From behavioral demand response to dynamic pricing, there are a lot of rate structures and programs utilities can adopt to send signals to customers to reduce their power usage. The key to success is in finding the right one.
Many utilities are experimenting with time-of-use (TOU) rates, and some have gone further. By 2019, all new residential customers in California will be enrolled by default into a TOU plan. Ontario and Italy have also made variable rates the default. Utilities across the nation are now using behavioral demand response programs as a low-cost way to manage the grid. Demand response programs that pay for load reductions have been around for decades.
So how do utilities decide what works best for them? At Portland General Electric, you test them — all of them.
PGE is now in the midst of an ambitious demand response project: to evaluate 12 rate structures that enable demand response. The utility is a dual-peaking power provider, and more than ever it is focused on price-based, non-firm resources. It already has a TOU rate, but could potentially replace it with one being tested now. But a dozen pilots?
Josh Keeling oversees PGE's smart grid customer analytics, and calls it " a real labor of love."
"It's a fairly complicated program that is a really critical part of our demand response portfolio going forward. We have a pretty diverse demand response portfolio and action plan over the next few years," he said.
PGE intends to quintuple its demand response resources in the coming years, a part of 77 MW of non-firm resources outlined in its integrated resource plan last year. So far, it has developed about 20% of that goal.
Keeling spoke about the pilot, and the utility's partnership with CLEAResult and AutoGrid, in a July Demand Repsonse Dialogue hosted by the Peak Load Management Alliance. And recently he connected with Utility Dive to talk about what is working for the Portland utility.
The utility is testing three TOU rates, peak time rebates, critical peak rebates, an opt-out behavioral demand response programs, opt-out peak time rebates, and a few combinations at different price levels. "Just from a logistics standpoint, it's complicated," Keeling said. Essentially, the utility is analyzing 12 randomized control trials to find what programs are most effective.
"We have a pretty diverse portfolio being proposed in the IRP and non-firm price-based resources are a pretty big part of that," he said. The utility's portfolio of demand response remains relatively small, with a focus on residential customers. There is some capacity from the small commercial and industrial space as well, but Keeling said that is an area where the utility would like to acquire more resources.
But for now, the 12 pilots are focused on residential customers, which means marketing has become one of the keys to success for PGE.
Selling DR to an EE crowd
Signing up enough customers is a challenge for any demand response program — without sufficient participants, the resource is small and uncertain.
The Pacific Northwest, and Portland in particular, have a reputation for conservation and environmental stewardship. And as Keeling explained to the PLMA crowd, “we have a good amount of energy nerds out here, and so there was a lot of interest.”
But the utility quickly discovered that marketing energy efficiency products and programs is very different from demand response.
"There is high engagement in energy programs generally, but this is a relatively new concept," Keeling said. "The region has a lot of experience marketing and deploying energy efficiency, but demand response — and especially price-based demand response — there's really not a lot of research out here, and we had a lot of learnings in terms of what messages resonated well."
“We came into it thinking we could talk about it the same way we talk about efficiency, and while there are some synergies there is obviously a lot of nuance to that," Keeling said. “As the program grew, and we fine-tuned the messages, PGE saw bigger returns in engagement and enrollment.”
In general, messages around control, pricing and value are having a strong impact, moving beyond the conservation aspects. And the utility's use of price-based programs made the pitch easier. "There's no cost to entry, nothing ti buy. You just start participating," Keeling said.
AutoGrid's Jeffrey Norman works on the pilots with Keeling, and said it was a learning experience and an innovative project. "Anecdotally, we all believed there were messages that would work," he said. "It was fun as the project evolved, to see as the messaging was tightened how it drove adoption update."
Customer segmentation is extremely important to understanding how different messages resonate with different groups.
PGE is a dual-peaking utility due to its location and mix of resources. A growing number of homes have air conditioning (the percentage has risen quickly in the last 20 years as temperatures have risen), and many have electric heat — a holdover from the heyday of cheap power from the Bonneville Power Association. But specific populations have a tendency towards electric heat or AC, and the utility discovered different messages resonate among those group."
"We do both behavioral segmentation as well as load based," Keeling said. "It's critical for enrollment."
"The other interesting thing was learning how to have a much more proactive communication strategy," he said, "since most of the communications we have with our customers are either on a regular cadence or are reactive to actions they take."
The conventional wisdom with utilities has been that customers don't actually want to talk to them — bill complaints and outages are the big calls, in addition to stopping and starting service. So the worry has been that too much communication, be it mail, email, texting or automated call, would turn off a customer to program participation. But Keeling said PGE has seen relatively little evidence of that.
"When you're running a behavioral demand response program or a peak time rebate program, you are proactively communicating with customer on a regular basis, and Keeling said there was "some concern, internally with operations folks, that it would be more painful," but beyond some anticipated attrition, the results have been positive.
The evolution of a DR customer
PGE currently runs some direct load control programs for its commercial and industrial customers, and wants to expand those to include turnkey options as well. The utility also has a Bring Your Own Thermostat program, and is about to launch a water heater demand response program for the multi-family sector.
"But right now, the interaction between those programs isn't really happening," said Keeling. After the pilot phase ends and programs are selected to scale up (the utility has no plans to proceed with all 12 pilots), the next step will be finding ways to connect technology-enabled programs and rate-based programs.
"You have pricing and direct load control — what does both look like? It's a complicated question and I don't think anyone has figured out the best way to do it yet," said Keeling. "We see a lot of opportunities."
PGE works with the Nest thermostat, for instance, which is working to optimize home comfort schedules to electricity rates. But in its pilots, the utility is looking for load management programs that don't require the purchase of enabling devices.
"Pricing is a really strong recruiting tool for getting customers into high touch programs," said Keeling. "Once customers understand what demand response is...BDR is a great recruiting tool, a low-cost way to bring them in."
There's a progression, said Keeling, in moving customers into higher-control programs. Introduce them to the demand response concept with "a no-lose situation," like behavioral programs or peak-time rebates.
"And then move them to something a little more involved, like time of use rates, where there is some skin in the game. And then from there you move them into the more high-cost options, where they're buying enabling technologies or doing retrofits," he said.