Motivating metrics for household water-use feedback
Graphical abstract
Introduction
The recent advent of smart water metering enables the collection of higher resolution water consumption data than under conventional metering. By logging usage at highly frequent intervals (e.g. of seconds or minutes), smart metering creates a far larger database of water consumption according to time of use and allows for analysis of end-use patterns. Deployments of smart water meters are growing rapidly in many developed countries, with over 250,000 units either planned for, or already in operation by Australian water utilities (Beal and Flynn, 2015). The water industry is therefore now beginning to embrace new levels of data and means for its analysis and communication. However, while many water utilities are focusing on the benefits in terms of internal planning or network efficiency (see Boyle et al., 2013), not all implementations have been extended to provide more detailed water-use feedback to householders. An underexplored opportunity therefore remains for collected data to be analysed for translation into significantly more detailed water consumption feedback for household consumers.
Householders may not be entirely aware of their water usage nor of opportunities to save. Therefore, consumption feedback may help to address these knowledge gaps. Recently, Beal et al. (2013) found that householders which received feedback on how their water consumption compared to other homes could more accurately match perceptions of their water consumption to their actual usage. Beyond raising awareness, the provision of information on opportunities to save can further improve householders’ knowledge of how to save; and knowledge and water-saving skills have been found as important determinants of water conservation behaviours (Corral-Verdugo, 2002, Corral-Verdugo et al., 2012).
The provision of consumption feedback has been shown to help customers monitor and more effectively manage resource use (see Abrahamse et al., 2005), and positive impacts through feedback have been found via research across a variety of domains, including household energy and recycling behaviours (Schultz et al., 2014). Through raising awareness and changing consumption behaviours, feedback on household water-use also has the potential to contribute to more sustainable consumption patterns and practices, and potentially reduce the demand on constrained water resources.
A number of recent research studies, which particularly involve smart water meters, have shown that various different forms of water-use feedback can effectively reduce household water consumption (Wetherall, 2008, Doolan, 2010, Willis et al., 2010, Erickson et al., 2012, Anda et al., 2013, Fielding et al., 2013, Joo et al., 2014). These studies tested for quantitative changes in water consumption and reported savings ranging from 5 to 10%. The recent Home Water Update (HWU) study, to which this present work relates, also signalled initial savings of 8% through feedback of detailed water and end-use metrics in paper-based reports, although with the moderate sample size and measurement periods, these savings could not be confirmed as statistically significant (see Liu et al., 2015, submitted for publication).
Despite the reports of overall quantitative savings and some documentation of specific changes in behaviour, relatively few feedback studies have conducted a detailed post-intervention evaluation to understand how householders engaged with the different forms of water-use information provided. In the few cases where comprehensive post-trial surveys and/or interviews were actually conducted (e.g. Doolan, 2010), householder information preferences were investigated only briefly among a wider set of trial related issues. Erickson et al. (2012) took the novel approach of investigating pilot usage of an online water portal via internet click-counts, which showed where users were spending most time, and thus sheds some light on user interactions with available data and functions. However, both in this and most other studies, it is unclear what additional or alternative types of information householders might have preferred. It is furthermore uncertain in each of the studies which pieces of information specifically motivated changes in behaviour towards water conservation.
Fielding et al. (2013) took another innovative approach by using study sub-groups from the outset to compare impacts of customised end-use data against procedural information and against descriptive norms on ways to save. The distinct feedback groups obtained significant savings relative to the control group, but not relative to one another (Fielding et al., 2013). The study scope did not include an evaluation of how householders within the different groups interacted with the different forms of feedback provided, or their other preferences.
Parallel work from within the field of sensor technologies (Froehlich et al., 2012) represented a first attempt to obtain wider insights into householder information preferences for more detailed water-use information. However, the study’s use of prototypes meant participants were unable to engage with any real, customised, water-use data reflecting their own household consumption.
Research is currently underway to explore alternative ways of analysing household water usage data collected via smart water meters. This includes improvements in end-use categorisation via the automation of flow trace analysis using pattern recognition and event probability functions (Nguyen et al., 2013); as well as the detection of water-use signature patterns, which cluster meter reads on the basis of calendar dates, times and volumes to identify target behaviours for reducing consumption (Cardell-Oliver, 2013). Within the energy sector, there have already been further advances in terms of automating smart meter data analysis and producing customised household energy reports for mail distribution (e.g. Laskey and Kavazovic, 2011). This suggests that further developments can well be expected within the water sector to facilitate the creation and provision of more detailed consumption feedback to householders.
Amid these growing prospects for more advanced water-use feedback, our review of existing literature found a clear need for additional work in this emerging field to understand the forms of water-use feedback that householders might prefer and could help motivate savings. This remains essential in shaping the role for feedback via smart metering to promote water conservation. While scarcely researched for the water sector, this need has been more clearly recognised within the residential energy sector, where detailed feedback is also more established.
Energy-related research has specifically cautioned that care must be taken when choosing the information the consumers are given and with the way in which it is presented…that the information provided to consumers is relevant and enables them to make sustainable decisions about their energy use and so the study of what kind of consumption feedback consumers understand and prefer is required (Karjalainen, 2011). Within the energy sector, insights from behavioural science (e.g. the fields of influence and persuasion) have also been recommended to help develop the right ways to present the right information for maximum impact (Laskey and Kavazovic, 2011). Such considerations also need to be applied to the design of water-use feedback, whereby well-designed feedback can help householders to better monitor their consumption and ultimately save water.
Our review of the existing literature highlighted two important knowledge gaps to be addressed in this paper. First, the existing literature lacks a comprehensive overview of the extensive water-use feedback possibilities specifically made possible via smart metering. Second, more work is required to understand the specific forms of information that householders prefer, and could help motivate behaviour changes toward more sustainable water use in the home.
The research topic is depicted in Fig. 1 below, where one circle represents the set of household water-use feedback information that is technically possible; while the other circle represents the set of informational interests and behavioural responses of householders. The area of overlap is designated with the term motivating metrics and corresponds with information that is possible, and is additionally of interest and/or motivating to householders. The key objectives of this research are to:
- (1)
Develop a conceptual framework presenting the feedback possibilities enabled via household smart water metering, which therefore clarifies the first circle in Fig. 1; and
- (2)
Present the results of an evaluation and householder preferences survey conducted in conjunction with the Home Water Update (HWU) smart metering feedback trial conducted in NSW, Australia (2012–2014) (see Liu et al., forthcoming), which contributes towards understanding the second circle in Fig. 1 and attempts to direct discussion in the field towards the area of overlap. Here, the main aims are to:
(a) Explore how householders evaluated and valued the detailed water-use information provided in the HWU feedback trial;
(b) Explore what types of water-use information/metrics householders prefer and find useful/motivating.
The HWU study’s distinctive feature was its provision of feedback involving a range of customised metrics of household water and end-use consumption not previously provided in other research. Therefore an evaluative assessment by these particular study participants allows for a potentially more informed householder evaluation of different forms of water-use feedback. While this research is a small-scale study and cannot provide the definitive answers, the results represent a valid contribution in demonstrating the variety of responses that exist among householders and uncovering some of the key considerations in the design and presentation of more detailed feedback. The mixed method approach further combines quantitative responses from the survey with qualitative insights from the interviews, and thus explores the issues from multiple angles to gain a more accurate picture.
The rest of the paper proceeds as follows. Section 2 presents a framework for smart metering water-use feedback possibilities, and is used to highlight knowledge gaps requiring further exploration. Section 3 describes the study research methods. Section 4 presents the HWU evaluation survey results, further complemented with illustrative quotations from participant interviews. Section 5 summarises the key conclusions with a discussion of the study implications for smart water metering deployments, limitations, and directions for future research.
Section snippets
A framework for water-use feedback: content, medium, frequency and duration, and context
In order to understand the opportunities and level of interest among householders for different forms of detailed water-use feedback enabled via smart metering, a guiding framework is required to help conceptualise the problem. In this section, a framework for water-use feedback is introduced. The framework emerged through a review of the characteristics of feedback described in related energy and water sector literature (particularly Fischer, 2008, Froehlich, 2009, Karjalainen, 2011), and a
Background to study
The study forms part of an Australian Research Council Linkage Project investigating the role for smart metering in a sustainable urban water future. The study location is Tea Gardens/Hawks Nest in regional New South Wales, Australia. The suburbs are serviced by MidCoast Water (MCW), a council-owned water utility based in Taree and Forster. Smart water meters were previously fitted to 141 properties in 2009, from which households were recruited to participate in the Home Water Update (HWU)
Results
This section reports on the findings from the HWU evaluation survey and telephone interviews. All reported percentages are from the evaluation survey. The survey questions and results discussed here can also be found in Table 1. Additional insights from interviews and quotations that illustrate participant views are provided where relevant.
Brief summary and discussion of results
This paper presented an overview of the extended possibilities for detailed household water-use feedback enabled via smart water metering. The need to understand householder preferences and responses to determine motivating water-use feedback was highlighted. The framework in Fig. 2 thus illustrated important feedback design considerations and areas of research that might guide future implementations of smart water metering and water-use feedback programs. Dimensions of the framework were
Acknowledgements
This research was supported under the Australian Research Council’s Linkage Projects funding scheme (LP110200767). We thank Graeme Watkins and Chenxi Zeng of MidCoast Water for access to the study householders, whom we also thank for kind participation in the study. We also acknowledge contributions from Thomas Boyle and Candice Delaney to the design of the HWU study. We also thank the reviewers for their helpful comments and suggestions for improvement on a previous version of this paper.
References (66)
- et al.
A review of intervention studies aimed at household energy conservation
J. Environ. Psychol.
(2005) The theory of planned behavior
Organ. Behav. Hum. Decis. Process.
(1991)- et al.
A novel mixed method smart metering approach to reconciling differences between perceived and actual residential end use water consumption
J. Clean Prod.
(2013) - et al.
Smart metering: Enabler for rapid and effective post meter leakage identification and water loss management
J. Clean Prod.
(2013) - et al.
One size does not fit all: averaged data on household electricity is inadequate for residential energy policy and decisions
Energy Build.
(2013) - et al.
Quantifying the influence of residential water appliance efficiency on average day diurnal demand patterns at an end use level: a precursor to optimised water service infrastructure planning
Resour. Conserv. Recycl.
(2012) - et al.
An experimental test of voluntary strategies to promote urban water demand management
J. Environ. Manage.
(2013) - et al.
Understanding behaviour to inform water supply management in developed nations – a review of literature, conceptual model and research agenda
J. Environ. Manage.
(2009) Consumer preferences for feedback on household electricty consumption
Energy Build.
(2011)- et al.
Development of an intelligent model to categorise residential water end use events
J. Hydro-Environ. Res.
(2013)