Advanced modeling techniques can help loyalty program sponsors understand their members better and extract more long-term bang from investment and marketing bucks.
Understanding member behaviors has always been important for loyalty program sponsors who focus on the estimation of associated revenue and cost. Today, the dominant analytical approach is to rely on summary redemption data. But for how long will that persist, when more advanced modelling techniques, such as decision trees (available within the Willis Towers Watson LoyaltyAdvisor framework) and customer lifetime value (CLTV) modelling, offer the lure of more personalized member insights that, in turn, can help financial performance?
Since loyalty programs have continued to grow in both popularity and scale, such details will increasingly matter. Finding the correct balance between unlocking additional revenue opportunities and granting points – that can represent a very material liability on a company’s balance sheet – is taking on added significance.
Knowing, for example, whether certain members hoard points for a big splash or, alternatively, pounce on available deals and promotions as they come up will have a significant impact on benefits design and costs.
A step forward
The reason that decision trees and CLTV modeling are particularly valuable is that they group members into a number of clusters based on similar accrual and redemption behaviour, rather than look at them in aggregate. The future performance of the program can then be forecast more accurately from this reduced number of groups.
Such a move carries little, if any, early mover risk for loyalty programs. CLTV is hardly a new concept after all. CLTV modeling has been widely used in retail, banking, insurance and other sectors to estimate profit and loss at an individual customer level.
Characteristics that are suited to loyalty programs include the ability to model different accrual and redemption behaviors as discrete groups and then predict the group that a given customer may be in, as well as their propensity to change between groups in future. For example, one model might predict how likely the customer is to stay engaged with the product, while another might predict spending patterns while engaged. By combining these models, users can predict the overall expected revenue and liabilities for a given customer, as well as enhance their understanding of how each customer arrives at this position.
So what does this mean potentially for loyalty program sponsors on a practical level?
1. Deeper understanding of risk and benefits
Fundamentally, CLTV used alongside decision tree modeling approaches, can provide a much deeper understanding of risks and benefits. By comparing the marginal revenue and marginal cost that results from each additional point accrued, sponsors can identify the most profitable members of a program.
2. Program optimization
A greater understanding of member behavior would allow sponsor companies to more accurately estimate how members’ accrual, redemption and engagement would change as the terms of the loyalty program change – and from that, to be able to optimize the program to offer terms and features that best meet the needs of the most valued customers. For example, how would customer engagement and redemption change as the expiration rules are varied, or if the awarded points were increased for a given transaction type?
3. Targeted marketing and communications
Sponsors can more precisely target marketing and communications, based on having greater insights into the relative performance of various segments of the membership population.
Marketing departments can already target advertisements to specific demographics, but most won’t have access to information on how actual or proposed loyalty program changes will impact members’ earning and redeeming behaviour. With an analysis conducted at the individual member level it becomes simple to segment these behaviours and identify better performing segments to target.
Steal a march
Any loyalty program sponsor aims to maximize economic value. The techniques for doing so are moving on quite rapidly. Those that adopt models that enable them to get more into the psyche of their members will potentially steal a march on competitors that continue to rely principally on more basic aggregation methods.
Manolis Bardis and Andy Rigby are lead advisers to loyalty program sponsors at Willis Towers Watson.