Download full publication

Related Sectors:
Related Disciplines:
Behavioural Economics
Related People:
Dan Elliott

Getting engaged

Regulators across sectors are requiring companies to engage with customers more than ever when developing their business plans. Insights from Behavioural Economics (BE) can be useful in designing innovative approaches to meet the challenge. At a recent workshop two clients Frontier has been helping with their engagement programmes shared some of the lessons they have learned.

It’s good to talk

It has long been recognised that a particular weakness of economic regulation is that the regulated company may have an incentive to improve its financial performance by cutting corners on service quality rather than by increasing efficiency.

Over the years regulators have adopted a range of methods to address this incentive. But it has become clear in the last decade that prescribing service levels, targets, etc. can have significant drawbacks. At some point the pursuit of ever higher service standards is likely to hit the law of diminishing returns. How then do we know whether the targets and standards being set are effective and efficient? Do they actually serve the interests of the customers the regulation is there to protect?

The seemingly obvious answer is to ask the people to whom it matters: regulated companies should engage with their customers to find out what aspects of the service they provide matter most to them. Businesses especially need to determine whether customers are willing to pay what it costs to provide the better service that they demand.

To that end, regulators are asking the companies for which they are responsible to talk to all their relevant stakeholders and use the information they gather to design their own “optimal” performance regimes. The energy regulator, Ofgem, and the water regulator, Ofwat, are particularly keen on this approach and are now being joined by the airports regulator, the Civil Aviation Authority.

However, engaging with customers in regulated sectors has a number of challenges such as low awareness of the services being supplied and a perceived lack of choice. Insights from Behavioural Economics (BE) can help companies to design innovative approaches to address these challenges.

Another reason why BE is valuable is that they help utilities to meet wider social goals., BE insights can be instrumental in getting people to change their habits, for instance by promoting energy and water conversation. This can result in major investment savings, thus improving affordability.

On 14thNovember, Frontier held an event in its London offices, bringing together expert speakers and an audience of senior decision makers from the three sectors that have adopted this approach.

Jon Ashley, director of customer engagement and outcomes at Ofwat, talked about how it has sought to get water companies to better reflect customer priorities in their regulatory business plans. He also addressed the importance of robust and proportionate evidence and the role that customer groups play in challenging this research. Kathryn Greenhalgh from Heathrow and Sarah Jenner and Frank Grimshaw from United Utilities (UU) then spoke about the specific challenges of their respective engagement programmes. Their insights were supplemented by observations from Frontier’s Annabelle Ong and Rob Francis.

Lots of lessons

The exchanges revealed many lessons applicable to any business considering applying BE as a tool to improve customer engagement, especially in a regulatory context. They also highlighted the need for BE approaches to reflect the fact that customers engage with different organisations in very different ways: a hub airport is a far cry from a regional water and sewerage company.

Despite the differences between the two businesses, the workshop identified some key themes:

  • It is essential not to presuppose the customer’s priorities. Identifying those priorities is the essential first step if we are to be sure the results of the research are a proper reflection of customer views. When pulling together a regulatory plan, companies need to be clear how customer priorities have shaped the plan, not the other way around. In addition, companies need to be clear about the areas that customers can truly influence and make sure that research is focused on those areas.  This may seem obvious, but in practice it can be a challenge.
  • If the research is to place valuations on customer preferences, it needs to be meaningful to the respondents and designed in such a way that it can elicit marginal assessments of changes in performance levels. This is particularly important for issues – typically longer term in nature – that customers do not generally consider and those that involve weighing up probabilities, such as a service provider’s resilience.
  • There is no unique way to measure customer valuations of service changes. Insights from BE are especially useful in the design phase of this research.
  • Not only different methods of questioning but also alternative research approaches will produce different valuations. A variety of methods should be used so that companies can develop an overall evidence base that is robust.
  • There is a vital role for judgment in the ultimate design of any incentive scheme. In particular, if you have conflicting pieces of evidence, simply taking the average may not be valid. The fact that the evidence is contradictory may itself be of importance.

The customer comes first

Both UU and Heathrow emphasised the need for research to be customer-led. Heathrow stressed how new customer studies to inform the regulatory plan did not start from scratch but built on an extensive body of research that it is constantly updating. The integration of new research into ongoing operations chimed with Ofwat’s views on how engagement should be carried out. UU underlined the two-stage nature of its work, first researching priorities in an open way that does not direct customers to particular outcomes before proceeding to a valuation phase.

Our speakers identified some significant challenges that need to be overcome in designing consumer research. These included:

  • Customers may not be good at placing a value on programmes focused on long-term improvements, such as enhancing network resilience or expanding capacity.
  • Customers may not be very aware of what the company does, especially if they don’t deal with it very often. This may be a more serious issue for a water company than for an airport.
  • Customers tend to value more highly programmes that have a human impact over ones that have a more abstract environmental effect.

All of these issues present a real challenge to the design of effective customer engagement. It is often not sufficient to take a simple survey approach, because customers just do not think about the issues at hand and so cannot provide valid answers.  This means the survey design can have a big impact on customers’ responses. This can be particularly important when customers hardly every actually experience a given problem (e.g., unplanned water supply interruptions), which makes it difficult for them to place a value on avoiding it.

Research puzzles (and farms)

These problems call for innovative approaches that help the customer understand the issues being addressed from a more personal perspective and to understand how they would really feel when faced with the incidents being considered. This sometimes means calling on the skills of creative design agencies as well as survey companies and behavioural economists to design “immersive” approaches to familiarise the customer with particular issues through models, demonstrations, puzzles, interactive games and so on. Importantly, this is quite different from deliberative workshops. Our immersive research allowed customers to engage with difficult topics over 1-1/2 hours and to provide robust views.  We achieved this by developing physical materials that help customers visualise the issues being discussed and so engage with them in a more realistic frame of mind.

The examples below show some of the visual aids that were produced to help customers think about key issues. A model farm was used to show very graphically the relationship between farming practices and river quality. To bring home the impact of water service interruptions, a board provided a clear reminder to respondents of all the ways they rely on a continuous water supply in their day-to-day lives.

The importance of BE in designing research approaches came out both in the design of survey questions and the consideration of the evidence that could be drawn upon. For instance, Frontier highlighted the role of key BE influences of priming and framing on survey results. We gave as an example how nine different approaches (3 frames x 3 primes) were taken to a single survey to test how sensitive the results are to the way the questions are asked. In this case, the results were shown to be sensitive to the way the problem was framed, demonstrating the care needed in both the design and interpretation of survey results. Another lesson is that simple averaging of results may be invalid and only serve to conceal more fundamental issues with the underlying evidence.


While simple improvements to survey design can help make the research more robust, companies should use multiple methods.  For example, UU and Frontier explained how we had used “natural experiments” to corroborate the results of willingness to pay (WTP) survey work.

In this case, UU had measured the willingness to pay of customers to avoid prolonged service interruptions or incidents of sewer flooding. But we used the natural experiment of customer contacts with the water company in the event of these incidents as a cross-check.

The WTP research also tested different approaches, for example single question surveys instead of options allowing customers to trade off multiple objectives. There is some a priori evidence to suggest that the former approach may lead to larger estimates of WTP than the latter, without any suggestion that one is more accurate than the other.

BE to the rescue

All of these approaches need to be brought together to feed into the business plan. The conclusion from these exercises is that companies need to produce plans that are customer-focused as well as evidence-based. That evidence, especially as it relates to customer valuation of changes in service quality, requires a familiarity with the field of behavioural economics. While BE may feel new to utilities, it has the potential to provide significant insights into what customers want and value. It can also provide new and interesting ways for regulated companies to engage with their customers. Who doesn’t like playing with a model farm?


Comments are closed.