Seeing is believing: is data visualization the low hanging fruit of energy reduction, or too good to be true?

By Julia DeIuliis

What if energy consumption could be reduced by ~2% without developing new technology, adding regulations, installing new equipment, or disrupting communities- and with minimal investment? If 2% sounds insignificant, remember that it would save households over $3 billion on electricity.

This is the promise of data visualization. Energy reports can display the absolute level of energy usage, as well as comparisons with historical usage and peers. For example:

Exhibit 1: OPower report

Exhibit 2: OPower Energy Report (detail)

Exhibit 3: Nest* Energy Report (detail)

According to the Environment Defense Fund, these visualizations can reduce usage by 1.8% on average- and if targeted at households most likely to show reductions, average reduction rises to 6.5%. A comprehensive review of 20 studies on energy usage feedback between 1987 and 2008 showed average savings between 5-12%. OPower claims their reports are one of the most cost effective energy interventions:

Built on decades of behavioral psychology research, data visualization works by both showing the potential financial savings from demand reduction and introducing social pressure with peer comparisons. While granular usage feedback requires smart meters, which have received some customer backlash over privacy concerns, all standard utility bills have the potential to show customers these basic metrics: their bill versus the average community and past bills, and the potential savings from small reductions.

Moreover, this information can be delivered via mobile apps or through SMS, indicating the potential to work in emerging markets. Of course, data visualization would work best in relatively developed emerging markets, such as Bogota, that have a somewhat established energy grid, as opposed to urban slums.

While this approach holds promise, I have two key concerns:

1. Impact may reach a “ceiling” as customers with above average usage reach the community norm, and then feel less pressure to lower their energy consumption. As opposed to sensor-based systems, which can dramatically lower consumption of even contentious customers, data visualization only brings people towards the middle and enables small changes.

For example, one report estimates that nearly 5% of electricity in residential homes is “leaked” by powering devices in “standby” mode, such as TVs turned off. Yet, few consumers will bother to unplug appliances, even when aware of the potential savings. Smart outlets provide a much better solution to tackle this issue.

2. Who is going to pay? Customers seem unlikely to pay upfront for an energy report. While upfront costs are low, a cost is still implied. If utility companies pay for the report, they must assume that they’ll be able to serve more customers with the same assets. For this incentive, governments must hold utilities to high levels of efficiency, before greenlighting the development of more energy plants.

*Full disclosure, I worked for Nest over the summer, and plan to return after business school.



8 thoughts on “Seeing is believing: is data visualization the low hanging fruit of energy reduction, or too good to be true?

  1. Pingback: Most Commented through Friday, March 8th | Sustainable Cities: Urbanization, Infrastructure, and Finance

  2. Julia, you bring up a very interesting topic. Studies have shown that “Nega-watts” (non-consumption) is the most environmentally friendly and cost effective way to become more energy efficient.

    In our Launching Tech Ventures class last week, I met one of the early employees and current Product Managers of Opower. He described how they used gamefication and other behavioral psychology (one of their first advisers is Robert Cialdini who wrote the popular book Influence) mechanism to encourage people to consume less energy..

    1. In response to your first point, Opower is experimenting with ways to use leaderboards and other features to continue incentivizing top users to save — rather than charts comparing them with the average user, it would instead compare them to the top 20 users. Ultimately, the bigger gains in energy consumption decreases will come from those who currently use more energy (the marginal cost theory).

    2. Utility companies face regulation that requires them to reduce energy consumption, and they are the ones signing up for Opower services. There needs to be continued governmental standards to encourage utilities to use these services. For many utilities it is also cheaper to decrease consumption rather than invest in creating new power plants.

    Overall, Opwer has seen tremendous success with their existing platform of encouraging negative energy consumption. However, as they expand into smart grids and other complementary areas, they are currently facing increased competition. We’ll see what happens!

    • Thanks for the thoughtful response. I’m clearly a big fan of Opower. I think their track record of deals and relationships actually provides a pretty strong competitive advantage. It’ll be interesting to see what happens once they expand into new geographies. I’m so incredibly interested to see what happens when they expand into adjacent markets that aren’t based on data visualizations.

      In response to your point #1, and the idea of a leaderboard, I still remain pretty skeptical. There’s certain things people just won’t do: no one is going to go back into the house when they realize they left the lights on, and they’re not going to unplug their TV every time they’re done using it. Most of us don’t even bother to recycle!

  3. By John Macomber

    An informative and provocative piece about visualization at city scale is here:

    “The way we visualize and compare cities says much about our understanding of how they work. As part of our ongoing exploration of what makes a city, we wanted to survey how people are using data to describe the political, geographical and social realities cities face. Below, we’ve compiled some unique visualizations. Some of these center around cities in the common sense of the word, focusing on large urban areas, but we think these images as a group help expand the understanding of the diversity of all kinds of municipalities…. As we continue to engage in open data work, we hope to contribute to the kind of information that powers these visuals and help create the resources for the next wave of municipal understanding.”

  4. By John Macomber

    The voluntary modification of individual behavior is one potential benefit of data visualization, as the original post and the first response both explored. Can it also help collaborative consumption?

    Is there added upside in collaborative consumption that is made easier or more compelling via data visualization? This would be a situation where users could use visual cues to navigate data and not only shift the quantity of resources uses, but also shift the timing. For example if all the people in the hotel are willing to stagger shower times, then there is less need for capital expenditure on a giant hot water tank and the hotel can make hot water at times when electricity or gas are cheaper. Or, the Zipcar reservation screen could show bar charts by car for both the hours when cars are reserved or available but also change their pricing to make less busy hours get more usage…or even to allow you to buy a reservation from someone who already had one, if you just had to get the car right this minute. Less steel parked, more mobility from the same steel.

    Traffic optimization platforms are a third way to combine data visualization with geo-coded information and crowd sourcing in order to get more utility out of one fixed capital investment: the roadway. Waze is a good example of this.

    Links: Collaborative Consumption (Rachel Botsman in HBR Blogs)
    Zipcar how it works
    Waze “Waze is the world’s fastest-growing community-based traffic and navigation app. Join other drivers in your area who share real-time traffic and road info, saving everyone time and gas money on their daily commute”

    • This is an interesting point, and one I hadn’t fully appreciated before: the ability of improved data visualization to influence all of our decisions.

      We all make decisions all the time, based in some part on whatever information we have available. Data visualization can make that information more clear, more compelling, and more actionable. Some of the actions we know are good for sustainability seem too difficult, too inconvenient. But- to use your example of car sharing- what if we got a pop-up notification that cars were available a block away from where we were going?

      Of course, that would require us to acknowledge something we don’t often want to acknowledge: that we’re not as good as making decisions as we think we are. We don’t just require good information- we need it to be presented in a better way.

  5. I had seen the Nest thermostat in some architecture magazines and installed in some places but I had no idea that it was more than just a nice design. I find quite interesting the whole concept.

    Regarding your concerns, I am not sure about 1. I personally wouldn’t feel very compelled about “competing” with my neighbours (which I might not even know), or at least I would lose interest very soon; it looks more like an anectodical “fun” app. I find it not very convincing because of the relative comparison… you can be the same environmentally-conscious person living in a more-or-less sustainable city like Cambridge or in a sea of McMansions in Houston, and your position in that scale is probably different. For me, in the long term, the key is absolute energy savings (Exhibit 2 and 3): it is much easier to be sustainable when it also makes economic sense; in this case to reduce electricity bill. Even most of the reviews in Nest’s website focus heavily on the bill savings they have had after adopting Nest and receive an input on their energy performance.

    As for the second concern, I see it in two levels: one in terms of policy and regulations, and a second layer in individual behaviour.

    I think governments should be able to require some kind of energy reports, at least for sale transactions. The issue I see is that this can be understood as a new “tax” when selling/buying a property, and that is when the Real Estate market in the US is not fully recovered. However, in the big picture of things it is probably just a few hundred or thousand dollars in the total value of a property and the potential customer could make a more informative decision in order to save in future utilities for years or decades. No one finds surprising that every new car model has to pass pollution tests specifically for the US market, and those costs will be transferred somehow to the customer (of course, a car is an industrial, standarized product while every other property would require a customized energy test). The EU has implemented recently the Energy Label (it has been for a while in cars and appliances) in new buildings and sale/buy transactions of 2nd hand properties. I think it is an useful way to regulate the market and accelerate energy efficiency in construction; properties with good ranks are likely to increase their value (the same way that LEED-certified properties increase their value in average more than the real costs in energy consumption reduction) and those with bad marks to reduce it. Owners with “bad ranked” properties will feel more compelled to improve the energy efficiency of that property, if only to raise its value or appeal in the market.

    There is a second layer, for individual behaviour, and I think it is in this case where systems like Nest make sense. No one can control whether you keep the TV on standby (even if you know that this is wasting energy), the same way that you can drive an hybrid car in a very aggresive way (thus increasing the fuel/energy consumption) and using no public transportation even if available and convenient. Like in the first concern, I think it comes back to savings and economics: when the average customer feels that a system like Nest is a real way to help him or her to save energy and reduce their bills, they will adopt it at the right price (when those savings outpace their investment). I wouldn’t be surprised that there is also a behavioural pattern and in some major purchases, customers prefer to pay a bit more even if they are not going to recover the full savings in the long term (I am thinking about building a new house and using a thermostat $200 more expensive than another, which is not much in the total cost). It reminds me of the extreme popularization of diesel cars in Europe (which in places like France or Spain are like 70% of new car sales today); or to a lesser extend (in terms of current market share), hybrid cars. There are reports stating that many customers will never recover the price differential among the same car in petrol or diesel/hybrid versions (a diesel car can be easily 2 or 3,000 euros more expensive than the comparative petrol version, and usually more costly to maintain; so it makes sense only if you drive a lot), yet when buying a relatively expensive item like a new car, many Europeans are willing to pay a higher upfront cost with the confort of thinking in future fuel savings.

    • Daniel,

      Thanks for the thoughtful comment.

      I had a few specific reactions:

      1. I think the “competing with your neighbors” element is more compelling than you give it credit for. Not necessarily because of the desire to “beat” your neighbors, although I think that’s significant for some folks; rather, I think it’s compelling because it shows you what’s possible. It shows you that it’s possible to lower energy use by 20%, because your neighbors are doing it, and they’re unlikely to be people who never use AC or keep the lights off all the time. I hear your point about the absolute usage being compelling- but I think its even more compelling to know that not only can you save $100, but your neighbors already are.

      2. I love the idea of a government mandated energy report- although I hope if consumer demand is high enough, companies would provide it voluntarily.

      3. I think you hit at an important point: that the timing of a message can be as important as the message itself. When you’re buying a house, an extra $250 for a thermostat seems relatively small. When you’re just going about everyday maintenance, $250 seems cost prohibitive. If energy suppliers focus on addressing people when they’re in the midst of purchases, the message may be much more compelling.

Comments are closed.