Nudging energy behaviours: Combining the Internet of Things and behavioural science to help residents save energy

Housing Advisers Programme case study

2021/22 cohort 

Set in the wider context of the current cost-of-living crisis and rising energy prices, it is important to offer advice and assistance to council residents around their energy use and costs. The council worked with DG Cities to conduct extensive research with different Internet of Things (IoT) technologies on the market that address issues relating to energy use and the cost of living crisis. A supplier was chosen to provide a smart energy monitoring device which recorded energy use by device and in real time, and delivered this to participants with a user-friendly phone app for live usage, usage tracking and goal setting, and monthly feedback letters. 

The project objectives were to support council tenants to increase their energy efficiency awareness and technological literacy and to better manage and reduce their energy usage where possible. To deliver these objectives, the team developed a monitoring and evaluation approach, with a Theory of Change, developed and refined a user journey and complemented the Sense device and app with a) a behaviour change strategy and b) setting up an ‘Energy Community’ group to enable participants to share and connect with each other, as well as to receive the behavioural interventions.  

In the 12 months leading to March 2023, electricity prices in the UK rose by 66.7 per cent (ONS) leading to growing concern around energy bills. The Royal Borough of Greenwich partnered with their innovation partner, DG Cities, to identify a series of innovative, IoT technologies that could address these concerns and deliver support to residents around their energy usage and costs. By combining the use of such technology with a behaviour change approach, centred around nudges and other behavioural interventions, we sought to aid residents in developing a greater understanding of energy use, associated costs and ways to reduce them, and connecting residents to one another to support shared learning and increase motivation/engagement. 

We conducted extensive desk research around different IoT technologies on the market that addresses issues relating to energy use and the cost-of-living crisis. We considered a variety of solutions, some of which included smart heating controls and environmental sensors. We defined our criteria as a solution that was simple to use and unobtrusive, and that could be deployed quickly and easily. We wanted to procure devices that were as accessible as possible in order to encourage sign-up and ensure people could use them effectively.   

We partnered with Sense Inc, a US company that provide smart energy monitors powered by AI and machine learning. The devices use AI and machine learning to identify different devices within the home, measuring and recording their energy usage in KwH and providing estimations for how much the devices cost to run. The device is linked to a smartphone app that is downloaded by residents, where they can view their energy usage in real time.

The app delivers monthly reports to the user, providing information such as total energy usage and cost, trends in energy usage over time and reporting on the energy performance of certain devices. The Sense monitor is unique on the market and its user-friendly app interface is designed to be accessible to users of different accessibility needs. The device is relatively small and is installed directly into the resident’s fuse box. Once installed the device links to the smartphone app and requires no maintenance. Sense provided us with 40 energy monitors as part of the trial. 

Alongside the devices, we developed a behaviour change strategy for residents, centred around an ‘Energy Community’, a WhatsApp group for participants where they could connect with others on the trial and share experiences and information. We have delivered tips and advice around reducing energy use, tips for using the app and getting the most from it, as well as setting a series of nudge challenges based on trends we were seeing in their energy usage data. 

The behavioural change strategy was developed with an extensive literature review of best practices, from well-established theoretical models (including Nudge-plus theory and the COM-B model) and in conversation with academics in the field.  The resulting strategy was focused on the principles of: the role of social norms, the value of personal goal-setting commitments (and feedback), the use of trusted messengers, and of salience and attention to the app interface. 

An overarching element to the trial was a monitoring and evaluation (M&E) plan, based around the Council’s objectives relating to a reduction in carbon emissions and the road to net-zero, as well as cost savings for both the Council and residents. The M&E plan was developed with a Theory of Change approach with accompanying logic model, and with a series of measures (for identified impacts and outcomes) including: change in energy usage, increased resident awareness and confidence of energy management practices, increased resident intentions to improve energy efficiency in the home, and improved resident confidence in digital technology. The ongoing monitoring includes analysis of the Sense inc. data tool with device data down to the minute level and by device, and with participant surveys for pre, during and post-project for evaluation purposes. This will deliver a series of findings and determine factors for success and recommendations for any wider roll-out. The upfront survey captured a range of household information, resident wellbeing ratings, their views on energy use and climate change, their expectations for how useful different elements of the project would be, and their aspiration for how much energy they would like to see (given the level of changes Sense have seen elsewhere). 

There are a number of emerging positive impacts and outcomes as a result of this project: 

Energy Efficiency Awareness: 

Throughout the trial, residents generated a greater understanding of their energy use, specifically around individual devices. On average, each home identified over seven individual devices ranging from microwaves and ovens, to hoovers and space heaters. Once a device was identified by the Sense monitor, the energy usage and estimated code of that device was recorded and reported back to the resident in real time, as well as being tracked for daily/ weekly/ monthly views, improving their knowledge of electricity usage in their home by providing more detailed, granular information.  

The survey findings are emerging and will be important to understand if and where awareness has increased. However, in advance feedback is captured through the Community Group, for example one resident has reported to us;

I cannot get over how clever this gadget is! It is so interesting learning how much power I consume.

Technological Literacy: 

One of the main barriers we identified was technological literacy and the confidence of residents to engage with and benefit from new technology. This project gave the opportunity for residents with lower technological literacy to experience new technology, in particular the more advanced use of smartphones, and gain confidence and understanding. The app required a level of resident engagement and included things like setting up an account, verifying an email address and labelling devices. Some resident feedback indicated that their confidence in using tech had grown and they had benefitted in this way from the trial. This is another impact which will be tested across the participant group through the next survey data collection period.  

Improved management of energy use, and electricity reduction where possible: 

There are emerging findings on energy use reduction. For example, a recent nudge challenge to encourage participants to look at devices that are ‘Always On’ has seen a reduction in electricity use of up to 20 per cent for one household, and 4 per cent for the group overall compared to the weeks prior. Importantly, this will continue to be monitored and is an emerging finding only. Overall, nearly half of participants have seen a significant reduction in their average weekly electricity use so far, since they started on the trial. 

Furthermore, the ‘Energy Community’ has seen several residents report how they are finding this management and app use, including the following quotes: 

This is fun. I have gradually turned into a device detector…

I am up to 8 devices found. The machine is definitely not British; it only just found the kettle.

Behaviour change - I now ensure that electrical devices are on only when in use. No more on standby. Hoping to achieve my monthly target.

Comprehensive data analysis across and within the participant group will be continuing across the next stage as more participants have the device for a number of months and to control for factors such as seasonality and household type. DG Cities look forward to being able to report on the project outcomes, as well as recommendations, in the next stage. 

The trial is continuing, following further support from Greenwich Council to continue to work with residents and gather lessons learned. Residents are still taking part in the Energy Community and receiving support as well as energy saving tips and advice. We are also continuing the data monitoring and evaluation phase to draw further conclusions and continue to deliver bespoke advice through the Energy Community. 

In the future, DG Cities will be developing a proposal using evidence from the trial to provide a framework for further roll-out of the trial, providing the Sense monitors to a wider group of residents. This report will include learnings around recruitment and evidence of the positive impact of the trial. 

Recruitment has been the main challenge associated with this project. From the outset it was challenging to gain access to a wide pool of residents for whom this trial would be appropriate. In order to recruit for a project such as this, involving new technology and installation into people’s homes, it is important to try and reach a wide target audience and to deliver an effective user journey with relevant touch points and information sharing as needed by individuals.  

We tried a variety of recruitment methods including door knocking, social media, distribution of flyers and attending community events and regular group meetings (like coffee mornings) on social housing estates. While this generated a lot of initial interest we found that there are often barriers to residents responding to follow-up emails or completing requested forms (such as consent forms), particularly when there was a period of waiting between the first contact and follow up.  

We found that the most effective recruitment technique was a face-to-face approach where we could speak to residents directly, talk to them about the project, demonstrate the app and answer any questions. This style of recruitment was the most effective, particularly when the resident was able to sign-up on the spot. As part of this approach, we found that hosting drop-in sessions at community hubs on specific estates offered an informal, neutral setting in which to engage with people.      

 

Sense is new to the UK market and the application is constantly learning about UK devices and their consumption statistics. This trial is helping to further improve the product and service for wider rollout. Transparency with participants up front has been vital in helping residents engage and support improvements in the overall service. 

We have also found that the Energy Community WhatsApp group has maintained a good level of engagement by its participants, with the majority posting messages at least once in recent months and with group responses every week or two. This is helped with our bi-weekly information and nudges, and that people have been willing to ask, share and support each other. 

Lavinia Parker, Strategy Officer: [email protected] 

• The energy reduction trial website
• Sense's website