Rethinking superannuation data and member engagement
Many super funds are investing heavily in data profiling and predictive analytics solutions which promise to unlock a tidal wave of insights, often geared toward improving “member engagement”. Indeed, the term “actionable insights” has been dropped countless times across the latest financial services conferences.
But mostly, these analytics solutions are just providing more data — not insights – and it is no wonder.
Standard data and risk profiling generally focuses on basic member and financial data, such as age and account balance. Predictive analytics often requires complex models and is focused on classifying likely outcomes across a segment rather than an individual member. While these can provide important information to develop product strategies, there are additional levels of data that can provide context-specific insights that are truly “actionable” with the member.
As a result, rather than striving to increase member engagement, the focus should be on improving engagements with the member.
Often it is said: ‘If we’d known that we would’ve treated that member differently’. This is mostly because a member characteristic was not identified or measured in context prior to a particular event. However, the business knowledge is already embedded in the member data; it simply needs to be unlocked.
So how can existing fund data be translated into ‘actionable insights’ that can be used to benefit the member? The key is to treat each member as an individual.
Each member may have already been put into various segments based on age, balance or employment type, but the key to knowing members is to create and manage a profile for each and every one of them.
The starting point is to identify exactly what data is available. This is best achieved by gathering and classifying three lenses of data:
Basic member details
These are the core profiling factors and can include age, account balance, employment type, insurance levels and length of membership. While these factors are not weighted or ranked, they can be used to better segment and analyse the member engagement and sensitivity profiling.
Member engagement level
Often engagement is measured via surveys and external studies, or simply categorised based on age or gender classifications. However, funds already have the data that can provide an engagement score for each individual member. This includes the level and frequency of member-online activity, personal contributions, investment switches and whether the member has sought financial advice.
Member sensitivity level
This is based on two key elements:
identification and measurement of events outside business as usual, such as adjustments and complaints; and
the specific characteristics of the member and their account, such as if they have been on-claim or received a financial hardship payment.
The ‘sensitivity’ element is frequently overlooked but is so often a trigger point in member’s leaving a fund.
How member profiling works
The objective of member-level profiling is to establish a clear, easily calculable and repeatable weighting method that can be applied across all members, creating a weighting and classification that measures the level of engagement and sensitivity for each and every member of the fund.
There is no one-size-fits-all approach, however, 18 profiling factors for each member is recommended; six each to profile: the member’s account, their level of engagement and their level of sensitivity.
Once the factors have been provided with a total weighting, the ranges within each factor need to be identified. These ranges must be designed to ensure that each member fits into one (and only one) range for each factor.
See separate case study below: ‘Profiling in action’, to understand how factors can assist.
Weighting each factor simply means ranking its importance. The weighting can use a percentage term measurement, with each member scoring a value ranging from 0 percent to 100 percent.
It is important to understand that the factors, ranges and weightings are all subjective. Stakeholders will have differing views as to what to include and the percentage allocations; the key is to agree on the measurement and use it consistently.
Importantly, keep it simple. The measurement process needs to be simple enough to understand, yet comprehensive enough to provide meaningful results. The usage of six assessment factors for each profiling class aims to strike this balance.
Key benefits of member profiling
The member profiling factors gathered - along with the engagement and sensitivity classifications - are a simple, consistent method for measuring all members of the fund. But underneath the simplicity of the concept is a powerful asset that can assist in a wide range of business decisions. If 18 factors with just four possible ranges are gathered, there are actually 68 billion possible combinations for each member.
Some of the tangible benefits that the member profiling concept can deliver include:
Real-time, detailed information for the advice team
Enhanced quality checking processes to be introduced for ‘sensitive’ members across operations
Targeted strategy for assigning inbound member interactions (call, email) to the most applicable operators
Targeted strategy for marketing campaigns based on member engagement scores
In theory, every form of member interaction could be enhanced using member-level profiling.
By rethinking engagement to focus on the member, funds can decode the buzzwords, translating the concepts into practical and repeatable analysis that can offer real and measurable business benefits.
Better understanding the member data that they already have, and recognising the additional data they need to seek out, means that funds can better interact with individual members and in a more meaningful way.
Case study: Profiling in action
Fund A identified a system issue where the previous weekend’s deduction of quarterly administration fees were duplicated for all accumulation accounts with a balance over $100,000.
Using the member profiling lense, they identified that 2,500 members of the fund were impacted by the issue.
Using the member engagement lense, they identified that 125 of the 2,500 member’s checked their account balance via the fund’s online site more than once a week, plus 55 of these member’s had used the fund’s financial advice service
Using the member sensitivity lense, they found that of 95 members had a rating over the ‘highly sensitive’ threshold of 80 percent. Of these, 30 members had been impacted by another remediation event within the past 12 months.
The fund was able to develop and fast-track a targeted communication strategy for the members classified as both ‘highly engaged’ and ‘highly sensitive’ as well as those impacted by the previous remediation event. In addition, the members who had used the fund’s financial advice service received a direct call and explanation from their adviser.
Subsequent analytics found that less than 1 percent of impacted members requested a rollover to exit within the next six months, a significant reduction in the 4 percent turnover that resulted from a similar event prior to the introduction of member profiling.
Mark Vaughan – Managing Director
If your organisation is investing in the capture, analysis and utilisation of member data and would like to get more from data, then please contact QMV for a preliminary discussion. You might also be interested in checking out Investigate. Investigate is an automated data quality management solution, which now manages data for over 10% (and growing) of Australia’s total superannuation funds.
Like what you see? Please subscribe to receive original QMV content !
You may also benefit from our free monthly pensions and superannuation regulatory updates.