🎧 #160: Case Study: KIPP DC Schools wants autonomy, deploys IDL, FDD, Supervisory control, CMMS

"Most School systems are not financially efficient. Stereotypically, the facilities are outdated, rundown, and poorly managed.  We’re looking for ways to do that better and I think the end goal being, how do we get the money that is saved there back into the classroom, and back where it is needed most?” 

—Nathan Morris

Welcome to Nexus, a newsletter and podcast for smart people applying smart building technology—hosted by James Dice. If you’re new to Nexus, you might want to start here.

The Nexus podcast (Apple | Spotify | YouTube | Other apps) is our chance to explore and learn with the brightest in our industry—together. The project is directly funded by listeners like you who have joined the Nexus Pro membership community.

You can join Nexus Pro to get a weekly-ish deep dive, access to the Nexus Vendor Landscape, and invites to exclusive events with a community of smart buildings nerds.

Sign-up (or refer a friend!) to the Nexus Newsletter.

Learn more about the Smart Building Strategist Course and the Nexus Courses Platform.

Check out the Nexus Labs Marketplace.

Learn more about Nexus Partnership Opportunities.

Episode 160 is a conversation with Nathan Morris from KIPP, Stephen Dawson from Normal Software, and Terry Herr from Intellimation.

Summary

Episode 160 features Nathan Morris from KIPP, Stephen Dawson from Normal Software, and Terry Herr from Intellimation and is our 8th episode in the Case Study series looking at real-life, large-scale deployments of smart building technologies. These are not marketing fluff stories, these are lessons from leaders that others can put into use in their smart buildings programs. This conversation explores KIPPs partnerships between Normal Software and Intellimation to better manage school facilities and increase savings to bring money back into the classrooms. Enjoy!

Mentions and Links

1. KIPP (1:34)
2. Intellimation (3:03)
3. Normal Software (4:44)
4. Comfy (5:32)
5. Niagara (36:12)

You can find Nathan, Stephen, and Terry on LinkedIn.

Highlights

Overview (1:14)

Introduction to Nate (1:40)

Vendor team (2:43)

Project results (3:31)

Introduction to Terry (4:10)

Introduction to Stephen (4:40)

Project goals (5:56)

Previous technology stack (7:00)

Impact of new technologies (9:32)

Solution options (10:30)

Why start Normal Software (17:00)

Phases of deployment (20:44) 

Thinking about IDLs differently (29:48)

How is the tech stack getting used (40:29)

Lessons learned (47:45)

‍

‍

‍

Music credits: There Is A Reality by Common Tiger—licensed under an Music Vine Limited Pro Standard License ID: S528430-16073.

Full transcript

Note: transcript was created using an imperfect machine learning tool and lightly edited by a human (so you can get the gist). Please forgive errors!

‍

Nathan Morris: [00:00:00] Most school systems are not financially efficient, at least as it relates to the facilities. Uh, you know, stereotypically, they're outdated, rundown and poorly managed. So working on how to find ways to do that better. And I think the end goal being, how do we Get the money that is saved there back into the classroom and back where it is needed most.

Reallocate it in an equitable way.

James Dice: Hey friends, if you like the Nexus podcast, the best way to continue the learning is to join our community. There are three ways to do that. First, you can join the Nexus Pro Membership. It's our global community of smart booing professionals. We have monthly events, paywall deep dive content, and a private chat room, and it's just 35 a month.

Second, you can upgrade from the ProMembership to our Courses offering. It's headlined by our flagship course, the Smart Building Strategist. And we're building a catalog of courses taught by world leading experts on each topic under [00:01:00] the smart buildings umbrella. Third and finally, our marketplace is how we connect leading vendors with buyers looking for their solutions.

The links are below in the show notes. And now let's go on to the 

podcast. 

Welcome to the Nexus podcast. This is the latest episode in our series, diving into case studies of real life, large scale deployments of smart building technologies. And we emphasize real life here because we're not here to create a marketing story.

We're not here to create fluff. We're here to share lessons from leaders that others can put to use in their smart buildings programs. So today we have a story from KIPP public charter schools. And the voice you just heard was Nate Morris, Director of Facilities at KIPP DC. Uh, Nate, can you tell us a little bit about your background?

Nathan Morris: Yeah, absolutely. Uh, so I, uh, once upon a time was a teacher, uh, working in the classroom, um, anywhere from kindergarten all the way through middle school, uh, for a few years. And, [00:02:00] uh, from there I Uh, started working for DC Public Schools on their facilities, uh, team, uh, doing large scale modernization projects, um, and very large scale, uh, capital improvements on schools.

Uh, I've been with KIPP for, uh, this is my fourth year. KIPP DC operates eight campuses, uh, In Washington, D. C., we have just under a million square feet of real estate, uh, as part of our portfolio, 20 schools, uh, just over, somewhere over 7, 000 students now. We, uh, we, uh, have students from pre K all the way through 12th grade.

James Dice: Can you talk about your, your vendor team first? 

Nathan Morris: We have a third party facility management vendor. Uh, that staff has full time staff in each of our, uh, buildings at each campus. Uh, and then we also, we have a, a variety of different vendors. So just speaking on the ones that I think [00:03:00] are most relevant, uh, uh, we have Intellimation and their team, uh, helping us on HVAC controls and a variety of different, um, sort of building systems and software tools.

And then our mechanical vendor. Uh, who also handles preventive maintenance, as well as any sort of, uh, repairs or Modifications as it relates to HVAC. Uh, and how many square feet? Just under a million. 

James Dice: When did this project start? And what 

have been the total 

results since then for the district? 

Nathan Morris: Visibility, data, uh, to probably oversimplify the answer.

But having, uh, insights that we were previously lacking, knowing what is working and what is not and where our attention needs to be. Um, in addition to that, I would say we have made some, some serious steps forward on energy efficiency as well. 

James Dice: Awesome. All right. So let's bring in part of your team here.

[00:04:00] Um, let's start with Terry. So Terry, you've been on the, on the, the Nexus podcast before, way back in, Uh, episode nine, I believe. Uh, can you, can you reintroduce yourself 

to everyone? 

Terry Herr: Sure. Intellimation is a, is a, um, I would call us an MSI now. I've, I've been in the building automation space my entire career.

So, I don't know, it spans about 30 some years now. In 2014, we made a major transition, I'd say, from an MSI. Focusing on primarily what we call data driven retro commissioning and optimization. And we're still at it. 

James Dice: Awesome. Uh, Steve. Go ahead, Steve. Can you introduce yourself? 

Stephen Dawson: Absolutely. So, I'm Steve. I'm the founder of Normal Software.

And, you know, I think a little like James actually, I founded this company because we would see all these great ideas floating around about smart buildings and software and data and all that. And then you'd sort of go out there and Wonder where it all [00:05:00] was when you actually started, you know, looking at buildings in particular, and we found it normal to really try to accelerate the deployment of some of these advanced controls, advanced optimization, advanced, uh, You know, even, even not so advanced, just having the data and that's, that's our mission at normal is to make it simple and easy to connect as many buildings as we can to the applications that are going to drive those kinds of outcomes that, that Nathan was talking about, uh, before normal, I was a co founder of.

Company called Comfy, where we had an early sort of occupant control app. And then was at Siemens for a couple of years, learning how, how the big guys do it. 

James Dice: I feel like I could have a whole, whole conversation about that topic, but we'll save that for later. Save that for the AHR Expo here. Um, so Nate, let's talk about the story of this, this project to give people a little bit more depth and background here.

So why did you sort of start down this path and what were your goals? [00:06:00] 

Nathan Morris: When I started at KIP DC, coming over from DC Public Schools, um, you know, KIP DC was about entering its 20th year, uh, and it had been in buildings that were primarily, uh, or newly renovated, and had been in a developer mindset for, you know, there in the entirety of the organization, and stuff was starting to get old.

And was going to start failing soon. So I was sort of tasked with moving KIP from the developer mindset into an asset management mindset. And so this project I think was a big part of finding out how do we take care of our assets and how do we find out what we have, what kind of shape they're in and how we can most effectively and efficiently monitor them going forward.

James Dice: So Nate, when you think about your technology staff stack, what was it like before this? Um, it sounds like you came in and [00:07:00] got tasked with this asset management role and they were like here. Um, and just based on what I, what I know about our industry, eight different campuses, eight different buildings, it's probably a.

A hodgepodge of different types of systems. 

Nathan Morris: We, I think we actually, uh, have more, uh, building automation systems than we do campuses. We have one campus that actually has two operating within the space because a lot of our buildings are, uh, as we like to call them, Franken buildings, where we have just continued to tack on and add square footage as needed to allow us to expand, uh, which is nice in theory, um, but has created a real challenge from a management and oversight perspective.

Um, in terms of other technology stack, I'm kind of laughing because I don't know that we really had much technology other than the stuff that we were required to have for our buildings to actually function. Um, you know, we had a pretty arcane work order platform that was being used, uh, that [00:08:00] did not actually spit out usable data in any form or fashion.

Very qualitative, uh, and really, really lacking on the quantitative sense. And then a, uh, sort of a. A bevy of spreadsheets, nice spreadsheets, but at the end of the day, still a spreadsheet. Um, so that's where we were. 

James Dice: Probably a few clipboards that you haven't mentioned. 

Nathan Morris: Yeah. And not to mention stacks of O& M manuals, uh, buried in basements and hopefully having not gotten wet already.

We did have, uh, I should give credit where credit is due. And I think Terry's aware of this too, but we had, uh, the software, the software tools that did exist were, um, not necessarily a great fit. They were also very expensive. Um, and only half stood up and not being used by anyone because there was no real life application or need for them.

But, uh, you know, we kind of got sold on an idea. Somebody had a, a nice presentation and, [00:09:00] um, it never actually produced much. And that's not necessarily a knock on the tool, but I think a great illustration of the tool's only as good as, as you know, the users that are. going to be relying on it. 

James Dice: Absolutely.

And if you had to sort of summarize the effect that this was having on your, your task, you were tasked with, you know, turning this into a sort of asset management focused organization. Um, certainly on the O and M side, what would, what was the impact of this sort of disparate tech, all these disparate technology tools, disparate control systems, lack of technology.

Can you just summarize the sort of impact on what you were trying to accomplish? 

Nathan Morris: Yeah, I would say the impact was, you know, The landscape was incredibly subjective. Uh, it was hard to know what, what actually worked well, where, because all we had were anecdotes and again, qualitative data from people, uh, who, I would say school staff members.

And school leadership [00:10:00] teams often have strong feelings about a lot of things. Um, and which makes them successful at their job, but also makes it really challenging when you're trying to face down something that has some mathematical qualities to it, like facility management. Uh, so really trying to put our finger on where to start and where the need was greatest, where we were going to have, uh, the most important impact right away.

James Dice: What were all the options you were considering for solving this problem? And then when did you sort of realize that you needed to engage Intellimation to sort of help you walk through this journey? 

Nathan Morris: Yeah, that's a good question. Uh, You know, there is no shortage of firms trying to sell you stuff. Uh, there are plenty of people who will tell you they have the solution.

Uh, they know how to fix what you're dealing with. They can make your life so much easier. Um, and I think it was actually just that, that was sort of helping to steer us back to Intellimation. Um, I had [00:11:00] worked with Terry and his team. Uh, in my previous role with the district, um, and. I, I, I knew what he and his team stood for and how they operated and just the, uh, sincerity and authenticity.

And frankly, you know, I, I trusted them, uh, to look out for our best interests. Um. And I, it was a matter of just getting the other folks on my team to see that and understand that, um, working with Terry and his team would benefit us down the road and help us find the tools that were, uh, the right fit for what we were ultimately trying to achieve.

James Dice: Totally. And Terry, can you walk us through, so, so Nate and the district engage you, can you walk us through like your, the roadmap that you had in your, in your mind for them and what sorts of technologies you felt like they needed at that time? 

Terry Herr: Like many owners with multiple buildings, they had a, [00:12:00] um, they had, I don't know, a number of different building automation systems.

I don't know the exact number, but as Nate said, there was, there's one building, I believe it has three in it, different ones, but they, but they had a mix. And they were all siloed, isolated. So we're, you know, generally our approach is similar. We had, we had spent some time, as Nate said, on, uh, with, with, with DGS over quite a number of years building and figuring out the right sort of application stack.

So it always starts with an independent data layer, a middleware layer, um, to bring all the data together, to trend all of the data in a, in a neutral, you know, non BAS fashion. Um, and, and it always includes a, a, you know, central BAS, you know, a typical BAS front end, but a single one instead of five or six different ones, which is what they had.

We did go through a process of. [00:13:00] Evaluating three products in that, in that space. We did that really for this project, but we did it for ourselves as well. We're always trying to stay abreast of what's, you know, the best in class product. Um, so BAS is, you know, they, they needed one. They had five and we wanted just one.

So we had a single view. So, um, and then of course the independent data layer, um, is part of that. And then, you know, one of the good things about CHIP is that they were, um, willing to sort of build out what I would call the complete digital facility management application layer, and we're, I think we're not even quite done there.

We, we've got a, we've got a, we've got a BAS, we have, um, we're, we're, we just went through a process where we're evaluating three different FDD tools. For, for the application, you know, part of the IDL course we did that, that was normal framework and we did do the data model and we actually spun that up.

In [00:14:00] this case, it's, it's in, it's in AWS, but it's in their AWS. So they own, they own the data. They own the infrastructure. They own the data model. Um, there's a BAS that's been deployed. We read it all the graphics for all the sites. So it's a single, uh, Um, VIEW, SINGLE, um, WORKFLOW. And again, we're in the process of deploying FDD.

We did, we did a pilot on one tool, and then we decided that wasn't the right tool, and we're, we wanted to, uh, evaluate two others. Um, and I think we're about to pull the trigger, um, or make a recommendation on, on that. Um, they've been, we've been spending some time on a CMMS, because they're, they're gonna choose, and I think we started out with, I don't know.

I mean, that process has been going on for, for a while. There's a lot of applications in that arena as, as well. Um, I've looked at them years ago, but anyhow, that'll, that'll be included. We added, um, [00:15:00] Grafana as a, as a dashboarding tool because we really like Grafana. We think it's, it's, you know, well, first of all, it's open source, it's free, and it's really powerful.

We've not seen a BAS come close to the types of dashboards you can build with it. So we're, we're a big fan of that. Um, so that's included. A time series database, again, an open source one. So we've used As much open source tools as we could. Um, you can't get it all done with open source. So I think that's the basic stack right now.

James Dice: And Terry. So I want to bring Steve into this conversation a little bit back when you and I talked four years ago, normal didn't exist. You've been sort of keeping up with this IDL space since then, right? I'm sure you've tested out a bunch of different. different vendors and to sort of plug that hole in the stack.

And, um, can you talk about what, what it was like to, you know, you know, I know you use Steve back in the building robotics days, but Steve starts normal. And so why'd you [00:16:00] pull them into this, this, um, project? 

Terry Herr: And that was probably what Steve, two years ago, you felt like it was, um, a good time to actually build what you might call a trender 2.

0, you know, new and improved, um, and. So we started working together and testing that, um, Steve's a hell of a, um, software developer, so it went quick, it went well, went fast. Um, we saw the, uh, you know, what was possible. And so we have been, um, we've been using normal, well, we've been using normal since, uh, since, since the start.

And we think it's come a long way and we're pretty pleased with. 

James Dice: So Steve, yeah, a lot of people haven't heard of what you guys are doing. This is kind of a little bit of a coming out party, I'd say this, this podcast. And, um, yeah. So can you talk about why you decided this, this is your next chapter after, you know, Lehman's selling comfy, leaving Siemens, um, talk about that, that opportunity that you saw a [00:17:00] little bit.

Stephen Dawson: Yeah, absolutely. I mean, I guess it's kind of the personal side and the technical, you know, and business side. You know, personally, so for those who may, might not be familiar with comfy, what we started with, there was basically an occupant control app. So you could press these little buttons on your phone and say, warm my space or cool my space.

And, you know, we had implement that with backnet gateways that we built. Um, but I thought that was pretty cool. You know, that's kind of a new experience you're enabling in, uh, in these commercial buildings where, you know, traditionally we would show these pictures in our sales deck of like the lock thermostat or whatever.

Um, and we were sort of unlocking that. Uh, and you can get energy savings by, you know, having extra data about what people wanted. Um, I would, you know, the part of Comfy I really liked, though, was that controls part. I thought the, you know, getting in there and really changing the way the equipment was programmed and the way that it operated was You know, both sort of rewarding and technically interesting and, you know, had the most possibility for impact.

Um, comfy, I would [00:18:00] say, you know, due to what our customers were asking us turned, you know, kind of evolved over the years where that was still a piece of it, but it also did a lot of. Other occupant tasks, like, you know, how do I order lunch in this conference room that, um, you know, have their, have their uses for sure.

But I was kind of less passionate about that. So, you know, when I left Siemens, I knew I wanted to get back into the control space and like I kind of alluded to, I think there's a lot of great ideas floating around, like. Uh, you know, some of these ontologies, lots of advanced control strategies, uh, different ways of programming buildings.

But I just felt then, and I still feel to some extent, there's a lack of, you know, great software doing it, basically. And, like, there's kind of a gap between, like, the best practice that people talk about, and as, oh, we know, we know how to do that, and then what you can actually, like, buy and actually scale.

Um, and I think that that's real, you know, if you read the journals, it's, it's really there. And so, [00:19:00] um, yeah, that's what normal exists to do, at least for our little slice of the industry. It's to take this independent data layer and, you know, figure out, you know, the capabilities to really make it work technically, and also to make it work, you know, as a business and make it work from a deployment point of view of how do you make something that does that.

You know, I think it was interesting listening to Terry and Nathan's conversation, you know, how much. Terry is serving that sort of consultative role there. And that's absolutely needed. There has to be some, somebody like that in the industry. I know that's something you've highlighted a lot, James, because it's.

baffling and most people have day jobs who, you know, aren't going to go like Terry and, you know, demand demos for 50 different FDD products. And therefore I think it leads to, um, you know, just lack of deployment of tools that would actually be beneficial, but, you know, there's a big cost of, uh, for not doing that.

So that, that was our starting point. And we think that, you know, by building a great. Solution [00:20:00] really for people like Terry and Intelmation. I think that you're going to unlock a lot more deployment to the software because if nothing else, it should be easier to try out those 50 different FDD products so that instead of doing like some crazy six months and six months integration with whichever, you know, after a brutal RFP, um, you could just kind of do a bake off and the cost should be low enough that, you know, you shouldn't have to worry so much about the planning upfront.

I think that you can definitely achieve that. 

James Dice: Absolutely. All right. That's great context for sort of how this sort of stack came together. So you, Nate, you guys brought Terry in, Terry pulled Steve in. That's how we're all having this conversation. So Nate, can you talk about sort of the main phases of deployment?

I think Terry mentioned earlier that you're about 80 percent done at this point. So can you talk about like, you know, what are the phases of this? And then sort of what's, what's next to give people like, uh, uh, a certain current status on where this is at. 

Nathan Morris: Yeah, sure. Um, so I think the [00:21:00] primary, like, first phase was just seeing, uh, sort of determining working backwards from what we're hoping to achieve.

Like, what sort of things, uh, did we want to get out of it? Um, and what was going to help us run our campuses and our network more efficiently from the facility perspective? Uh, and Terry, please jump in here. If I skip any steps, I know that, uh, You may be aware of more than I was on some of these, but, um, yeah, I think then just sort of evaluating the different tools and seeing what was there, something that's coming to mind and something I know I was a broken record about early on with Terry and probably still am to some degree.

Uh, is just the importance for us, uh, to have autonomy with whatever tools, uh, we do select, you know, we are do not want to end up in a place. I, one thing I failed to mention earlier is that the arcane work order platform we do have that has mostly qualitative data actually, uh, is not licensed to [00:22:00] us. It is with our facility management vendor.

So. Should the day come that they take off, the data, in theory, could just go with them. Um, depends on if they want to share it with us or not, and if it's worth keeping. Um, but, you know, we like to build tools where the data is worth keeping, and then we want to be able to keep all of that. Um, and still have the freedom to switch tools, uh, as well as switching vendors, you know, building stuff that can live in house, so I think, um, that was a big focus for us in that evaluation of the different tools and, and options that were out there.

Um, and then Terry, can I, uh, maybe I'll pass to you to talk about just like the deployment and, and sort of how you guys work through it from 

there. 

Terry Herr: We almost always start with trending all the data as fast as we can, um, immediately. So just because the data, the trend data has a ton of value and typical building automation system maybe is trending 10 percent on average.

So, um, so we, we deployed. [00:23:00] Normal and started trending data. Probably the second piece of that was we started integrating into, uh, one BAS front end and redoing all, all the graphics. That was probably the second stage. Um, maybe simultaneous there afterwards. Um, we started building the data models, you know, that data modeling is a, is a challenging task.

Building graphics is a challenging task. Um, and, you know, I think the graphics are always there, it's the foundation of a BAS. Um, I do think we probably overemphasize them, but everyone's used to the fancy graphics and so we, we've got to build that. We, we really like trending all the data and making some good use of that.

So, so that part was, um, was, you know, the The trending of the data, the graphics, then the data model, um, and then the optimization. So we always include optimization as part of our [00:24:00] process. Um, so we started once we have enough data to look at how things are operating and what's, uh, what's working, what's not working, what's, what's, um, faulted.

We, we, we, we come up with this list of, okay, these are the things that are broken. You need to fix. And here's the opportunities for optimization. Um, and then we start deploying optimization. The, and, and, uh, this one, this is an interesting project in that, um, although FDD tools there are plenty of, optimization tools there are much fewer of.

And we've learned over the years that the optimization is probably worth perhaps more than the FDD is. Um, certainly there's rebates for it, there's energy savings. FDD is a great tool. In fact, we think now we kind of lead with optimization and follow with FDD. So, um, but at KIPP, um, you know, [00:25:00] and they were willing to do some, some testing or piloting.

And we've actually currently deployed or testing three different optimization solutions. And Steve has recently added a, uh, a great feature to normal. Um, which is an SDK to program in JavaScript. So, essentially, um, you know, an engine that we can program optimization guideline 36 optimizations right into the normal framework.

So, we're doing that at, at one of the schools. Because that, that space, optimization space, we, we think is fairly young compared to the FDD space. So, um, you know, it, it's, it's There's, there's evolving products in this space. Um, but anyhow, so, so optimization, you know, is, is what's next. Um, and I would say the CIA MMS might be one of the final pieces [00:26:00] of it.

Um, where we'll have a suite of applications running all, all against a neutral, you know, owner owned data, data model or data model and, and IBL layer. 

James Dice: Nate, can you talk about the, so you, you've brought Terry in this role, he called himself an MSI, sort of a trust advisor, trusted advisor to sort of put together this solution for you.

That's great, but I really want to hear from you around, um, the certain certain design decisions he's made and why it's important for you as you develop this stack. And one of them is this role of the IDL or this this layer that Steve and Normal are providing. What is it sort of, why is it valuable to you as the the end user here?

Nathan Morris: Yeah, I think it was actually Mirrored quite nicely in some of the other pursuits that we were making where going back to my previous point about having autonomy and building our, our [00:27:00] systems and, you know, just the structure of, of our team and, and, uh, tools and processes in a way that would allow us to plug and play to the greatest extent possible when the need comes, you know, um, anecdotally, we had a, uh, a vendor Uh, that was deeply trusted, had worked with the organization for several years.

Um, and tragically the owner of the, the organization passed away, um, and it was taken over and, you know, the service went south and, uh, really a sad story. But at the end of the day, you know, we had to move on and find someone new. So just being able to have that as somewhat of an insurance policy, but also just, um, being able to, you know.

Uh, be nimble on, uh, on that front. So, you know, working with Terry and, you know, what made, I think what allowed and fostered the, the deep level of trust that we have in him and his team [00:28:00] is also knowing that. He's not, uh, selling anything, you know, he's not selling a specific tool to us. He's not getting payouts based on what we choose, uh, to my knowledge, just kidding.

Uh, but yeah, I think so understanding is talking with Terry and understanding the independent data layer that would allow us to, like Steve was saying, um, have a bake off and try a number of different solutions. and be able to see that in action without, you know, the agonizing setup part to where, uh, exhaustion sets in and you are just need to pick something and you are just tired and you pick a subpar, um, or mediocre product because you're tired of looking or somebody wants to make a decision.

I think it's also worth pointing out something else that Popped up, um, as Terry was speaking in terms of optimization pre FDD, you know, optimization is how we were able to, uh, get our organization on board, you know, with a [00:29:00] combination of cost savings from energy, um, reductions, energy use reduction, but as well as, you know, uh, building energy performance standards requirements in DC, the BEPS, um, um, law that's come into place.

You know, it is a very real looming deadline where fines will be, uh, levied towards, um, commercial building operators who are not meeting, uh, energy efficiency guidelines. And probably comes as no surprise that DC is on the, on the cutting edge of that, uh, sort of legislation. 

James Dice: Absolutely. Yeah. So you don't really have time to pick one of these, like Steve said, and have it go six months and then realize it's not the right application.

Um, it makes a lot of sense. Steve, can you talk about just like circling back on what Nate just said about the IDL? Talk about sort of how you're thinking about the IDL differently because you're not the first, normal is not the first, um, IDL on the market. Yeah. But it's the sort of the latest one. So, like, how are you thinking about what normal brings to that conversation?

Stephen Dawson: And there's a couple of ways to come [00:30:00] around it. And, you know, I think it is worth kind of double clicking on that independence idea because people have different ideas of, you know, what you ought to be independent from, I think, in the more technical community, people want to be. Independent from your BAS vendor.

That's usually the way it comes up, right? It's like, oh, I've got Siemens, I've got Johnson, I've got Jace's, you know, I, I want to be independent. I don't want to be independent from that. So that when I have a portfolio like Nathan, you know, I've and Kip do that. It's, you know, you can do these, you know, consistent reporting across your portfolio.

And that's certainly an important dimension of, of the independence. Um, on the other hand, I think when you get into buying decisions, it's. And when you talk to owners, my take is that commercial independence, you know, independence of service and data portability and ownership are actually the kind of independence that, that they value, because, you know, at the end of the day, your average owner, isn't going to go [00:31:00] reprogram their BAS in any case, it doesn't matter who made it.

Right. And so they're not going to log in and start writing a PPCL or GCL plus, or, you know, you take your pick of. What, what the automation language is, um, they're going to rely on their vendor team. And so this, that having the ability to go, you know, switch out vendors and not be locked into their current provider is actually really valuable.

Um, and. So that's the kind, you know, it's, it's, and it's maybe more of a true independence from, from their point of view anyway. Uh, and I think that, you know, to some extent that really comes through in how we've thought about going to market with normal, which is to go to market first, actually with a non SaaS offering that is primarily delivered through integrators.

You know, like Terry was saying at the beginning where, you know, the product is deployed in AWS, actually normal has no access. Really beyond, beyond support, um, [00:32:00] and then really digging in on, you know, what, what does it mean to be a provider there that can try to achieve some of those outcomes for the owners of commercial independence?

And, you know, you very quickly come in and over, all right, the actual data integration, pulling back new points or pulling points from meters or whatever, you know, that's a component, but it's, it's not really enough. I think that that's not. You know, a novel product. Okay, so then you augment it with this data model, because now you start to be able to do, you know, some consistent reporting, um, On top of that.

And I think that that's absolutely required for their, for any sort of, you know, independent data model products that you can restructure the data, manage the data quality, that, that kind of thing. Um, but you know, I think you still are not quite, quite there actually, as far as what a true IDL product is, I think you need the final step, which is really to bring in programmability, you know, into that data layer so that you can, uh, well, there's, there's many reasons you need that, you [00:33:00] know, but it's, it's a.

Industry in a world of sort of 90 percent solutions where you can buy products off the shelf that do, you know, you try to buy one that does as much as what you want as, as you can find. Right. But there's always something funny going on everywhere. Every project is something funny, basically. And, you know, if you are locked into a model where either you have to say no, you know, to that project because you can't do that last 10 percent or, or it's, you know, exorbitantly expensive because, you know, you have to put it on the product roadmap and really productize it and then maintain it forever, right?

It's, um, you know, that's also not, not quite getting there. And so that, that's at least the final piece that I think, uh, is really needed for these. Independent data layers is, you know, they have to embed some amount of programmability and extensibility in there so that you can structure that, you know, that, that last 10 percent can drive 90 percent of the cost, I think, because, yeah, you know, you have to do custom development and then you have to [00:34:00] maintain it forever.

Right. And so it's very easy to run into a situation where, Oh, we've got this crazy protocol, or, you know, we have this IWMS that nobody's ever talked to or whatever, you know, that can just totally blow any sort of project out and so how do we structure that technically to keep it so that those. Easy integrations, uh, stay easy and can be maintained, you know, appropriately.

By the, by the appropriate person in the ecosystem, which I think is not the ideal vendor because there's too many, there's just too much funky stuff floating around. And so that, that for us is kind of the final, uh, piece of the product, which is this SDK to make it really easy to do those one off developments and then wrap it, wrap up, wrap them in the.

What you need to actually deploy them reliably and repeatedly, uh, so that that custom development is as small as it can be, basically, but still still meet the need 

Terry Herr: the, um, just a quick comment, the, um, one of the things, James, that I'm frustrated with in our, in our, in the building [00:35:00] automation industry is that every vendor feels like they have to invent a programming language for their product.

So we've got. Yeah. Um, I mean, every, every piece of BAS hardware programs with a proprietary programming tool. Um, so the fact that normal can program in JavaScript, which is, you know, open source, um, programming environment that, that. You know, college students learn and not PPCL or, um, or CCT or whatever, you know, um, proprietary programming tool that's, that's, that's required.

I think that's a major step. It's things that we look for, like we kind of want a best I think that's Best of breed, um, but not, um, you know, that, that doesn't, that doesn't have a vendor lock or require some, you know, uh, annual SAS fees or licensing of the programming tool every year. I won't mention that big product that requires that, which is a pain in the ass.

Um, so no, no [00:36:00] licensing JavaScript. 

James Dice: Can I, can I mention it? I mean, the first thing I think about here is you guys are talking about, um, a lot of integrators are using a product called Niagara. Can you guys just talk about the differences between what just sort of make it a little bit more black and white.

And then I want to get back to Nate and the users of this, but just. One more technical question around what are we talking about here compared to what most integrators are using today, which is Niagara? 

Terry Herr: Well, and I, I would, I would tell you that, you know, I'm, um, I love Niagara back in the X days and back when we had.

You know, LAN and BACnet competing for, um, and we had a lot of legacy protocols as well. Niagara was a great integration tool, and it still is. But I think the integration, the, the need for integration is so much less, it's, it's diminished now as BACnet has sort of taken over and we really have one protocol, thank God.

Um, so. [00:37:00] So Niagara Now is, is, I don't know, um, um, it's a, it's a favorite product. Uh, we don't promote it anymore. We think it's overpriced. Um, we think it's proprietary. We think they use a protocol that's non standard as well and proprietary. It's the only product that somehow our industry seems to think that is, is open when it's the only protocol left that's proprietary.

So, you know, Normal has a BACnet stack. Um, it programs in JavaScript. It talks from Edge to Cloud in MQTT. These are all open protocols. Um, all standard open protocols. Um, so, um, yeah, I, I think it's, um, you know, I think it's a nice step forward. I, I think Steve's done a great job. And, and, um, and it's not, it's not a SaaS.

It doesn't require, um, you know, licensing fees every year. And programs in a standard programming environment. 

Stephen Dawson: And if I could jump in, you know, the way [00:38:00] I sort of look at Niagara, right, is if you look at the 90s and the protocol wars, I think that that's where it got it started, where you wanted a single pane of glass on top of all sorts of field protocols.

And that's really where it You know it how it got its initial traction. Um, I would not go so far to say is there's only one protocol. I think today when you whenever you talk to people, but but they have shifted. I think inside of, you know, HVAC, there's really BACnet. But, you know, on the other hand, I think the trend is also for more integrations across verticals and more integrations across buildings, you know, with different SAS platforms.

And so the need for What that integration box does, I think has shifted a little bit from. Something which, you know, the legacy was more on these field protocols, where it's very low level stuff of like, Oh, can we implement, you know, integrate Modbus and Canbus and KNX and all this other junk in the building, you know, but still at the building, whereas today it's, I think that they heard that the place to look is more on these API integration [00:39:00] platforms where you have an enterprise with, you know, tens of products with different APIs that you can't really change.

And like, they're all fine. Basically, like those APIs are all fine. You know, they're all REST or something sort of reasonable, but you still need to integrate them somehow. And, you know, in order to do reporting or to do controls on top of all of them. And, and I think that that's kind of a different problem than what you face, you know, at the fieldbus level and something that, uh, you know, normal certainly strives to.

Fix for people, basically. 

James Dice: Got it. Okay. So we took a nerdy diversion as we usually do on the show. I want to go back to our, our case study, our story here. Um, so, so Nate, if I can summarize that quickly, um, you guys brought Terry in, Terry and the Intellimation team, um, sort of designed this technology stack of the normal as the IDL.

A, um, single consolidated, um, supervisory control software or front [00:40:00] end, and then you have, um, a visualization application, an FTD application, an optimization application, and a CMMS. So all the people that have seen all of our, uh, um, Stack visualizations can now picture the entire stack that we're talking about here, but let's take it back to the, the, the user themselves.

And so you mentioned you guys have a third party facility management firm. Um, can you talk about, okay, um, how is this tech tech stack getting used and sort of how have you sort of digitized the FM process? Um, Using this technology stack, 

Nathan Morris: you know, I think there are other tools that sort of fall in that same, I would say, fall into the same realm, um, in terms of our, uh, facility management technology, um, sort of bucket thinking of like access control, which is a big one for us that we've been consolidating and just trying to, you know, achieve, um, as well as intrusion detection and having enterprise level, um, [00:41:00] Really modern control over these systems.

Um, so just to note that it does like touch a lot more, I think, than even we're speaking about here, some of the tools you mentioned, like, just to clarify, like the CMMS tool, we're not, you know, we're not there yet. I'm looking forward to having everything, uh, where it is, uh, uh, smooth operation, um, and you can start to see it coming into, uh.

Into focus. A couple, I guess a couple different ways I would think about, uh, that I think we think about it is, um, number one, just from like end user experience and how quickly we're able to address issues and how quickly we're able to make adjustments and sort of thinking of it in that, you know, Domino's pizza tracker way as to, okay, a teacher has this issue in their classroom.

How does it, you know, what are the steps that it goes along to get resolved? Um, and then, you know, the other sort of lane, parallel lane with that is [00:42:00] how can we use these tools to make data informed decisions about how our equipment is performing, where we should be doing, you know, how we should be doing our capital investment strategy.

Um, you know, and, and really there's always too many projects and not enough capacity or money to support them. And it has been unfortunately a very subjective process to date. A lot of times the loudest voices went out, the people who are most well connected and know whose shirt sleeve to tug on are going to get the projects they want.

And, um, we are trying to really, um, Sort of, uh, mix in some, some fairness and equity and, uh, and really get that to a point where we can defend those decisions, uh, with with data, um, to back it up. So I think those are really the biggest use case applications for both of these. And then from that, I guess, speaking [00:43:00] to your question about.

How our facility management vendor who, you know, I'm on the phone with them, uh, or in meetings with them multiple times a day, um, making sure that the tools are they, they can be trained to use them. They know where to go for what. Um, and then also, you know, standing up that sort of how to documentation on a lot of stuff on our end internally.

So that should the facility management vendor Um, you know, change or should our mechanical vendor change or should our, you know, pest control or whoever, especially when it comes to the CMMS, like it's going to be easy for us to do that, that bake sale, that, that plug and play and make sure that we're working with vendors who are competitive and, um, you know, providing great service.

Uh, so I think that is sort of how we're thinking about it. Um, and then obviously, uh, I'm sure Terry's sleeve is run ragged from me tugging on it about where to go for. [00:44:00] Uh, sort of what answer is based on what information we might be looking for. Some of it gets down into the nitty gritty and level of, I would call it pettiness of maybe even, uh, a school staff member or a school leader who is saying, my office is always burning hot.

Um, and we can go look at data to see what temperature is it actually. And how has it been performing and maybe actually which rooms are incredibly hot. A project that Intellimation helped us out with early on is, you know, we have kitchens, um, cooking equipment, commercial cooking equipment in spaces that weren't necessarily designed to house all of it and have since been retrofitted except for on the, you know, cooling front.

And so we have staff members who are in these rooms for several hours a day when the temperatures are in the high 80s. So it was easy to make a decision to get a mini split in that room and provide really, truly needed [00:45:00] comfort to those, uh, 

staff. 

Terry Herr: I want to, um, Nate, um, talk about BRF a little bit in terms of, of data because Nate's been very successful.

Um, one of the challenges. Um, and from a data perspective for VRF systems, their BACnet gateways have historically not provided, they provided what I would call superficial data about, about the actual cassettes, temperature, um, but if you needed to get, so a VRF system has these cassettes, which are fairly simple, fan, an EV valve.

Um, you know, and, and the refrigerant coil, but they have these compressors that are incredibly complex and the gateways have historically never given you compressor level data. You'd have to walk out on site with the, with the service tool, plug in the service tool and get that. Um, things are changing in that, in that world.

And KIP has had some real problems with VRF as a lot of our clients have, right [00:46:00] there. Um, it's a great product, but you know, it's a sophisticated product. If you don't install it right and operate it right, it can give you a problem. So there's lots of failures. Anyhow. Um, Nate really pushed this envelope.

We, we knew that, for instance, Daikin, um, so, so I think at KIPP, which is the first place we got this done, we actually have, um, VRF service level data as BACnet data, right? We have it coming from, from Daikin, that we're talking about 22 points of data from, from a compressor. We, we've never had before. All the, all the refrigerant temperatures, pressures, EV valve positions, frequencies.

Um, so we got it out of Daikin and we've got it out of LG via a co automation product. So we're seeing a major trend in the industry and, and so that's a really, um, I mean, to me, we, we now have service level data. So it's deep down data that we can [00:47:00] actually do FDD on, um, at the BACnet level as opposed to a service tool.

They're very handy for, for troubleshooting future, um, problems or compressor failures. Um, so we, so we'll trend that data with like normal every five minutes, like we do every other piece of, piece of data. That data was never, even the service level tools, when they would plug in, they'd get two weeks of data, not data forever, nifty, nifty.

James Dice: So Cool. Nate, let's talk about lessons learned from this. I want to sort of talk about this in a way that another school director facilities can sort of listen to this podcast. Maybe they skip over all the nerdy stuff, but they, they, they really sort of understand how to implement this from, from your perspective.

So if you were starting this again, what are the challenges or sort of, um, road blocks that you'd want to look out for or advise someone to look out for? 

Nathan Morris: You know, I think going back to the earlier point of there is no shortage of folks selling [00:48:00] solutions, um, and finding a way to understand what you actually need, um, because not all data is created equal, not all the more data is not necessarily always better.

Good data is always better. Um, and I think I'm using data as like a broad term here, but understanding what information is needed at your campus for your school system, for your district writ large, like what do you need to make informed decisions? And so even working backwards from that, what are the details that you would need to know?

In order to, uh, you know, make those decisions and then doing your homework. Um, and if, uh, the reality is like, I didn't necessarily. But I know that Terry does his homework. Um, and so finding partners who you trust that can understand the conversation that you guys were having when I don't, you know, I don't know those acronyms and I don't necessarily need [00:49:00] to.

Uh, I know, I think it's important for. Folks in, in, you know, working, overseeing school educational facilities to have a decent bullshit detector. Pardon my language. Uh, but no, when someone is blowing smoke up your ass, um, and know the right questions to ask and to look for, like read the fine print, um, I have seen some organizations that will sell you on.

You know, we'll, we'll pay for all these upgrades to your, to your facilities. Um, there's going to be huge energy savings and then we'll just take half of those savings. And you come to find out that the half they are taking are based on their projections that were made for 10 years. And then all the recommendations they have for the energy efficiencies that they're willing to pay for, lion's share of them are behavior based.

And teachers and everyone in education knows very well that behavior change is not easily implemented. But if they're taking the cut of projected, you know, savings, you may be [00:50:00] actually paying them for what they're doing because you're not actually, those aren't actualizing and coming into, um, into being.

So, Uh, not to go down the rabbit hole in that one experience, but I think it is really just about finding some, uh, partners you trust who, who know, uh, who know the landscape, know what tools work, what tools don't, finding other organizations. Um, and I think that's been really helpful too, is even working with, uh, partners that we have at, at KIPP, uh, Texas, who, you know, is a much larger.

A network of KIPP schools serving about 35, 000 students, um, in a number of different cities across Texas and just seeing how they're doing it. Um, so yeah, that would be, uh, those would be the pieces that stand out to me the most as really critical steps. 

James Dice: Can you talk about any challenges related to the, the business, making the business case for, because this is not a traditional way that technology is implemented in [00:51:00] school districts, if there is a traditional way that technology is implemented in school districts, right?

So how do you make the business case and sort of pay for. This sort of new way of doing, doing things. 

Nathan Morris: None of this would happen if we couldn't make that case, right? Like I, I think anybody working in a school system knows that, uh, being able to point out and speak to the bottom line is going to be really important.

Um, and why is this money worth putting in here? And I think, um, so a big piece of it is again, knowing the landscape, um, politically, even like where. What, what are the, the goals and priorities of not only, you know, the leadership team of your school district, but of the school board of the local politicians who are, you know, sort of, um, uh, setting, like setting the goal posts up and potentially moving them around at times and being able to speak to their priorities.

So connecting it to what people are [00:52:00] interested in. And as it relates to the money, it's oftentimes like the savings, right? Like what's the ROI on this? What's the payback period? Um, before we see the benefit of this, or, um, if there isn't that as clearly to speak to, you know, not all of it necessarily, a CMMS tool isn't necessarily going to have an ROI or a payback period.

It's more so about how much time are we spending on this now and how well are we doing it? Um, and if you can point to that, um, and for us, you know, I think it was. Internal trust as well. When there's a lack of data, it's hard to point at that, at, at that data to, to illustrate that things are not working well.

Um, so I think it's, it's a combination of showing what current struggles are, but also what is possible if we do it differently. And that goes back to again. Looking to those other school systems that are, uh, that are operating in different ways and they might be doing one thing really well and other things, you know, not as great.

So it might take a, [00:53:00] uh, a few trips around to, to find the combination, um, of, of stories and situations to get there. And, you know, I, I still don't know, uh, that we are, are fully there. I'm still on the hunt for, for new ideas and new ways to create efficiencies in our work. 

James Dice: Awesome. Steve, Terry, anything to add to this, the challenges and lessons learned?

Stephen Dawson: I think the ROI conversation is always really, uh, really good, good one. But, you know, the, I think the point to dwell on there is You got to figure out where the, this data platform is going to land at the end of the day. Right? Are, are you selling an outcome? Cause a lot of the, you know, talk in our space is, Oh, you got to sell outcomes.

You got to sell outcomes. You got to sell outcomes. You know, on the other hand, like what's the ROI of your email system? Like what's the ROI of Salesforce for, you know, SaaS company. Right. It's like, I don't know, you have to have it to do your job. Right. And I think that a lot of these IDL conversations, you know, number one, the cost of them is going to go down.

They're not as expensive as people think. [00:54:00] And number two, I think it's just people aren't going to make a spreadsheet for that. They're going to realize that if you have 20 buildings and eight campuses, what does it mean to professionally manage that environment? Well, it means being able to look at the frigging data if nothing else.

And, you know, I think that, uh, I think that that's going to be more of the conversation where the ROI conversations will come in then is how do we, how do we leverage this to improve our vendor process so that we can. You know, have a bake off. How can we really hold, hold our different vendors accountable using that data to make sure that we're getting what we paid for, which is, you know, an evergreen conversation, I would say this industry.

But I think once you cut that properly scoped that data piece, it's just going to be something that. You know, if you want to be a facilities manager in the 21st century, you're going to need it. 

Terry Herr: But I would argue that thermal comfort, indoor air quality are also not ROIs necessarily, but they come right, you know, right along with it.

Because we can show thermal comfort charts, indoor air quality, um, you know, [00:55:00] ventilation rates. So, that data really helps that. And I know in schools that's, particularly since COVID, that's become obviously important. If you can show that data across Across, you know, months, years. That, that visibility is, is really valuable.

Nathan Morris: Yeah, I would, I would echo Terry's sentiment about, um, air quality being a new focus amongst schools everywhere, especially with the, you know, student outcome setbacks that have come as a result of COVID. I mean, math scores across the U. S. are, have bottomed out, um, in terms of where they, they were and where they were headed.

Um, so being able to show anything that is going to have a positive impact, um, and data like that shows the outcomes of those impacts on, on student performance is going to matter. Um, and to Steve's earlier point, like any, any school [00:56:00] system. And probably even wider than that. I'm just going to speak of what I know here, but I have yet to meet anybody who works, uh, for a school system in this world who has not dealt with a vendor that's not, not meeting the bar, uh, and where they feel they're being, you know, They are paying for way more than they are actually getting and maybe their hands are tied on stuff and they have to call this one vendor even when they come out they don't get the service they need or they don't get the follow up they need and so finding that story and being able to say wouldn't it be nice to have that autonomy that independence having you know by way of the IDL and anything else you To look at other options and really find somebody who, if we're going to pay this much, we might as well get our money's worth.

James Dice: Brilliant. Well, thank you so much for the three of you for coming on the show. This was great. Three different perspectives, software development, integrator, master systems integrator, and, and the building owner or the director of facilities. So, uh, [00:57:00] thanks for, for humoring me as we hit. this from all these different angles, but, uh, I think this will be a great conversation for people to listen to, to sort of follow along in your footsteps.

So thank you, gentlemen.

Rosy Khalife: Okay, friends, thank you for listening to this episode. As we continue to grow our global community of change makers, we need your help. For the next couple of months, we're challenging our listeners to share a link to their favorite Nexus episode on LinkedIn with a short post about why you listen. It would really, really help us out.

Make sure to tag us in the post so we can see it. Have a good one.

​