In behavioral programs, where small businesses make all kinds of changes, attributing energy savings is not straightforward.
Colin Fraser is a Data Scientist with EnerNOC focusing on experimental design and evaluation for behavioral energy efficiency and customer engagement programs. He has been involved with the design and rollout of several large scale randomized controlled trials for energy efficiency programs with some of the largest utility companies in the world.
Small and medium enterprises in the US spend around sixty billion dollars each year on energy, but remain largely untapped as an energy efficiency market. One exciting approach to deliver cost-effective energy savings among smaller businesses is behavioral energy efficiency, which utilities such as Pacific Gas & Electric and BC Hydro are using. In these programs, small businesses receive personalized feedback and tips about their energy use, often in the form of monthly reports, prompting them to save energy by changing their behavior as opposed to paying for new, more efficient hardware.
"Evaluators could be performing nearly the identical regression analysis, but with the logarithm of energy usage as a dependent variable." – Colin Fraser
Measuring energy savings so that utilities and regulators can compare approaches and decide which to prioritize for funding from limited budgets is critical. When savings estimates are inaccurate, less effective programs might get funded, making the cost of energy savings more expensive for everyone. As a result, there is a lot of scrutiny on how savings will be calculated, and established protocols-which can be tough to change-are usually applied.