This site uses cookies to provide you with a more responsive and personalised service. By using this site you agree to our use of cookies. Please read our cookie notice for more information on the cookies we use and how to delete or block them.

Bookmark Email Print page

Using analytics to understand KiwiSaver customers

Author: Adam Follington

Understandably, the early years of KiwiSaver has seen providers focus on the basics – getting systems working, putting reporting in place, developing products.  However with these building blocks in place there is now an ever increasing focus on the members (or customers) themselves, trying to better understand them, communicate with them and appropriately influence their behaviour.

Ideally, this analysis and research should be focussing on one primary goal – to increase customer engagement. In turn this engagement will result in greater retention, natural organic growth and ultimately enhanced profitability.

The success of KiwiSaver means there is a wealth of customer data which can be used to this end.  KiwiSaver providers would be foolish to ignore the potential of this data and the opportunities it brings.  “Big data” organisations the world over, such as retailers, banks and websites, are increasingly awakening to the benefits of cold, hard data analysis in conducting their everyday business.  Analytics does not replace the need for the fundamentals of business, such as good management, processes, systems and people – but nevertheless it is a very useful and effective tool to have in the armoury.

Whilst data analytics has become a trendy technique it is not new.  Actuaries have been using analytics for centuries to understand and predict customer behaviour.  Until recently this was focused on the areas of insurance and superannuation but increasingly actuaries are applying their skills in a wider array of fields.

Wanting to use analytics and using it effectively are two different things.  It is important that any analytical problem is approached in a structured way.  The following is a suggested approach:

1. Understand the problem you are trying to solve

Analytics is always more successful when it is trying to answer specific questions.  This could be focused on your portfolio, such as:

  • How can we segment our customers in order to better understand them?
  • What value is expected to be driven from different pools of customers?
  • How do we contact customers where we have little contact details?
  • What are contribution rates / dollar amounts across different socio economic segments?
  • Which customers are nearing retirement and might want to think about investing their money post-retirement?
  • How long will it take me to recoup the investment in a new website and mobile app?
  • What do the most successful financial planners have in common?

Or it could be broader questions around the impact of the scheme as a whole.  For example:

  • Drivers for choice of fund: why do people choose certain risk types or companies? What are the drivers? What are the drivers for not making a choice (default option)?
  • How does KiwiSaver help augment the State Pension? In which customer segments do we foresee the largest deficit?
  • Awareness: how many people actually know how much they have saved? How much (or little) idea do they have about how much money they need after retirement?
  • What are KiwiSaver retirees doing with their funds? 

 

2. Understand your data

There are a number of sources of data available which may be useful in solving your problem. The obvious one is your base administrative data and that is a clear starting point.  However that can often be enhanced using other sources such as other internal data including contact centre usage, financial planning information or even dormancy.

Additionally there is the opportunity of analysing data matched within your organisation.  If you provide other products to this customer (e.g. banking, insurance) then data sources can be matched to build a bigger profile.

Lastly, external data cannot be ignored.  This could be census data or industry information used to attribute the data with average profiles.

 

3. Know your facts

Make sure you can walk before you run.  Before attempting any complex analytics it is useful to make sure that the very basic levels of analytics have been covered off and are well understood.  So this might be ensuring that you have a good profile of your customers developed:

  • How many are in each age group?
  • How many are contributing the minimum amount? 
  • Where are your customers based – do you have a good map of where your customer’s live?
  • Do you know the profile of the employers – small, medium, large?

Such basic analytics, while perhaps not as sexy, can nevertheless sometimes produce useful insights and are a good tool for giving senior management a quick overview of your customer base and getting buy-in on more complex techniques.

4. Understand what is happening

Do you have a clear understanding of what is happening in your portfolio, even if you can’t explain why it is happening?   This should be linked to the problem you are trying to solve and should be readily identifiable by analysis of movements in your data.  For example, how many customers are withdrawing funds for their first home purchases?  Were the member inflows in line with expectations?

5. Predict what will happen

Using different analytical and actuarial techniques it is possible to predict customer behaviour based on historically observations.

So, for example it will be possible to build a financial model which can predict, with reasonable accuracy, the funds flow from individual customers going out to the customers retirement age (and even that can be varied based on assumptions of government behaviour!)  Such lifetime models are used by banks and insurance customers to determine the lifetime value of customers and portfolios.  This information is used to justify marketing spend and investment in systems and can also be used to identify the highest value customers and hence where any dollar spend is best focused.

And predictive analytics can also be used for more surgical procedures.  For example, it could be used to build up a profile of likely lapsing customers based on the profile of previously exits.  Combining this with a lifetime value model tells you which customers to focus on in any persistency campaign.

6. Influence what will happen

Using the insights gleaned from analytics take action to deliver the desired outcome.  Perhaps you have learnt that different types of customers respond to different messages when encouraging a greater contribution rate. Using this information, and matching to your portfolio, you can deliver a campaign tailored to individual customers.  Even a small increase in a success rate is likely to justify any investment costs.

Analytics has a key role to play in helping businesses solve problems and deliver outcomes and the opportunity to embed analytics into your business to enhance customer engagement is undeniable. With the mass of customer data, and within an increasingly competitive environment, KiwiSaver providers are ideally placed to take advantage of data analytic techniques. 

Whilst a few organisations may have already applied some existing filters to establish a better understanding of their KiwiSaver customer base, clearly much more can be done to achieve genuine engagement and long term loyalty.

The question is who is going to take the lead?

Forward Focus March 2013 contents:

 

Related links

Stay connected:
Get connected
Share your comments

 

More on Deloitte
Learn about our site