, and you could remove a lot of questions from the buyer's mind, you get a premium for it above just the standard cap rate from cashflow. You get a premium because the risk is lower. If you could show we have this usage and this, this many leaks and here's, here's this, here's our traffic patterns, here's et cetera, et cetera. And then we'd give you multiple ideas to talk about. You definitely get more money from that.

Drew: So it's your data just by packaging it correctly, including it when you trade the property, you will see a large premium return on that outside of the operating expense reductions. It's interesting because there's a lot of automation in this space for sure. Once you start to collect and as data starts flowing from this system, oh, now from that system, oh, from that system, there's definitely a lot of automation that can occur, obviously.

Real-World Examples of Data-Driven Decisions

Bill: It's interesting because it's something happened just this week with one of our commercial clients where like there's an anchor restaurant tenant on the ground floor corner of this facility. So mixed use, yeah.

Drew: Yeah, exactly. And specifically on the restaurant, there was an interruption, let's say, something on their network. And it impacted like their audio visual system, right? And so we'll just say that's one system, their AV system is one system.

Bill: Well, because of the connectedness of not just that system, but other systems in this campus and the fact that we were collecting on that data, we were able to correlate all the way back to like root cause in a way that wouldn't have been possible if those systems were even connected, but in isolation of one another.

Drew: So the side benefit was system resilience and uptime because it was redundant network monitoring.

Bill: Exactly. How does that statement?

Drew: Yeah. I mean, in this particular case, it was a failure that was like an operational failure. It had more to do with staffing and access controls kind of thing. So it's just interesting because it was the confluence of data from multiple systems that contributed to understanding why this even happened.

Bill: And it's just interesting because I find this to be kind of approachable in a way that sometimes this sometimes it can feel so ethereal and like, oh, what is all this data? What do you mean? It's floating where? But this is a real world example where it's not magic. Multiple systems connected in common with data flowing and we're collecting that data allowed us to be able to understand like, oh, check this out. At this exact time, this happened. Let's go back and and correlate that to this other system here and compare what's happened there. And the surveillance system was involved in that particular investigation and things like that.

Drew: So at the end of it all, in this case, it wasn't a new automation process that was implemented. Rather, it was.

Drew: A new set of standards for physical access for them in these spaces, because that's where the capability is. You have to collect data for six months to realize it. I think you just gave us a real-world example of how you got immediate, because this is a brand new building, immediate return on collecting the data. You were able to not only know it went down, but know why and save your tenants a big headache.

Bill: And the cool thing is- The restaurant brewery did not want to be offline and the problem was remedied right away.

Drew: Yeah. Yeah. I know the cause and we don't need to dive into it, but had they not been collecting data, who knows how long it would have been.

Bill: Yeah. And the great thing is- Music's off, but there was a lot more than that.

Drew: Oh yeah. It was a lot more than just music.

Automation, System Resilience, and Future Trends

Bill: Yeah. Well, and the great thing is like over time, as more data is pumped into the system and you can apply, let's say apply machine learning on that data, the possibilities increase greatly in terms of what you can do in response to those amalgamation of events, right? In this case, it was very early on, right soon after being connected, there were some manual decisions.

Drew: Let me ask you a clarifying question, because you said data was pumped into the system. Did you mean pumped out by a system into a data repository or did you mean to say pumped into a data lake, not into a system? I don't want to confuse people.

Bill: Systems generate the data, right? It wasn't being pumped into a system, it was being pumped into the client's data lake.

Drew: Well, yeah. I mean- Or a network monitoring system, like what was the system you were talking about?

Bill: Yeah. I would say that that's more specific. That's good. That's a good specification is like, this was more about a network monitoring system that was able to capture the data.

Drew: Okay. I wasn't aware it was. I thought you were trying to say it went into a data lake.

Bill: Not even yet. Yeah. It's so early in their implementation that they're still in this process that we're running through right here.

Drew: Yeah. Okay. Yeah. It's a brand new building and they're still in the Connect and Collect as the building comes live, as tenants move in. There's a connection every time.

Bill: Yeah. And so, I mean, the tee up on this topic for Collect included this concept of stop guessing. You can stop guessing because you actually have data now. And so, that's a perfect example from just this week. You don't have to guess. You know that at this date, at this exact time, this thing happened. And then you can correlate it through other systems to say, yeah. And in this case, there's even video footage of it happening. You know? Because that's a whole nother system generating data.

Drew: I think we talked about video security data in the last episode.

Bill: Yeah. Absolutely. The thing I see here that is great for the industry is 10 years ago, not only was there less data being generated, but there were fewer systems connected and less data being collected. So, think about how different that would have been. It was just like, it was the Wild West when it came to digital and networks 10 years ago or 10 months ago, for that matter, and a lot of properties, right? So, we've been on this trend for, what, seven years now? And the book just came out, but it's full of history and real stories. So, I'm really, really optimistic about how much better this all will be for commercial real estate.

Drew: Do we have some pain coming? I don't know. Pain. Change. So, change for some is painful, but I think it's all upside. I think that it looks better and better and better.

Bill: It's one of the laggards in the world relative to technology, but it is, it being the commercial real estate industry, still the largest asset class in the world. It's a huge industry. It's ripe for improvement. So, we love talking to people that want to help us, you know, rising tide raises all boats. We want people to embrace data and digital.

Conclusion and Next Steps in the 5C Process

Drew: Well, and it's interesting here. I mean, you know, we're trying to focus in specifically on collect, but we've sort of given a peek into that next step of coordinating. In our example here today, we've given a little bit of example of coordinating, but we'll dig more into that next time.

Bill: But I mean, I guess I was going to say collect is sort of a means to an end. You could kind of say that about all of this, but it certainly feels like it here. Like, oh man, I feel the tides building here, right? Oh, we're collecting now. Here's the data.

Drew: Well, I remember a year and a half ago, we did that big market study and almost across the board, commercial real estate owners told us they wanted business intelligence. And when we dug into what that was, they wanted the ability to make data-driven decisions.

Bill: Well, problem is there's no data. There is data, but it's largely around leases. What about data from operating systems? So that is where we evolved. That's where the book came from. Then we realized we've been doing this for a long time. We've only honed it in on certain numbers of systems and certain verticals inside of commercial real estate. And we are broadening that with this discussion and we're glad you've joined us for this. Hope you're in for a fun ride with us.

Drew: Yeah, absolutely. Well, that's probably a great jumping off point right there. Thanks for joining us today for all about the drill down into this collect phase of the 5C process and join us next time. We'll dig into what it means to coordinate. So until then, thanks for joining us, everyone. We'll see you next time.

Bill: Yeah. And as always, don't forget to like and subscribe and tell your friends to hop on in. Thanks all.

Drew: Thank you.