Persuasive Nudging
Towards Better Member Engagement
The increasing availability and usability of data can empower superannuation fund trustees to analyse member engagement at an individual level rather than segmenting classes of members. This can enable the trustees to apply behavioural design principles to nudge members towards target behaviours appropriate for their engagement profile.
There is a lot of talk about member engagement in the Australian superannuation system, and considerable effort and budget directed towards increasing leverls of engagement. One important observation I have made is that member engagement can mean different things to different people.
On the one hand, segmentation and alignment of product with member engagement levels (as indicated by exercise of choice) has informed regulatory reforms for superannuation products. Member engagement was the basis for creating MySuper for the less engaged, and ATO management of lost (or completely disengaged members).
On the other hand, there is an understanding of member engagement which is focused on increasing levels of engagement. This form or member engagement sees greater member involvement as an end towards which the industry should direct efforts. This often has unconfortable proximity to brand loyalty or member retention rather than member value proposition and outcomes.
I guess it really depends on what one means by member engagement, as there is value in either understanding. A more nuanced approach to analysing member engagement can balance these somewhat competing understandings of member engagement.
Data and the New Information Economy
The proliferation in data production & capture over the past few years is real, and will continue. There’s usually a pretty graph to insert here which shows an exponentially growing trend of endless transition from billions of gigabytes, terabytes, and now zettabytes.
While this can make for a great soundbite, I’ve got to say that our holiday photos, tweets about sports games and videos demonstrating how to apply makeup aren’t much use to trustees of superannuation funds.
Much like crude oil, the value of data lies in its utility. The refinement of meaningful outcomes from all of these zettabytes isn’t easy. The risk of distraction in such a large and noisy place is significant.
Even more than distraction, there are the risks presented by inadequate data quality and accuracy. What’s often assumed to be reliable data is all too often found to be inaccurate or incomplete. Having robust controls in place is increasingly important.
The information economy is built on the development of technologies which can facilitate better decision making. Better decision making is more efficient and informed decision making.
Analytics also assists in our ability to consider the future with some confidence. Even more powerful than predictive analytics is the ability for data driven technologies to impact, and influence human behaviour. The power of data driven behavioural design is profound.
If we consider the use of a design principle such as the Hook Model, or intermittent variable rewards to maintain attention (such as are used in poker machines, or your Facebook news feed), or the use of data driven prompts to promote better health, we’re seeing a major shift towards a world where human behaviour is influenced by data driven technologies.
It should then, come as no surprise that there is a growing concentration of power in those who control data. It should be equally unsurprising that there is a need to regulate the controllers of data to protect consumers, and ensure adequate competition exists.
This New Information Economy is the environment in which we need to consider member engagement.
Member Data Profile
In many ways, there’s nothing new about using data to inform decisions about member engagement. There is a well-established suite of applications which contain a profile of members.
This is our comfort zone. While many of us are only now coming to terms with managing the information in these systems, there is a significant shift on the horizon. The opening up of how data is used to give consumers (and members) greater control is a trend which will become increasingly important.
PC Inquiry into Data Availability and Use
Earlier this year, the Productivity Commission released its report into Data Availability and Use. The fundamental premise of the recommendations is that data sourced directly or indirectly from individuals, remains accessible to the individual (although not necessarily property).
Therefore, individuals should have a Comprehensive Right that would enable them to share in perpetuity joint access to and use of their consumer data with the data holder (including granting third parties the right to access their consumer data).
This appears to be happening, and the breadth of the change is significant. The recommendations cover everything from financial services, health, education, publicly funded academic research and anything else which we put online!
The opening up of access to data will present both competitive opportunities and threats. It is also bound to raise serious ethical questions of data privacy and security.
Privacy Attitudes and Ethical Considerations
The Office of the Australian Information Commissioner recently published the results of some research it had undertaken into the attitudes of Australians to privacy.
The results provide some generalised insights about community attitudes to data privacy.
What can we take from this? Financial institutions are generally trusted, yet there is still a significant segment of society which doesn’t trust financial institutions with their personal data.
There’s a growing concern with data privacy, which is predictable as a reflection of the growth in the importance of our personal data in our everyday lives.
There is resistance to the sharing of personal data between companies. However, the fact that 21% of Australians (a significant proportion) are not opposed to their personal data being shared is also significant.
It’s clear that there is a lot of diversity in the community which isn’t clearly visible when you present some high-level figures like this.
Individuals will hold diverse views about the privacy of their personal data, even throughout their own lives and in different contexts. Such diverse views about data privacy can be managed, if we take an individual member centric approach to engaging with members.
Member Individual Profile
The selfie is something which has risen in prominence over the past few years. However, it’s not something new. The fact that pictures of one’s self can now be posted online is reasonably new. The social affirmation which likes and comments can give, provides a solid dopamine hit, and guarantees that we see more selfies.
The growth of the selfie demonstrates the power of developing an individual identity and narrative for members.
Rather than making generalisations about member engagement, we are now moving towards an environment where the profile of each individual member can be constructed to consider certain aspect of the member. Some kind of data selfie.
There’s no limit to the number of profile indexes which Member Data can be used to develop. I’ve listed a few simple profile indexes which can be created for a superannuation fund member:
- Retirement Planning Profile (to inform financial advice)
- Engagement Profile (which goes above and beyond a simple marker such as investment choice)
- Ethical Profile (social media can be great as an indicator or any non-financial investment objectives)
- Fraud Risk Profile (to assess the likelihood and impact of fraud against a member)
The more relevant data which is available as a reference in informing these indexes, the more valuable they can be in identifying target behaviours which are consistent with the member’s individual profile.
While there’s no standard, I’ve suggested a few components which might inform a typical engagement profile.
- Awareness (levels of awareness, or basic connection between a member and trustee)
- Involvement (active interactions where the trustee and member work together to make decisions, such as the involvement of members in high level strategic asset allocation investment decisions)
- Independence (where a member takes on responsibility for decisions independent of the trustee)
- Demographics (are commonly used, however are valuable as indicators)
- Frequency (the regularity of interactions can demonstrate the amount of attention which a member is able to dedicate towards being involved)
- Behavioural Biases (such as hyperbolic discounting and overconfidence might be visible in available data)
- Life Cycle (with it being well known that the level of engagement of members will change throughout their lifecycle).
So just to reiterate, an individual profile can assist in identifying target behaviours which are appropriate based on the member engagement profile.
In many instances, the member individual engagement profile might indicate that only awareness might be appropriate, and that the member is ill suited to be encouraged to engage in behaviours which require significant time, knowledge and effort.
Better Engagement by Targeting Behaviours
The beautiful thing about the trust structure is that it is designed to accommodate the lack of time, information or ability in an individual which may place them in a position of dependence — or low engagement.
The trust structure in the superannuation system has been designed to allow for both high and low levels of engagement, and shifts between these levels of engagement at different periods.
Rather than focus on increasing the number of interactions between a member and trustee, member engagement can be most effective when specific behaviours are promoted, based on the member individual engagement profile.
There is a set of behaviours which we’re all familiar with. These behaviours fall within a scale which starts at awareness, progresses to involvement, and ultimately independence.
Persuasive Nudging
There is a proven science dedicated to the design of technology with the intention of influencing human behaviour. It is very effective, sometimes dangerously so. It can be applied to promote better member engagement through targeting specific member behaviours.
The prospect of Nudging (or even mandating) individuals towards better outcomes has gained traction over the past decade, driven by developments in behavioural economics and finance.
While there is an important role to be played by Defaults and the Trust relationship, the increasing ability to individualise analysis can enable a more nuanced approach in using persuasive technology to target specific behaviours.
I’d like to introduce you to persuasive nudging…
The model proposed here is an adaptation of the Behaviour Model for Persuasive Design developed by Stanford University Professor BJ Fogg. This model has provided the basis for the commercial success of many smart phone applications.
Fogg’s model asserts that for a person to do something — whether it’s buying a house, clinking a link in an email, or going for a 30 minute jog— three things must happen at once.
The person must want to do it, they must be able to, and they must be prompted to do it. A trigger — the prompt for the action — is effective only when the person is highly motivated, or the task is very easy. If the task is hard, people end up frustrated; if they’re not motivated, they get annoyed.
Having identified a target behaviour , the model suggests that the nudge towards the target behaviour will almost certainly be effective if three aspects are satisfied:
The Member must be adequately motivated
There are some basic psychological responses which govern a member’s motivation levels for any activity. The pleasure and pain response is particularly influential in the short term. Longer term motivations are often driven by prospective hopes and fears, while social acceptance or rejection can also play an important role. In targeting account consolidation, a particularly strong motivation may be the communication of the potential benefits of consolidation in the long term, or to promote social conformance by highlighting that so many other members have consolidated their accounts.
The Member must have High Ability to perform the target behaviour
Simplicity is manifest in reducing the required time, money or effort to complete the activity. Advances in information technology are wonderful in increasing the ability of members to complete target behaviours.
The Member must be triggered to perform the target behaviour
Finally, our limited attention makes it critical that a member is triggered to perform the target behaviour. The individual member profile which the increasing availability of data has allowed us to create can be very useful in identifying appropriate life events as trigger points, which can be used to notify a member that a particular activity should be engaged in.
Behavioural Design is a very powerful, which I believe is underutilised by the superannuation industry. BJ Fogg summarised well in describing the most important nine words in behavioural design as being:
“Put hot triggers in the path of motivated people.”
I’ve suggested a range of big themes and ideas which can help individuals better save for retirement, but they can also result in additional costs — and inferior outcomes — for members if not done well or exploited.
Ensuring that our efforts to accommodate varying levels of member engagement add value to the retirement outcomes of members needs to be at the forefront of our thinking.
This article is an edited transcript on a presentation delivered to the Association of Superannuation Funds of Australia Spotlight on Member Engagement in Super Forum on 11 September 2017 at the Sofitel Sydney Wentworth.
