Case Study Interview: DTE Energy

Presented by the Association for Demand Response and Smart Grid

View a recording of this case study interview

Download a copy of the accompanying PowerPoint presentation here.

Download a copy of the transcript here.


National Action Plan on Demand Response (NAP)

Case Study Interview


DTE Energy – Derek Kirchner

October 17, 2020


Dan: Welcome. This is the latest Case Study Interview in the lessons learned series produced by the Association for Demand Response and Smart Grid. My name is Dan Delurey. I'm the Executive Director of the Association for Demand Response and Smart Grid. And we'll introduce our guest in just a moment, but let me first do a little bit of an overview and backgrounder.

This series of case study interviews actually had its birth going all the way back to legislation passed by Congress in 2007. In some of the work done pursuant to that Act, one of the things that was identified by the demand response community was the idea of trying to gather and report on lessons learned. And so with support from the US Department of Energy, Office of Electricity, this series was begun and continues with that focus.

We call it a Case study Interview Because it's not a full-blown case study, and we do some of those as well at ADS. The idea of the case study interview, as well as the case studies that we do, is to look at what the lessons learned are to try to tell the story of what happened in a particular program or project, and to talk about things that were unexpectedly run into, to talk about things that perhaps would have been done differently if done all over again. And so we try and capture that and provide it to you as a practitioner or as an interested party in the area of demand response and smart grid.

So today I'm joined by Derek Kirchner who works at DTE Energy. And for those of you who haven't kept up on acronyms, that is formerly Detroit Edison. And Derek is going to talk to us today about a particular demand response pilot that DTE recently undertook. Derek, welcome to the case study interview today.

Derek: Well, thank you, Dan, and thanks for having me. We are formerly Detroit Edison or DTE Energy, and now DTE Electric Company, so lots of name changes.

Dan: Okay. Thank you on that. So, Derek, let's jump right into things here. I know the name of the program that you're going to chat with us about today is Smart Currents, but what exactly type of program is it?

Derek: It was a dynamic peak pricing pilot that we ran. We were testing a number of in-home technologies from in-home displays to full-scale appliances. We had a number of treatment cells in the pilot that we'll get into here through the program. And we were testing customer acceptance of the pilot, customer acceptance of the technology, and customer acceptance of the pricing scheme as well.

Dan:  Now, this was a program for residential customers?

Derek: It was a residential program. We developed the dynamic peak pricing rate. We had a 5,000 customer limit on residential and a 100 customer limit on small commercial. The pilot itself was focused on the residential side of the program, so we got about 1,400 participants in the pilot. We did not market to the commercial side given the focus of this was on the residential side and it was part of the SGIG Grant and the Consumer Behavior Study. We didn't have enough interest on the commercial side to really explore that, so the results that I'm going to talk about, and kind of the pilot that we went through, is going to be focused mainly on the residential side.

Dan: And you've mentioned the DOE Smart Grid Investment Grant Program or SGIG. So part of the research that was done here, I understand that was part of the requirements that went with receipt of the grant?

Derek: Yes. So the requirements, what we call our Consumer Behavior Study, and kind of the methodology and the rigor which we put with the pilot, were necessary to qualify for the Smart Grid Investment Grant. So with the federal money came a little bit more rigor than a normal pilot.

Dan: I understand. So you began to talk about the top goals of the program or, why you were doing it and what you were trying to find out. Can you elaborate on that a little bit?

Derek: This grant, the SGIG grant and the pilot really started in 2008 and 2009. And there had not been a lot of work done in the space testing customer acceptance of technologies and pricing platforms. I think we've made quantum leaps since 2009 to today in what we know about what customers will and won't do with pricing signals. But back when we started this we were still kind of new to it, or at least here at DTE, we were very new to it. So we wanted to conduct a real-world, in-the-home test of smart home and smart grid concepts, technology and really customer acceptance, to see can we get customers to shift their load, to shift their behavior and help us lower overall costs to deal with our summer peak.

Here in the Detroit area we have a summer peak of somewhere around eleven and a half to twelve thousand megawatts, and then a winter peak in the six and a half to at best seven thousand megawatt range. So for us it's a very short window of time, 5 to 10 percent of our year, which can drive a vast majority of our operating costs because we have to build for that peak, but we only operate it three to four months out of the year. There's a lot of air conditioning load here in Michigan, but it runs for very short periods of time.

Derek: We have about 10,000 megawatts of owned generation here at DTE, somewhere in that 10,000 megawatt range depending on how we count renewables and some PPAs. So we're always kind of short in the summer. We're going in on a purchase basis. We cover our winter peak, but it's a very different mix of units when we get into that summer service.

Dan: Let's talk about the time line here. Is this still in existence? Or are we after the pilot and in some different stage?

Derek: We are in a little bit of a different stage. The tariff still exists, the dynamic peak pricing tariff is still on the books, and we're looking at options for moving that forward into the future. But as far as the technology goes, we are kind of out of the technology business. In 2011 we worked with a vendor that provided the technology and the support to us. And since that time they've gotten out of the smart home business. So the stuff that we installed is out there in the field, it's in customer homes, but we're not sending signals to it and supporting it on a going-forward basis.

Dan: So a customer can still be on the tariff, but they're on their own in terms of how they manage their usage?

Derek: Correct. Per the tariff we still have to call them the day ahead of an event. So when we go into a dynamic peak price event, we have to provide notification by 6:00PM the day before. We still have to meet that requirement, but as far as the technology in the home or doing control, it's on the customers to go through and do that on their own.

Dan: Given that this was a tariff, then you did have to go through your regulators to get approval of this. And so how did that go?

Derek: We did. And when we did this we were kind of in a period where we weren't going in for a main rate case, so this is an experimental tariff. We had the Michigan Public Service Commission give us approval on it and we could show that there were going to be no cost shifting or impacts to any other rate class at the time. So we went through and established this as an experimental tariff. We've had a rate case since then. We've kept the experimental tag on it for now just because we were still in the middle of the pilot when we had our last rate case. But, I think the key in dealing with our commission specifically was to get up there as early as possible in the proceedings and start the education on what we were attempting to do with the pricing and with the pilot and explain how customers were going to be able to benefit from this rate. When we designed the tariff it was designed to be revenue neutral to a standard residential tariff and a standard residential load profile. So, our discussions were, this is providing an opportunity for the average customer to save money by making changes in their behavior. On average if you switched to the rate and you didn't change your behavior, you shouldn't have been penalized. So I think that was important in dealing with our commission. Selling that this was really an opportunity to save for a vast majority of our customers if they make some shifts in how and when they use energy.

Dan: And I know that staffs at public service commissions vary in size and scope and breadth and so on. And I think you're referring to a few years ago. But, you know, it sounds as though you were sort of educating them and talking to them about demand response, whereas it seems in some states, it's the commission pressing on demand response.

Derek: Yes, especially at this time, in 2008 and 2009, we were leading some of that education. Our staff was very impressed with what we were going after and they were up to date on some of the smart grid stuff. I think when they got into the specifics of the rate changes, it was a dual education back and forth, but they were really supportive and it definitely helped us get this in and get it enacted with success to our customers.

Dan: Well, let's go back to the beginning. And what happened? What did you run into? What were the challenges as you started to design and develop this?

Derek: Back in '08 and '09, we were on the leading edge of some of the smart home technology. We handled a number of issues with meters and devices binding. So I worked on the program and I was lucky enough to be one of the employee pilot homes. The initial install was fine and, a week later I was calling tech support back out to try and work out some kinks. So, some of the things in terms of what did we learn and how did we do it in the deployment, we started with employees as test subjects to get in there and get the real world work and install work done. As employees we probably don't complain to the commission as much as a consumer would, or at least that's the thought.

So it was easier to test out and work out some of the bugs with us. And we did that for about three months before we started rolling out to the general public. Interestingly, we do have a smart home demonstration center in our building where we had stuff hooked up and everything is working fine. And the second you got into an employee's home, trying to go through a brick wall versus a drywall wall or a plaster wall, we would run into issues. So the test lab is great, the meter lab is great, but you're never going to be able to replicate everything you're going to find out there in a customer home. And we learned that really early on even in the employee homes. And you start working a checklist through and debugging some of the issues. And we went into full deployment and you still find new things.

Dan: I want to be clear, Derek, on binding. I think what you're referring to is the meter and the in-home device communicating or is it more than just the communications?

Derek: It's really just the communications, the ability to send the signals between the meter and the in-home devices. So in our case we were doing from the meter to what was called the nucleus, which was providing the signals throughout the home to the actual devices. So there was the binding from the meter to the nucleus, and then from the nucleus to the rest of the devices to send the control signals through. So either of those two binds at any given time could come up with issues.

Dan: And then in terms of the end use or the appliances, were you connecting to anything that made sense?

Derek: We tested everything. And at the end of it when we got the results, the things that made the most sense were the programmable thermostats. The ability to send pricing signals to that thermostat and have it automatically adjust for the customer gave us the biggest bang for our buck in terms of impact. We'll talk a little bit about the results, but we were seeing about a four tenths of a kW to almost half a kW reduction in the homes at any given time. So somewhere in the neighborhood of 40 to 50 percent reduction for us on an average which was great. It kind of exceeded our expectation.

We did connect with appliances in the home, so there were a number of treatment cells, so customers could have what we called the kitchen suite, which were the dishwasher and the refrigerator, and the laundry suite, which were the washing machine and dryer. They could have all four. And we did see some impact and gain from having multiple devices connected in the home, but, the big driver was the thermostat. Here in Michigan, we mentioned earlier, it's a vast majority of the load in the home in the summer. It's anywhere from 50 to 60 or 70 percent of the load when that AC is running. So when you can affect that thermostat that's where we're going to get our biggest bang for the buck.

Dan: And thermostats are being increasingly used or sold and deployed. But were these things that consumers were able to install themselves, or did you actually get involved in that?

Derek: We were actually doing the installs on these because this was a pilot and we were using a technology that was binding to another device, and then binding to the meter for real-time information and communication. We were doing the installs ourselves, so these were not an off-the-shelf purchase.

Dan: Let's talk for a minute about how you marketed this program and how you recruited customers. You already mentioned that this was part of a larger research project of the Department of Energy that accompanied some of the Smart Grid Investment Grants. So were you told what kind of customers to go after on the basis of that overarching DOE study?

Derek: Not which type of customers to go after. It was more just making sure that the results that we were going to get we could prove that they would be statistically significant and be able to kind of use it and replicate it across our service territory. When we were doing this in 2008 and 2009 we needed that smart meter installed to be able to qualify this customer as being able to participate in the pilot. At that time we had just started ramping up our deployment. And so when we started we were probably picking from a population of somewhere around 300-350,000 meters. And some of them, because of the way we were doing deployments based on meter read routes and population, it was, kind of the same group of customers. We weren't getting a lot of variation, so we had to do some work to try and adjust as we went along and to design our consumer behavior study to be able to take the results and prove them out over time.

Now we've got over a million and a half - I think we just announced last week that we just crossed the two million meter threshold including gas modules - so it would be a little easier to do this in 2014 and have a little more variation, but back then we were making sure we could do all we could to put some variety in there.

We went through and did a direct mail program. One thing from a lessons learned perspective, make sure you have enough staff to respond. We were overwhelmed at our initial wave of mailing. And by the time we had caught up and gotten to some of the responses, some of the customers were kind of dissatisfied with us and weren't receptive to want to get back into the pilot. So we had to go out and do a second wave. And when we did the second wave there was an ability to sign up online. And when we automated that electronic enrollment, everybody was a lot happier.

Dan: And what were the messages that you were using? They obviously were received well by your customers if they really overwhelmed you with their interest and desire to participate.

Derek: Well, the main key was the ability to save money, making small changes in your behavior with a new tariff in place you're going to get an opportunity to save. That was the driver in a lot of the customer response. Half of our mailing went out with a message of “participate in this pilot and come and learn ways to save money and get on a new tariff.” The other half went out with the same message, “come and get on this pilot and learn how to save money and energy,” but, with a chance to get technology in your home that will help you do it. So as soon as we started putting in the, “hey, you can win free stuff to help you save energy,” the response rates went through the roof.

Dan: So this was pretty much across the board in terms of customer types?

Derek: It was. We did some pre-qualification before we did our mailing to make sure that we were finding places with electric service. We couldn't have it where we have a demand response program here with interruptible air conditioning where the air conditioning compressor is on a separate meter. We needed homes that weren't on that program for the thermostat to work and do the control properly and be able to do the measurement and verification on the back end. We needed customers that weren't on the IAC program, so we did a little bit of pre-qualification. But, yeah, when you start to send out messaging with free stuff, you'd be surprised at the number of customers that want to help you out, especially for a brand new washer, dryer, refrigerator, and dishwasher.

Dan:  So to actually do this, I mean I know it wasn't that long ago when a lot of utilities used to have their own field staff to do something like this. Was that the case here, or did you contract out?

Derek: We contracted most of the field work out. The technology vendor that we worked with did all of the install work for us. We had one field manager that was a DTE employee overseeing that work. So, on the initial rollout of the 15 to 20 people that were involved I'd say four or five of them were DTE employees, and then the rest were contractors, field contractors in field work. And then, post-pilot we've kind of ramped down to two FTEs. So we've got two full-time employees working on the program moving forward.

Dan: Any lessons or any lesson learned in terms of the contractor? I mean this may have been someone you were already using, or did you have to get someone special to do this?

Derek: The vendor that we selected on the technology side had their own implementation staff. That actually  made it work that much smoother. We worked with this vendor in the deployment of the technology side, and so they had a wide network of installers and delivery people in our area for normal home appliances. So adding smart appliances and smart thermostats to their repertoire was easy. That was helpful in the short run.

Dan: Okay. Now, let's talk about what you found out here. And you already talked a little bit about the impact assessment in terms of the kW that were reduced. Can you just repeat those numbers again per house on average?

Derek: So on average we were getting about a half a kW per house, which is between a 40 and 60 percent reduction at the time of peak. However, we're still waiting on final approval that all the numbers are in line. But, the results to date have shown that customers responded. And not only did they respond, but we've seen persistence in that behavior.

So we went back and did comparisons of event days versus just hot days and we're seeing maybe a 5 to 10 percent difference between what we call on an event versus what we call on a hot day. So not only are customers responding when we need them to, but on the days that we're not pushing the button or pushing the dynamic peak price, they're still shifting that load off peak for us, which was an unanticipated benefit. We hoped that we would have customers respond to the pricing the way that they did, but, there's always a little bit of skepticism in the back of my head that, will they really respond the way that we hope they will. And they did, which was, impressive and it left us on the team kind of nodding our heads that, if you build it, they will come.

Dan: I want to follow up on that, but back on the actual savings or reduction impact, in addition to kW reduction did you measure kWh savings?

Derek: We did, but the jury is kind of still out on that one. After the peak events ended at 7:00pm we did see a little bit of uptick in terms of usage versus the control group. So, for four hours your house gets warm and after 7:00 when the rate goes back down, you kick the air up a little bit more. So we're still evaluating the overall energy efficiency reduction of the pricing. I will say anecdotally we've seen much more shift than we have seen overall reduction. So customers are saving on their bills because they're moving the time around, but we haven't necessarily seen a large reduction. But, we didn't see an increase either.

Dan: So to customer acceptance, certainly you've already said that customers wanted it, and then it sounds like they were happy once they got it?

Derek: They were. We saw a pretty high level of customer satisfaction on those enrolled in the pilot, so we had a little bit of spread of word of mouth. And I think a big part of that was driven by the fact that everybody wanted some free appliances. So if I got a thermostat and I told you about it and you were my neighbor, you'd call in and try to get on here as well. So we had a few issues there where customers were a little upset that they couldn't get in the program, but we were trying to make sure we could stay with the rigorous study. So we stopped supporting the tariff in June, or stopped supporting the technology, I should say, in June of this year and have not seen customers dropping off. We've kept a pretty sustained enrollment number from pilot to full tariff rollout.

Dan: I want to ask a question about the discontinuation of the technology. But first I'm realizing that we didn't really talk about the monetary savings. That's been a theme as we've chatted here about that, but can you give any round number figures in terms of what people were saving?

Derek: Off the top of my head I want to say they were saving somewhere between 5 and 12 percent on their bill, I think was what the overall number came in at. There were those of us, and I'll say that because I'm remembering back to when I was on the tariff and measuring my savings, I was somewhere around 21 percent, but I made sure I was excluded from the pilot results knowing what I know and designing the tariffs the way that I did. But there were definitely opportunities out there to get substantial savings.

Dan: And going all the way back to the marketing message, did you use any numbers in your marketing message?

Derek: We really didn't. We were just trying to tell customers that there was an opportunity to save. And we didn't want to try and monetize that value because we didn't know what we were going to get and we weren't quite sure how these were going to work. Now, on the printed material we didn't really put anything in place. In the conversations that we had with customers, whether on the phone or in their homes, we were kind of in that 5 to 10 percent range, but we didn't put anything in print just because we didn't want to over-promise and under-deliver.

Dan: Well, let's begin to wrap up here by doing the look-back. But I guess the first thing I want to include in that is this discontinuation of the technology. So I'm assuming that you didn't foresee that and that that was unexpected?

Derek: We didn't. We put this in place and kind of thought that we were going to have something longer term. It didn't end up working out, but I think that in the industry itself we've seen kind of a proliferation of technologies that are available to customers, not just from utility-based programs, but, for the solutions for themselves, something like a Nest that you can go buy off a shelf. And to make those changes on your own without necessarily the utility helping out. Hindsight being what it is, we probably would have chosen differently, but, it still provided us with the data enough to say, look, this works. And we can make it work and we can make it work here in our service territory.

Dan: So you've already sort of talked about the fact that the reaction of customers was better than expected. And so that sounds like an overall positive when you're trying to find out if something works, you get sort of a clear signal that it does. But what other takeaways, as you're speaking to others through this case study interview, what would the lessons learned be that you'd want to convey to others?

Derek: So I think the number one thing we learned is that customers are willing to help you out. They're willing to make the changes if you provide them with the education and the tools to do so. The technology definitely helps. The ability to kind of set it and forget it and have it done for them makes it that much easier. We ended up getting a bigger bang for our buck for that group of customers.

But, even having the education and having the tariff to help them make changes in their behavior, definitely whether it's a pilot or whether it's a deployment, make sure you have something in place to support that technology. If it's a vendor solution, make sure that there's going to be ongoing support for it. If it's something that you're doing on your own, making sure for the long run that it's supportive.

I think as an industry, sometimes we can run a bunch of pilots and the numbers come back really good, and then all of a sudden, something has changed and we kind of shelve it a little bit. I would be careful of that. Your greatest customer expectation and they expect you to continue to deliver regardless of what might make overall sense.

Dan: I assume you thought that there'd have to be a certain degree of handholding for customers when you're doing something new. Was that the case? Did that turn out kind of like you expected, or was it more or less than you expected?

Derek: It did. It turned out about what we expected. We actually had established our own internal call center. Our pilot customers they had their own 800 number to call. That was staffed by someone from DTE 24/7. And we kind of anticipated that being a full-time job for someone for the first year or two of the pilot and it was. They weren't overwhelmed. It wasn't a thousand calls a day, but there were enough calls to keep somebody busy between technology troubleshooting and billing questions and why is this doing this in my house, and how do I override a dynamic price signal on my thermostat. That was helpful. That's another kind of lesson is to make sure that the technology is great and the tariff is great and you can install it, but don't just kind of clap your hands and say, hey, we'll just see how this works for the next two years. There's going to need to be ongoing support and make sure that back office is ready to go.

Dan: So we always end these case study interviews, and you've talked a little bit about this, but we always end with these two questions. And that is what would you have done differently if you had it to do over again? And the corollary to that, what do you wish that you had done that you didn't do?

Derek: The one thing I'd love to do over again is just to get the technology piece right. We saw the biggest gains in customer acceptance and customer reduction with a piece of tech that we could work with. And missing that point or missing that piece in the home, I wish we still had that.

What do I wish that we could have done? When we were looking at this we wanted to explore a peak time rebate to see was it carrot or is it the stick, and can you get a bigger bang for your buck one way or another. We didn't have the capability at that time to kind of do the ten-day base line on a peak time rebate and provide the data analysis to get that credit back to the customer. But, I would have loved to have seen us been able to test the pros and cons of both pricing treatments.

Dan: Well, Derek, what haven't we talked about that you would like to mention as we conclude here?

Derek: We talked about it a little bit at the beginning, but I think that as you're looking at doing something like this, that the number one thing you can do from a regulator standpoint, from a stakeholder standpoint, internally, externally start the education and start it early with kind of what you expect and what the customers can expect and be up front and honest about it that. We had those discussions with staff. We had a meeting, once-a-month meeting, and we're up there enough that they understood what we were trying to do and what we could expect from it and what our customers could expect from it.

I mentioned it a little bit earlier, we built the technology or a smart home demonstration center. We brought our engaged stakeholders through, we brought our skeptic stakeholders through to say, look, this is how the meters are going to work and this is how the technology is going to work and this is what it can do. So having that, having those conversations on the front end makes life so much easier on the back end because you're not answering those questions as you go along. When those stakeholders are engaged, it makes it that much smoother.

Dan: Well, Derek, congratulations on the program.

Derek: Thank you.

Dan: And thank you for sharing it with us today. That was really interesting and definitely had some lessons learned to pass along.

If you are interested in other case study interviews or other case studies, you can go to the ADS website at, and there's lots of other information there as well. So thanks again to you, Derek, and with that we will conclude this case study interview.

© 2016 Solar Electric Power Association    ::     1220 19th Street NW, Suite 800, Washington, D.C., 20036    ::   contact us

Periodic updates on news & events related to demand response and smart grid.

Powered by Wild Apricot Membership Software