Earth Notes: EGC: HaaS Lite WP1 D15 Customer Research
EGC Work Package 1: Deliverable 15: Customer Research, Public Report.
D15: Customer (Social) Research, Public Report
Summary and introduction
Deliverable 15, part of Work Package 1.
It is difficult to engage people in big enough, persistent energy savings.
The essence of this research strand is to uncover messaging and mechanisms that reliably work for normal busy human beings, at home, to engage them in those energy savings and to ensure that those energy savings stick.
Typically people just don't make many of the (energy) savings that they could, even when easy and quick and one-off with recurring significant benefits. So the 'customer'/'social'/'human' strand is essential for a sustainable HaaS business model.
The overall aim of the HaaS Lite project is to find ways to make heating more efficient (in carbon and cost terms) and more pleasant for users. It aims to find schemes that will engage users in the first place (eg the large portion of the population that have never switched energy suppliers despite the ease and immediate savings for most in doing so, or short-term tenants that the Green Deal was meant to help but was too heavyweight for), and that will keep them engaged. HaaS needs to overcome the apathy that normally bedevils measures which are not purely technical and automated. These schemes have to be profitable, and not fall victim to Jevons' paradox [Jevons].
"... if smart metering is to effectively reduce energy consumption there is a clear need to develop and test innovative new feedback devices that have been designed with user engagement in mind."
HaaS Lite is attempting to identify or create such 'devices', physical and behavioural, tied to a financially viable payback.
Apathy, Nudges, Gamification
The ulterior motive of HaaS Lite is to reduce climate change, by reducing carbon emissions from domestic space heating.
This 'green' message is directly toxic to a significant group of people who may see it as attack on personal freedoms. A reasonable and more universally-acceptable proxy, at least in the UK, is money-saving. Or similarly, but not exactly equivalently, avoiding waste. Both are well correlated with carbon emission saving.
However, many people can't be bothered even with something as simple as energy supplier switching, which for most UK householders would be a matter of minutes' work and with recurring and sizeable savings.
Ofgem claims that "comparing and switching supplier or energy tariff can make a big difference to your gas and electricity bills — with annual savings of around £300 available" [OfgemSw]. Also, at the end of 2015 GoCompare's research claimed that "2.5m (15% of) homeowners have NEVER switched energy supplier and are potentially missing out on collective savings of £728m [GoComp2015].
The tabloid view of the world is that not only is the whole thing probably a con (unless it makes your house's value go up), but also user apathy is insurmountable from feckless youth to grumpy old men.
A slightly more sinister, indeed Machiavellian, view of the world, still with traction in high places in the UK for example, is the notion that with the right 'nudges' (almost like subliminal marketing) you can get people to do the right thing. Project CHARM explicitly tested this notion for domestic energy efficiency [CHARM] along with social norms.
Another significant hope is that 'gamification', ie turning everything into a game to harness competitive instincts, will somehow move the punter past apathy into informed engagement and indefinite commitment. In any case people seem fascinated by their digital traces, and that may be a way of engaging them [CHARM].
None of these turn out to be invincible foe nor magic pixie dust, and to meet our aims we have to deal with a fair amount of illogicality and second- or third- order effects, of which Machiavelli might indeed be proud.
Issues Considered in More Depth
Some of the key issues found in existing research and experience are explored here; please see the sources/links for more details.
Market segmentation: is it needed and who should be targeted?
Ultimately, if carbon-reduction targets are to be met, virtually everyone (eg in the UK) will need to improve their home heating efficiency, and all current utilities will need to materially decouple profits from carbon emissions.
However, there are some sectors of the market that have been poorly served by existing measures, eg non-owners often unable to make significant building fabric changes and needing a fast payback of a year or so to make improvement efforts worthwhile financially.
Are consumers motivated environmentally?
This depends in part on the demographic being addressed though there is some common ground: "while there are a few areas where there are similarities in ethical behaviour among a broad range of consumers, in the majority of cases there is no silver bullet; companies need to take a nuanced and tailored approach." [EcoConsum].
This implies that segmentation will be needed to reach/engage a broad section of end users who could use HaaS and save energy, even to achieve similar outcomes across those segments.
Focus on private renters as exemplars
This is especially true of the (growing, soon to reach 25% of the UK market) private rental sector, where typical tenancies are one year or so, and the "split incentives" issue is writ large. (The landlord pays for improvements, and the tenant reaps the rewards with reduced bills, but is often oblivious to energy performance when selecting a property, thus no rent hike is possible to recover the landlord's expenditure.)
A scheme robust enough to work in this segment has a good chance of working in much of the rest of the market, where end users have more money and time and experience, and more commitment to a property too for example.
That is why HaaS Lite has a focus on this segment.
Are people rational: Homo economicus or not?
No [HomoE], at least not economically, and similarly not in other areas.
It was reported at the TEDDINET-Ctech Symposium [TEDCtech] that when giving savings feedback to office workers, smaller-valued numbers representing the same level of energy saving had reduced impact (eg displaying energy use in £ rather than kgCO2 was less effective). But having normalised the numbers to (say) equivalent 'units' then cost/money displays were more effective/motivational than CO2. So, humans are not rational or not numerate, or maybe just not holding conversion factors in their heads. And here at least money seems to beat carbon as a motivator.
Energy savings also tend to have a prolonged effect spread over years, and humans discount future benefits more heavily than is numerically justified. One bird in the hand now is far preferred to two birds in the hand in a year's time, instinctively, whatever the interest rates and risks are.
Are avoiding waste and saving money equivalent?
Benjamin Franklin is reported to have said that:
A penny saved is a penny earned.
But "loss aversion" refers to people's tendency to prefer avoiding losses to acquiring equivalent gains: "It's better to not lose $5 than to find $5" [LossAv].
People are also more motivated by frugality (avoiding waste) than thrift [CHARM].
And there are many studies where if a complex piece of arithmetic is framed instead as a social situation, eg including notions of fairness, accurate or near optimal solutions are often quickly arrived at.
The important point here is that mathematically equivalent views of efficiency, comfort, change, social benefits, etc, may not be at all equivalent in terms of how they resonate with people going about their daily lives and responding instinctively. Framing is important.
Do people know what effective measures are (eg light vs heat)?
In the TEDDINET-Ctech Symposium [TEDCtech], both the i-Stute and APERIO projects noted that people's lack of knowledge and/or 'wrong' folk physics (AKA 'energy myths' [EON2017]) meant that for example, they concentrated on lighting and electric appliances as 'energy' and did not understand heat as important.
Thus we cannot expect people to just know the important things to do to save energy but not (say) risk excess humidity and condensation, and indeed personally at home we find these issues complex. So there will need to be clear but non-annoying steers to get these right. Eg choosing when and how and how long to ventilate right is one key skill with some first- and second- order (ie obvious and subtle) elements [NEA-HIVE].
Anecdotally I know of one couple that always ran their dishwasher in 'eco' mode to save energy and be ecologically responsible, until it gunged up and died, putting them off. Manufacturers of dishwashers (and washing machines) now describe a weekly(-ish) hot wash as a 'maintenance cycle' to get the bulk of the energy savings and keep the machine healthy.
Such subtle and second-order rules of the road may have to be part of the regular engagement and feedback for a system that is going to work long-term.
Do people 'do their bit' and stop too soon? Negative spillover
What is "good enough"?
"Don't be distracted by the myth that 'every little helps'. If everyone does a little, we'll achieve only a little." - David MacKay
"The comfortable perception that global environmental challenges can be met through marginal lifestyle changes no longer bears scrutiny. The cumulative impact of large numbers of individuals making marginal improvements in their environmental impact will be a marginal collective improvement in environmental impact."
Since most people do easy and cheap things for the environment before difficult and expensive things (Diekmann & Preisendörfer, 1998; Kaiser, 1998), they may in practice justify not doing the more difficult and costly – and usually more important – things ... [SimplePainless]
Discussing this report [WhatCanYouDo] discusses the 'spillover' effect, including:
WWF's report offers an examination of this "spillover" effect, exploring the potential for one small pro-environmental behaviour to lead to another (positive spillover) as well as the potential for a pro-environmental behavior to have the opposite effect, discouraging a person from adopting other pro-environmental behaviours (negative spillover).
The effect is real, and if HaaS is going to achieve its secret carbon-cutting goals, it must be done in a way that avoids negative spillover if possible.
Can apathy be overcome?
Initially, and to persist savings.
Do nudges work?
Does gamification fix everything?
Gamification may only engage a small subset of users, eg young males, and may not improve outcomes even amongst them, so is not a panacea [CHARM].
How long do automated and non-automated measures typically last?
The challenges lie not just in getting people to start saving energy, but in persisting those savings, especially non-automated measures, once the initial novelty has worn off.
Asymmetric messaging may also be necessary to avoid making the people who happen to be doing best from thinking that they are working too hard and slacking off! [AsymMsg]
Fully-automated 'technical' measures reportedly last ~10 years, whereas those requiring continuing human intervention maybe 4 years [???]. (Programmes measuring the persistence of manual measures tend to stop recording data after a year, which distorts the results to look like manual measures fail immediately after; the 4-year value is an inference.)
Once people have engaged in principle, is is helpful to actively guide them to make the best savings that they will be comfortable with [EcoCoach]. It is also to provide appropriate feedback and/or incentives to reach and maintain the desired saving. The agency, ie sense of control, that eco-coaching brings, encourages continued experimentation (important because needs will evolve over time). This agency, and the sense that settings were tailored to their personal needs and habits, also seems to leave users more committed to sticking with the level of comfort vs savings they have selected.
This agency, evolution and commitment combination is likely to be key to HaaS success.
Effective Useful Life (EUL)
Whatever measures we employ with HaaS, our target should be something like a physical equipment life of ~10 years (to keep costs down, and avoid adding WEEE - waste electronics - to landfill unnecessarily), and we should aim to match that with the human element.
A ~10 year life may also match the tenor of investments that green YieldCos may like, but if not then some securitisation slicing and dicing may be possible.
We should aim to monitor, verify and report on these EUL measures [SERA-2009]; many types of investors in such schemes would likely want that tracking. Doing this well may to prove to be where some new ground is broken.
Privacy issues, personal data and security: smart metering
We feel that it is important to give users control over data generated by their activities.
We also understand that some users may be happy to trade some personal data for, for example, improved energy saving by allowing crunching of their usage patterns in the cloud combined with weather forecasts.
We also take security seriously, not wishing people to be spied on nor have their house messed around with by bored teens in Moscow or Croydon.
These two also speak to graceful degradation to completely unaffected operation if a user's Internet connection goes down.
Devices such as Nest routinely export data from the house, and similar devices have been known to fail when the home Internet does. Some existing heating controls leave users very open to having huge bills run up and systems broken remotely.
One of the attractions to us of smart metering is the ability to track energy usage (and thus savings) locally, and only upload minimal summaries to maximise privacy where required.
We should try to accurately gauge home energy users' views on these topics, as they may steer or constrain the flavours of HaaS system that can be put together. For example, competitive gamification elements maybe easy for those who already share details of every meal and trip away from home on Facebook; less so for the paranoid banker.
Decoupling: Can We Cut Carbon and Grow Revenue?
As discussed above, OpenTRV's ulterior motive is to achieve cuts at scale in domestic carbon footprint, while improving health and comfort, and providing an attractive and viable business model for utilities and suppliers of saving tech such as OpenTRV. Does this look possible?
HaaS Lite does indeed look possible from this 'customer'/'social' strand point of view, but the subtleties and layers of user messaging and interaction to get right may be harder than expected for a single-year project. Indeed this seems as if it will the most tricky of the three strands, and the hardest to objectively measure.
Agency, evolution and commitment are highly likely to be key.