Bringing energy efficiency data analytics to new data sets, such as natural gas interval data, makes advanced building operations behavior more informed and more meaningful. This is simultaneous heating and cooling, new benchmarks, and leverage on new ENERGY STAR scores!
The following is a slightly edited webinar transcript. To see a recording, please visit the EDF Climate Corps Tools & Guides page.
Sarah Berman: Good afternoon, everyone. Thank you all for joining us today for the EDF Climate Corps 3E Webinar Series, a summer long free webinar series focused on Energy, Efficiency, and Expertise featuring thought leaders in the Energy Management sector in support of the 2018 Climate Corps Fellowship Program and of course the wonderful EDF Climate Corps Network. I’m Sarah Berman and I’m very happy to be the moderator for today’s webinar. In just a moment, I’m going to be handing the presentation over to Tom Arnold, Co-Founder and CEO at Gridium who has prepared a presentation titled “Natural Gas Data Meet Interval Analytics.”
First, since we have some participants today who are new to our network, I’m going to take a moment to introduce Climate Corps and describe our program. EDF Climate Corps is a network of professionals united to advance climate solutions. Environmental Defense Fund matches leading organizations with top graduate students to advance the ways that organizations manage energy and spearhead sustainability. Fellows are engaged in a broad array of projects from clean energy generation and procurement, to energy data analysis, to sustainability strategy and goal setting.
Since the program’s inception back in 2008, we’ve placed over 950 graduate students into more than 460 organizations in both the US and China, at host organizations such as Kickstarter, Gap Inc., Dow Chemical and the New York City Mayor’s Offices of Sustainability. The results of these projects have been tremendous and fellows have identified $1.5 billion dollars in energy savings and have left a lasting impact on the companies they worked with.
If you would like to learn more about hosting an EDF Climate Corps next summer, please visit our website. And now, I’d like to introduce our presenter today.
Tom Arnold is the Co-Founder and CEO of Gridium. Prior to Gridium, Tom was the Vice President of Energy Efficiency at EnerNOC and Co-Founder at TerraPass. Tom has an MBA from the Wharton School of Business at the University of Pennsylvania, and a BA in Economics from Dartmouth College. And when he isn’t thinking about the future of buildings, he enjoys riding his bike and chasing after his two daughters.
So, following Tom’s presentation, we will have time before the end of the hour to take questions from the audience.
As a reminder, EDF Climate Corps does not endorse Gridium nor any of the vendors or presenters of the 3E Webinar Series. This is a forum for discussion and for sharing of resources focused on Smart Energy Management Practices. Also a recording of this presentation will be available on our website. We have several more webinars on a variety of Energy Management topics throughout the rest of the summer. And last but not least, we encourage you to tweet about us using the #EDFCC.
Tom Arnold: Thanks Sarah and thanks for the opportunity. Well, what we’re going to talk about today is interval analytics meeting gas data. This is a topic that we’ve been working on for a long time. In prior webinars, we’ve discussed mainly electricity analytics. I’ll cover some of the main points of electricity analytics but try to focus a lot of the discussion on natural gas.
For those of you that don’t know Gridium, Gridium is a software company. We’re based here in San Francisco, and our mission is to turn building data into actionable insights for people that run buildings. We do have energy management software, and in fact some of you and host organizations that use that software, but everything I talk about today could be done on spreadsheets or other types of analysis tools or even less nifty software from other vendors.
Let’s talk about data for a second. What we’re going to talk about today is meter data. And, everyone on the call is concerned with energy and trying to look at projects both large and small. And increasingly, those organizations are trying to develop data-driven strategies for the actual projects that they want to take a look at.
Why is meter data is so important?
Well, one of the things that is very difficult about buildings is every building is a snowflake. Buildings are entirely custom built, there are no common themes. Even on a thing like a building management system, you’ll see proprietary protocols, you’ll see different ways to get the data, different ways to control the data. If you think about the air conditioning system, there are as many ways to design an air-conditioning system as there are vendors and architects and MEP firms. And so what’s very difficult about sustainability in general in energy management is finding a common data layer. And that’s what’s exciting about meter data because every building has a meter. And what it allows you to do is have a common frame of reference for understanding what’s going on in a building and trying to make recommendations for both small projects and large projects about what actually is next.
Historically, we haven’t had much data on buildings.
Most of the time, for energy management for 100 years, we’ve gotten 12 readings a year for 12 monthly bills. And that gives you some picture of what’s going on. You can for instance, calculate intensities in terms of square footage per year, you can look at trends, you can compare one building to another. But you get an opaque view of what’s actually going on. And one very exciting thing is with more and more meters going in and more and more of those meters being smart meters, and the data being shared, suddenly you get a very clear picture of what’s going on. On the right is approximately 35,000 pixels, which is as much data as you get from a 15-minute data recorder a year.
And suddenly with that, you can determine all kinds of different things.
You can find out what the energy is at 2:00 AM on a Saturday when no one is in the building. You can find out when you set your peak demand. And you can do even more fancy stuff when you take that data and analyze it with regression techniques so that you understand what exactly are the factors that make energy go higher or lower.
The other really exciting part about data acquisition today is, for decades, this has been a very expensive proposition and for most commercial customers, they’ve had to install their own meters, install their own data recorders. That has been a real barrier here, but there’s a big change underfoot. I think most of you are fully aware of this, but smart meters are really continuing to grow. This chart just shows you how many households “have smart meters.” And you can see across the US, more and more states have smart meters.
The situation in commercial is even better.
You might hear the term AMR, which is kind of a really old version of smart meters that are typically run on phone lines or other proprietary networks. And even places like Seattle, here, which doesn’t really have smart meters on the residential side does have smart meters on the commercial side. Same thing with New York, Massachusetts, other states, too. So most of the time, if you’re in a host organization and looking for data, you’re going to be able to find some data in terms of meter data from electric or gas, or combination of that just by looking at your local utility website.
That’s the other really exciting part about this, is the websites have gotten a lot better. These websites used to be horrific, you used to have to contact the utility representative and you would get a really ugly spreadsheet of data. More and more utilities across the country are adopting standards like Green Button Connect and they make it very easy for you to extract that data and share the data with companies like Gridium, companies in the solar business, all kinds of different companies. Our local utility here, Pacific Gas & Electric, they now have 52 places to share your data which is kind of amazing that so many people have built the integration into this automatic feed. You can also download directly from the website.
So that’s exciting, and hopefully what that creates for everybody here is an ability to say, “Hey, where is actually our data?” and “Can I go look at that to help data drive some of the strategies I’m looking at this summer as a follow up?”
Once you got the data, one thing that you got to recognize is that the data is just the start of a conversation. And, what we focus at Gridium on—and I hope a lot of fellows recognize as well—is the data is just a very small part of building operations.
The real opportunity here is bringing data to people. Because with a handful of exceptions, a few very advanced buildings in the world, most buildings still have a lot of human operator decision. And it’s the decisions that those operators make that determine when and how you’re going to use energy. And so your task is of course to do the analysis, to find the opportunities, to data drive your strategies but ultimately build bridges to operators in your buildings and bring them along into your world.
Now, we’re going to show you some data here, but one tip here is that there are no problems in building operations, there are just opportunities. Most engineers that actually run buildings want to save energy, that’s what they’ve been trained to do, but they’re not going to see this data regularly, unless of course they have a system like Gridium. So, part of the recipe is how can you talk about an opportunity?
Let’s just show one simple opportunity here. This is a snapshot of a building, and as you can see the brown line is the 15-minute interval data. You can see that the building is largely off on the weekends and then you’ve got 5 days, Monday through Friday. The blue line is therm usage in the building. So this is a building in PG&E territory out here in California where we get daily gas data from most commercial buildings.
Now, I know that there’s no open line here, but hopefully by now everyone has identified the fact that, “Hey, wait a second, the boiler uses the exact amount of energy, the same energy the same day, every day. That’s an opportunity, right? The big opportunity is the weekend. Obviously the building is off and yet the boiler uses as much on a Saturday as it does on a Wednesday. The other thing you might notice, which is the next opportunity is the boiler uses is amazingly constant each day. The temperature is not that constant, I didn’t put the overlay here but you can go back to the weather record and see that in fact Friday was a lot warmer than Tuesday.
So there’s obviously an opportunity here. There’s an opportunity for a weekend to setback and there’s probably also an opportunity for outside air reset on the boiler. That is changing the temperature of the boiler according to outside air or according to the actual load conditions in the building.
This is one just very simple example of how you can display this data and show an operator what’s actually going on and then have a discussion about, “Okay, what exactly is happening with the boiler?” and “What controls do we have on that boiler?” and “Is that an actual opportunity for it?”
Some of you might be engaged in audits—and this is a snapshot of recent data—but when you look at an audit, remember that when you are going through and understanding operating conditions as part of your audit, you’re only really seeing the operations of that day. You don’t have any perspective of what happened in prior history. And if you visit a building in winter and do a winter audit, you look at the boiler running and you’re like, “Check, everything’s going perfectly.” However if you arm yourself with historical data from meters, you can see very clearly what’s going on.
Here’s a visualization that we use that is very intuitive, and actually we used this for property managers, all kinds of non-energy people. But all we do is we take each days usage and color code it. So the way this graph works is each cell is a day, you can see the billing lines for the billing periods, and the reds are really high therm usage, and the blue is really cold or low therm usage. So you can see obviously in December, in January, it’s cold. You are using more gas, you see a little bit of fluctuation with weather, but the big thing that you see over here is in July last year, we’re still using as much gas in July as in March.
Now, I Mark Twain is said to have said that summer in San Francisco is fairly cold, but it’s not that cold. So this is a definite opportunity to go in and say, “Okay. We had issues with seasonal boiler shutdown before, maybe that’s an opportunity that we want to actually look at.”
This is daily data which across in gas is going to be the most frequent usage that you’ll find. Some of you will get even luckier. For instance if you’re in Southern California, you will get hourly data. And here you get just an even better view of what’s going on. This is a building in LA. And same kind of visual language that we use here in our analysis, you can see that the boiler use does go down quite a lot at night, you can see the very clear operating windows, you can see the weekends in between the orange box where the boilers, largely all set. But one thing you might notice is from May and to July, they’re using basically the same amount of gas from 5:00 AM all the way to 6:00 PM.
Now this is Los Angeles, so you know the air-conditioning is running at the exact same time. And, this is what we refer to as the New York City Taxi Cab Building Management Strategy where you got one foot on the gas and one foot on the brake. That’s how you control the car. This is the exact same thing. This is simultaneous heating and cooling where the air-conditioning is essentially fighting against the boiler and you get the same kind of consequences as that taxi cab driver does on his miles per gallon.
So again, this is another opportunity that you can see very clearly in the data and actually data drive a strategy here. And that strategy might be better controls, it might be just alerting the operator, it might be, “We got a very antiquated heating strategy here and as part of the big capital upgrade, maybe we’re going to pursue and electrify everything strategy,” or “maybe we’re going to get a better boiler,” or “maybe we’re going to do something even more advanced like an isothermal or cogeneration or something like that. But at least you got the data to understand exactly what is going on.
The last piece of this is (maybe a little bit more fancy) but if the technical folks on the call would like to work on regressions, you can get a lot of insights from building a simple temperature response model. And there are packages that will do this for you. You can use the tool like Gridium as well, you can even do it and have a basic spreadsheet approach. But often, what we’ll find here is kind of clues. Because you might look at heat map like this and say, “Ahh, why is this happening?” Especially in a large building with lots of zones and if the zones kind of drive the ship, it can be difficult to find out why. And sometimes by graphing this, you’ll see what’s going on.
So, this is just a graph of temperature response by therm and you can see what’s going on. There’s an upward sloping curve as you go left, cooler more gas, that makes sense. But what you’ll notice is the responses are different on different days. So Monday we use a lot more gas when it’s cold. That kind of makes sense if we’re setting back on the weekend, but you’ll notice also that the closed line is not flat. And that means that something is responding to temperature on days that are actually closed. And so that can be a good clue that, “Okay, if we’re really going to work on weekend gas setback, we want to make sure that we have very wide dead bands that no zone can really sort of start the boiler and start moving energy across the building. So, little items like this can really help as well.
So, how does this actually affect operations? Well, the first half is, “Do you have an opportunity?” And when using analytics to identify an opportunity, that starts at data driven conversation that you can take to an operator. And with that conversation, once you’re actually driving the data and talking about what the system is doing rather than what the person is doing, you tend to diffuse any potential issues and you tend to sort of get everybody on the same page of, “Look, here’s an opportunity. Is this something that we can actually get our arms around?”
And if you work on gas, one thing that you will find in our industry is a huge tendency to just dismiss gas.
And one thing that is interesting is, that’s a really financial view. Most of you are working on an energy management, and so you deserve your natural attitude and sort of the tools that you’ve been given are really sort of financially motivated in terms of projects. And so it’s very tempting to look at numbers like this and say, “Ahh, most of this is electric, let’s work on electric. That’s where the big lift is,” and just dismiss gas. But that’s really just kind of missing the opportunity, and hopefully I’ll explain why.
This is an example building where… it’s one of the buildings we looked at before. We’re going to use it just to play with some data. So here, gas is about 11% of their total spend. This building is running about $350,000 a year on energy and it’s $2 a square foot. So, gas is tempting to ignore. “It’s not much money, why should I pay attention to it?”
Well, let’s think about reporting for a second.
I think most of you are familiar with Energy Star. Most of your organizations either have to report Energy Star for benchmarking ordinances or did you so voluntarily, otherwise they wouldn’t be hosting you this summer. And you might have read that the Energy Star scores are changing. In prior years, Energy Star has sort of felt like a little bit like Lake Wobegon where all the kids are above average. And that’s because of the reference scores are from a survey of commercial buildings in 2003. And obviously commercial buildings have gone tremendously more efficient in the last 15 years. So starting in late August, the score, that reference will change and the average decrease in Energy Star score is going to be 12 points in the office category. That’s different across different verticals, so it might be different for hospitality, or food service, or schools. But in office in particular that score is going to go down dramatically, which has all kinds of flow-down consequences.
If you have a building that has an Energy Star score of 84, that new score might be something in the mid-70s, which means you now have a barrier on LEED, you now aren’t excused from certain ordinances. And if you are a part of an organization that reports out certifications statistic or requires in your leases that you’d be certified, you now have a problem. So this is timely, it’s going to affect all of your organizations. And now we’re going to sort of bring it back to gas and help you think about a high leverage point on that score.
So, think about that building before. 11% of the spend was gas, “Yeah, who cares? Let’s work on the 90%.” But wait a second, if I translate that gas into an equivalent thermal unit as electricity, it’s 44% of the kbtu in the building. And if you’re a carbon person, depending on what kind of electricity you are sourcing, of course, that’s the major portion of your carbon emissions as well. That’s how you count things.
So it’s a big portion of the kbtu, and systems like Energy Star are highly sensitive to changes in kbtu because there is energy source in the calculation. So, think about that building where the boiler was running all summer. This is an example building, and what we’ve done is just simulate changes in the energy use filed in with Energy Star and looked at the changes in Energy Star score. You can do it just on your own by the way. Create a Portfolio Manager account, run some scenarios, and you can do this for your building or a set of buildings. So, at full gas, i.e. boiler runs all summer, still pretty efficient building—84, certified, put the plaque on the wall, do your LEED Certification. You know there’s an opportunity but it’s not going to sort of damage your strategy until the new numbers come out.
If we drop that gas use in the summer 50%, Energy Star goes up to 86%. If we drop it 100% and take the boiler completely off, suddenly, it’s 88. And these might not seem like very big numbers but for anyone that’s been through Energy Star, getting 4 points for a simple operational change is a huge win. It’s why more and more people dig into gas data. I think this is going to be kind of a common trick where you got a low Energy Star score, you know you need ideas to do it, you don’t have time to replace the boiler or do a massive HVAC project, you’re looking for sort of operational leverage and looking at your summer time gas and boiler resets and just working on that is a huge leverage point on bringing that Energy Star score up.
We’ve talked about operations, we’ve talked about reporting.
For those of you that are thinking more about energy efficiency strategy, what you’re going to do (this is straight out of the Climate Corps EDF materials and maybe you’ve reviewed this) but if you’re thinking about framing the data acquisition and framing your data strategy, think also about the barriers here. We’ve talked about how you could deal with a lot of these first two items in the organizational priorities which is with lack of accountability and difficulty assessing the performance. So hopefully, you can have an idea of how you would go find if you had data available, make that data connect with an operator to identify opportunities. You can even put in systems like Gridium to get that data feedback loop going to an actual operator in you’re building.
What about actual investments? Well, if you make changes here, most of you are familiar with measurement and verification, and once you have meter data, you can use that to power data driven measurement and verification. So, this is straight out of IPMVP Option C, and the idea here is very simple. You create a baseline that is weather normalized, you do the ECM or the change, (Energy Conservation Measure) and you measure the difference. And again, you can do this in spreadsheets or others packages, you can do also do it in a system like Gridium. And big changes are possible.
If you’re kind of skeptical about whether that example building was possible or realistically, whether you could really get 4 to 6 points by setting the boiler back, here’s one of our buildings in San Francisco. And this is June 2018 versus June 2017. It’s a new operator. The old operator was more conservative, he saw this opportunity, he grabbed it and his therms are down 50%. And that’s in San Francisco where you still have to run a boiler because of our cold mornings. If you’re in a warmer climate, you could come down completely 100%.
So, this is a very clear way to do the project and make sure that if you do the project, you know that you’re going to get operator credit and feedback. And sometimes there’s going to be capital involved. If you really want to work on boiler setbacks, you might have an issue like hydronic valves or leaks, or controls that you need to install, and so you might actually want to tie this back to the financials to the project. And that is very important. The more that you data drive the strategy on both sides, the more success you’re going to have in the organization. And that means identifying the opportunities upfront using a data driven strategy, and it also means closing the loop at the end and making sure that you report back to the organization, “We thought this is an issue, we did the ECM, and here are the savings from that ECM.”
That’s the start, the flywheel going in your organization. Especially when you go back to the business case of your project, once you start delivering on good business cases, the organization as you go up, the organization is going to trust you more and more and more, which is exactly the goal that you have so that you can get more projects done.
Okay, that is a whirlwind tour of natural gas analytics, and hopefully that you got a sense of why this data is useful and how you might use it even in summer to talk about your projects.
Sarah Berman: Thank you so much, Tom. That was a fantastic presentation and relevant to so many follow up projects this summer.
First question, “How sensitive is gas use to weather and how does that compare to electricity?”
Tom Arnold: Yeah, it’s a good question. Well, as always the answer is it depends. Let’s look at this chart. If you look at this chart, the answer is neither electricity nor gas are sensitive to weather. And of course that’s not the answer. Gas should be very sensitive to weather. When we do billing analysis for example, a typical monthly bill will swing on gas from the same period last year will be 10, 20% whereas electricity, much more 1, 2%. So, electricity, at least in a commercial building, 40% of those is lighting so that doesn’t change. 40% of it might be HVAC, but a huge proportion of that is ventilation. And the ventilation rate doesn’t really change with the temperature.
So, if you’re not seeing temperature sensitivity in your gas data, you should be seeing curves that look kind of like this. If you’re not seeing that temperature sensitivity, you got a real problem and you should investigate it and you should see a lot more temperature response on gas than electric.
Sarah Berman: Fantastic! Thanks so much. Okay, on to the next question. “What periodicity of gas data is generally available and how fine does it have to be to be useful?”
Tom Arnold: This is going to depend by your utility. Utility tended to do… residential AMIs have higher intervals and those are typically daily or hourly. There is a small portion in very large commercial like hospitals or industrial plants, you will find 15-minute gas data. Most of what we see is daily or hourly. I think daily gives you a lot. It can help identify resets, it tells you whether things are really coming down. What it doesn’t is tell you about the daily operations. When you get this, you’re very excited because you can really get a sense of simultaneous heating and cooling, which is a big issue in a lot of buildings.
So, you are at the mercy of what you’ve got largely, but both are useful in terms of analyses. You can install your own meters. So if you’re in an area where there is no gas data, it is possible to get gas metering done. There are variety of non insertion methods that will try to infer what’s going through the pipe without actually disrupting the gas supply. That’s obviously going to be expensive and you’re obviously not going to get historical data until the meters are actually installed.
Sarah Berman: So this question is regarding school districts and municipalities and it says, “For making the business case to keep track of gas data, (not so much electricity) what are your recommendation to make the business case to smaller size governments, municipalities, townships, school districts, and the like?”
Tom Arnold: I mean the nice thing about gas is, if you were looking at that chart and seen that the boiler was running all weekend and shaking your fist, that’s your correct instinct, right? It’s very easy to identify opportunities on the phase that are a bit wasteful. There might be complexities there, but everyone sort of gets that you should run the boiler in a mild climate on the weekend. So, there is typically a very good business case there because a lot of the changes are actually operational in nature so it tends to be a little bit in the low hanging fruit category.
In terms of making that business case seen, I think that’s where you go back to the reporting. Which is, hopefully the organization realizes that the future is going to involve them publicly or semi-publicly, disclosing their energy performance against reference scores like Energy Star or other ordinances. And, this is a high-lever item because all the frameworks in terms of analyzing those buildings come on a normalized basis across different energy supply sources. So it’s a very high-lever item.
And then thirdly, if your organization is very advanced and really has a carbon framework, of course natural gas is your primary sight in emission and affecting that lowers those CO2 emissions and get a better understanding of where the carbon is actually coming from.
Sarah Berman: Great! Thanks, Tom. Okay, so this is a two-part question. “In the building energy analytics industry, what trends are you currently seeing and where do you see this industry going in the next 5 years or so?”
Tom Arnold: That’s a good question. I’d love to actually know that definitively. It’s easy on the outside I think to see building analytics as a very crowded space. Once you’re actually in it, you’ll find that it’s incredibly nascent. Even a company like Gridium, we’ve service over a thousand buildings but that’s a tiny percentage of the stock of buildings. And other small companies do kind of similar amounts.
So, my perspective on this market is actually it’s really quiet early on. I think you have packets in cities like Seattle, San Francisco, LA, where the large owners are highly motivated and really sophisticated, but some of you might be in less progressive organizations that are just starting their sustainability journey and aren’t familiar with these types of tools. So, even if you don’t have the wherewithal to try Gridium or similar platforms, you might be surprised about how much impact just pulling this data and visualizing it for people actually makes, because this is not a mature market in any sense of the word.
In terms of trends, I think everyone is out there proving the value of these and these installations, and so more and more people are using it and are getting success from it. And the market is growing at a good clip.
So I don’t know if that answers your question, but those are the trends that we tend to see in this market.
Sarah Berman: Wonderful! Thank you. And just following up, “In the next 5 years, do you have any inclination of what this industry is going to look like?”
Tom Arnold: I think that there are now… while this software is underadopted, they’re now in sort of category of pretty clear leaders that are going to help build the market together. And so I think you’ll see less frothiness that you’ve seen before in terms of what’s going on in the market side of things.
I do think that also there are some other trends here. It’s pretty clear that everything related to energy analytics on the control side of the building, so fault detection, mapping BMS points, getting really into the weeds and say, “Hey this valve is open when it should be closed.” That is emerging as a pretty distinct piece of the market in metering, getting submeters, getting real time meters, getting utility meters, getting this kind of spend analytics going through. That seems to be a pretty clear division whereas a few years ago, there were people kind of trying to unify both of those.
Sarah Berman: Okay. The next question also has two parts. So first, “At what point do you recommend a business using a service like Gridium versus maintaining and analyzing their own data?” Part two, “What size building portfolio should make the business case for a software?”
Tom Arnold: Good question. So, “at what point” is really a question of what your bandwidth is and what your capabilities are. The software is also free to try. So, that’s mainly how we go to market and how other people go to the market as well is, “Hey, give us your data and you can see the tool and see if it’s useful.”
Where the software tends to shine is in portfolios. So, doing this by hand in a single building is tedious but possible. Doing this when you’ve got a 100 or 200 building or meters, it’s just impossible. You need a computer to do it and you need the computer to also manage the information flow. Bob that runs Building 123 can’t really efficiently get data from the manual system, they want data sent for their building only.
In terms of the size of building, we have the full spectrum. Typically if it’s a single building, it needs to be roughly 100,000 square feet to make the math work. But an example, we have a large housewares retail customer that deploys us on every single store. And some of those stores are 2000 square feet and they are looking at exactly the same thing. We’re doing LED, measurement and verification studies for them where we’re asking why the store starts at 2:00 in the morning when the store hours don’t open until 11:00? So, technically you can do it on any size building, just work with your vendor to make sure that pricing makes sense on those types of arrangements.
Sarah Berman: Excellent! Okay next question, “Should we be looking for monthly or maybe weekly pattern shifts?
Tom Arnold: You definitely want to look at the tightest interval that you can. So if you have hourly data, you should be looking at hourly trends. If you’re looking at daily data, you should be obviously looking at daily and so forth. I think the broader, and the reason that regression analysis is so powerful is once you put regression factors into a basic model, you can develop insights about longer term trends. So for instance—we didn’t talk about this—but you can measure, just like you measure baseload on electric, (baseload is the minimum reading in a building) you can detect and measure changes in baseload overnight. So you noticed that these colors kind of change the third week or the second week in June, what happened? Why did our overnight gas go up?” It’s not very much gas at whole, but why did it go up?
So when you have regression factors, you can start to develop more fine-tuned diagnostics of what exactly is going on because you can strip out the noise and you can clearly see the signal of what’s going on in your data.
Sarah Berman: Okay next question, “What types of companies and industries have been early adopters of this technology and who do you see as the next opportunity?”
Tom Arnold: Early adopters have been real estate. That’s commercial real estate where sustainability has been a major factor in leases especially the high tech companies. That’s been a big portion of our base. This second wave has been corporates that have a sustainability mandate. Again, at the barrier, that’s largely tech where they are looking at sustainability as part of their business mantra and it’s where their employees are saying, “Hey, what are we doing about energy?”
So those have been the classical early adopters. And to go back to the previous question, that was the bulk of the market for a long time but actually that is one of the important trends is that more and more people across industries are using this. As an example, we now have a concrete and aggregate company using it, which is very interesting. They obviously have very high energy use, they have low margins, but they’re getting a lot of value out of the software in terms of managing when they’re using energy and how they reduce the expenses of energy.
So you are seeing industrials start to adopt this. And then again, I mentioned retail. There’s a lots of interesting things going on in retail. Obviously, a huge lever on operating expenses and people are realizing that the cost of these systems are coming down so much that you can actually efficiently deploy it in retail environments as well.
Sarah Berman: Okay next question, “Are gas EEMs more or less challenging for building engineers to implement?”
Tom Arnold: As usual, it depends. If you’ve got a good boiler system and you understand it and it’s modern, it’s very easy… often times when you see the boiler running in the summer, someone has just disabled the resets and it’s a simple matter of re-enabling the resets.
You will find some buildings where the integrity of the hot water system is dependent on high temperature. Essentially the valves need that high temperature to make sure that they don’t leak. If you want to watch some ploddy videos on this, just Google on YouTube and you can see people set in their boiler’s back and water going all over the roof.
So that’s the one kind of “gotcha” on this. And in fact, the example I showed in terms of savings, this was actually a capital project. In order to affect that, they had to replace those victaulic valves. But the payback was very good and they’ve accomplished that and got a good ROI on the actual project. Some boilers are really old and boy, if you’re in a historical facility, if you’re in a hospitality, these are going to be harder things for you to deal with. You typically have very big hot water demands, people are very nervous to touch anything to do with the hot water because it’s so important to the guest experience. So, you can have a lot more sort of operational pushback on these types of measures.
Sarah Berman: Great! Thank you again, Tom. And if anyone has follow ups, please do feel free to reach out to Tom directly.
Tom Arnold: Absolutely! My email is here and we love talking to fellows and alumni. Please reach out if you have additional questions.
Sarah Berman: It looks like that’s it for today’s questions. But I want to say thank you to Gridium for hosting today’s webinar. As a reminder, we have several webinars left this summer as part of the 3E Series so make sure to register and tune in.
Thanks again everyone for attending today and have a great rest of your day.