Energy use feedback: worth pursuing on mass scale?

Report: variables involved in motivating consumers

Phil Carson | Feb 27, 2012


A new report attempts to synthesize recent studies on large-scale, real-time energy use feedback pilots in the United States, the United Kingdom and Ireland.

"Results from Recent Real-time Feedback Studies," by the American Council for an Energy-Efficient Environment (ACEEE), concludes that energy savings depend on a number of variables.

The  ACEEE report uses different terminology than is often employed in smart grid discussions, and the 28-page report contains much nuance, so I'd suggest giving it a read. But a few points are worth noting.

First, the basis for the report. The introduction noted that a major question lingers over demand side management: "How to better engage customers with their energy usage? What methods work best to deliver information in order to enable behavior change? How long [might] any effect be reasonably expected to last? (The old "persistence" question.) And: How best to measure and evaluate any savings as part of a utility energy efficiency portfolio?"

ACEEE used the term "Intelligent Efficiency"—"the synergistic effect that emerges from the interaction between information and communication technology."

"Technological innovation is an impressive driver of efficiency gains," according to the report, "over time, however, it has become clear that without some understanding of human factors, the full potential of energy efficiency will never be unlocked."

The report suggested that smart meter pushback may stem in part from the one-way communication of end-use data to the utility. Further, the more devices that are added to a residential household's load, the murkier the use picture becomes and the more difficult it is to control use.

Whether an end-user is likely to save energy using real-time information feedback depends on their "sensitivity," a term that may or may not be akin to the "early adopter" motivation. Devices providing energy use feedback may aid the process, depending on their design and whether that facilitates the actions taken by end users. Engagement with energy use information feedback may be helped or hindered by "household dynamics." Persistence of behavioral changes remains a key area of enquiry.

One conclusion of the report that's relevant to U.S. utilities today: "The question remains of whether efforts to entice the 'less sensitive' to save through feedback programs are a good utility investment. Based on the very limited data from the nine pilots ... the cost of providing real-time feedback remains high."

The ACEEE report found that results ranged from zero to nearly 20 percent savings, with average savings of nearly 4 percent. An anomalous finding, not included in those numbers, came from Northern Ireland, where real-time energy use was combined with prepayment meters. Users in that pilot tended to be low income, the cost of their energy was high and the market offered no competition.

The study identified three "points along the human-technology spectrum" that pilot studies should take into account: technology adoption, effective installation and continued use. One solution: require professional installation of feedback-related devices and training in the devices.

The three U.S. pilot programs examined by this report derived from studies at Commonwealth Edison, Stanford, Calif. and a Google PowerMeter experiment (as you probably know, Google has now folded it PowerMeter effort) and a Cape Light study on Cape Cod.

The parents in the audience will not be surprised to find out that households with several occupants—in particular, people under the age of 15—found energy savings difficult. The presence of pre- or early teens made it difficult to maintain a consistent strategy.

Interestingly, participants in the three U.S. pilots said that energy use feedback made them more aware of their consumption, but it did not make them more aware of the resulting actions they took as a result of that information. That suggested to the report authors that actions that led to savings may have become unconscious habits. Also, those who saved the most energy were more likely to spend time on an online portal that provided energy use feedback information. "Therefore, the frequency and type of interaction with the home energy monitoring system appeared to be a key to getting energy savings," the report stated.

As I said, have a look for yourself. One challenge researchers have found is drawing conclusions across studies that aren't uniformly designed to produce aggregate insights. But the ACEEE report does focus on questions that many in the power industry are asking.

Phil Carson
Intelligent Utility Daily

Related Topics


Pretty dismal results as indicated by this study

The bottom line? Except for a prepay system that's been in place for decades in Northern Ireland, the average energy savings that results from real-time feedback and in-home displays is only 3.8% That's not a very strong case for smart meters in my opinion. It's pretty clear that utilities--or 3rd parties--are going to have to do better at extracting value out of high-resolution energy data for their customers.

I've been at this in one way or another for close to twenty years and have come to a few conclusions:

1) Kilowatt hours, profiled or not, mean very little to most people, even energy managers as I found out long ago. Dollars do get everyone's attention however. Normalize all energy data to dollars (interval usage, alarms, potential savings, actual savings, etc.) and you speak a language everybody can understand. Imagine how our energy usage behavior would be different in electric meters were inside the house and denominated in dollars rather than kWh?

2) Real-time feedback is overrated. We are all being drowned in messaging to the point that it's just becoming background noise. Maybe it's just me, but I don't have the spare bandwidth to respond to much of anything in real time. I need to look at the data, think about it, and plan a course of action, whether it's a big sale at Home Depot or a hot stock tip. Energy consumption data is no different. Showing people current kW or a kWh profile is not very actionable. Showing them a daily or weekly e-mailed cost profile, with specific analysis as to what it indicates, and with specific recommendations on how to reduce it, is actionable. And, those recommendatations should be prioritized by payback time, whether they involve behavioral change or an investment in technology. It may take a week to respond to the "message" but the net benefit is likely to be longer-lasting.

3) Positive feedback loops have to be shortened. If I do something smart to save energy today, I probably won't remember it five weeks later when I get my bill. If I can't get a weekly bill, I should at least be able to get a daily or weekly e-mailed report reinforcing the benefits of my positive actions.

4) People need yardsticks, but group behavior should not be the standard against which progress is measured. Although it's a lot easier to aim for a target rather than just improve, people should set their own goals, whether it's in weight loss, annual salary, or energy consumption. People can and do raise their own bars, but there's less motivation to do so if the target is just to be better than average.

5) Not everyone is motivated by the same things, so the challenge is to create incentives that can capture different drivers. Some people are unmotivated by the prospect of altering their habits to save a little money but will go to extraordinary lengths to "win" the same amount in a demand reduction program. Others are motivated by a desire to save the planet, or to reach some extreme level of performance, as with the subculture of Toyota Prius "hypermilers." To get maximum value from energy data, providers have to present the data--and the feedback--in a manner that addresses more than just the goal of "use less."

6) Finally, you can empower people with actionable information, but it's all for naught if the rate structure isn't conducive to rewarding the right behavior. The rewards should encompass the psychic as well as financial and must be materially significant. Saving 3.8% off an electric bill is not very exciting for most people, but saving 25% probably is. Making that happen is going to require some significant changes in the way the utility industry prices their product however. We need a major effort to internalize more of the external costs of our energy consumption and have prices reflect the true cost of providing service, as transparently as possible.

Greg Tinfow
Energy Informatics LLC

Energy use versus productive consumption

It seems to me that one of the problems with EE/DSM is that, particularly with respect to consumers, the goal seems to be simply use less.  Why?  The assumption is made that if my household uses more electricity, that is must be wasteful.  There seems to be no accommodation for productive consumption.  We know that there is a strong correlation between energy use and standard of living.  If my use of electricity, for example, is productive, then why should I not use more?  Watching the meter spin is not productive.  Perhaps energy use needs to be normalized the same way we use weather normalization for heating but using some figure of merit based on productive use.  The goal should be having machines (electronic or otherwise) that are most "efficient" in their use of inputs and generating facilities that produce the most output at the least cost and with the fewest impacts.  It is the constant mantra to simply use less that seems at odds with other trends in technology.

Rich Mignogna

Good point


I'd have to agree that the mantra of "use less" doesn't necessarily make sense. With dynamic pricing, end-users will respond as they see fit, paying more for peak use if they can afford it. Others will avoid peaks because they can't afford it. Reducing overall use may answer an integrated utility's capacity problem; should that matter to the end-user?

Do you have an example to flesh out what you mean when you write: "Perhaps energy use needs to be normalized the same way we use weather normalization for heating but using some figure of merit based on productive use."

Regards, Phil Carson 

Another Step on the Journey

Good article and I will take a closer look at the report. This points out several of the areas I have working with utilities on to try to get them to move forward with, and better success for the masses is definitely feasible if approached in the right manner.

Specifically customers react and engage based upon thier "behavior" and lifesyle, not based upon the utility operational challenges and benefits. And that reaction is not based upon a few $$s saved in the month it is how they precieve this benefits and how they fit into their lifestyle. And as your article points out doing this on a larger than pilot scale for the masses is costly unless looked at from as an enterprise solution basis.

So given that utilities are trying to gain knowledge in the customer engagement and relationship maintenance area, but have very little real background to work from at this  point, and they do not have the systems to support any sophisticated actionalble customer data what is the best approach to move the ball foward?

From my perspective is does not mean utilities need to go overboard and get in to behavorial science type programs or sophisticated persona classifications because this is based upon retail market dynamics and is as much hype as value. In addition you can't just meet the needs of the few "sensitive" people that are out their and gain the returns utilities require unless you reach more of the masses. This can be done in a cost effective manner if utilities effectively overaly technology that is proven to assist with customer information management and first apply that to their core business processes across the silos of the organization that want to commuicate with customers.   

Energy use feedback: worth pursuing on mass scale?

Ya think?

Human behviour is well understood, just not always predictable.  How many of us grumble at the price of gas as we fill up the SUV? But, in fairness, not everyone can make a Volt or a Prius work for them - too many kids, hauling loads, commute is too long.  The feedback is there every day as you drive past the pump in near real time, but it does not / cannot change the human behaviour overnight.

Walking around the house turning off the lights is becoming less excersise as we substitute higher efficiency lighting.  Savings from turning down the thermostat get offset by plugging in a space heater in the occupied room (unless your primary heat is resistance heating). 

Do we retrofit our heating system when the price spikes? Do we replace our car with a higher efficiency model whan gas hit $5? Behaviour / habits are hard to change even for the highly committed.

So... we let big oil get away with market economy, why not electric (and for that matter gas, water, etc)?  OH, right, we do!  Consumers will want to know, "What's in it for me?" Give the consumer what they want: feedback is ok, but they really need incentive.  After some fits and starts, there WILL be a study that predicts ways to save and then let the consumer decide if and when their systems and habits need to change.  Make it worthwhile, not just a sacrifice.  The incentives worked for big industries 20+ years ago.  Plant managers made choices based on the economics.  Consumers will too.

I hope the ACEEE study didn't cost their sponsors a lot of money.  Seems pretty intuitive to me.

Wally Binder