Getting Predictive Around Customer Loyalty

Digital transformation continues apace across businesses and industries, and so new technologies are being leveraged to boost customer loyalty and improve those programmes that focus on this. Predictive analytics, in particular, enable brands to build better and more targeted programmes by making predictions based on customers’ demographic, purchasing and behavioural data. 

Effectively using such data should enable organisations to develop targeted promotions and rewards for individual customer segments, enabling the business to improve its levels of personalisation – something that is increasingly in demand from consumers. 

Getting predictive

Mike Renzon, group CEO for inQuba, a customer journey management software provider, points out that loyalty in its simplest form means that companies can retain existing customers, attract new customers through promoter referrals, and successfully introduce these customers to new products and services. “However, it should be clear that no loyalty programme can make up for a poor customer experience and there is a real danger of the loyalty programme becoming the ‘product’ that customers value, even more so than the actual product or service the company supplies.” 

Therefore, says Renzon, a loyalty programme should be designed as the cherry on top of an extraordinary product or service. “Even better, it should be integrated into the product or service experience itself, in much the same way as Uber One offers a flat-fee loyalty programme for regular users of the service.” 

Cindy Carvalho, head of rewards and pricing at Investec Private Bank South Africa, notes that, just as in any industry or business, useful information can support decision-making and strategy. “When it comes to loyalty and rewards programmes, artificial intelligence (AI) and predictive analytics can assist companies in optimising the performance of a programme. These technologies do so by forecasting the effects of tiering with more confidence, accurately targeting clients most likely to take up an offer, and managing the liability risk of unredeemed loyalty points.” 

According to Orediretse Molebaloa, head of solution engineering at Infobip, a provider of predictive analytics solutions, the value of personalisation lies in being able to scrutinise certain data about the customer. “With this knowledge, you can hyper-personalise the customer experience, for example, by providing them with notification only at a time of their choosing, which adds an intimate kind of value to each customer.  

“In addition, it opens the door to offering revolving, targeted rewards that can be used to help change shopping behaviours. It is also worth mentioning that a more personalised experience can provide data that can be mapped against fraud detection technologies to improve your overall security.” 

Consent and privacy 

Regarding personal information, Molebaloa indicates that the key lies in properly managing consent and data privacy. Infobip, he says, ensures it aligns with key local and international privacy legislation and then also utilises encryption on its platform to ensure personal data remains private. 

Renzon agrees that data privacy is at the heart of trust, suggesting that there needs to be an open and transparent handshake between businesses and each of their customers. “Put simply, organisations need to be transparent and customers need to feel safe and understand and agree to the benefits that sharing their data will provide to them. 

“There is always a concern around ‘algorithm bias’,” adds Renzon, “but this can be managed with a ‘walled-garden’ approach, where AI only acts against content that is specific to a consumer or consumer segment and does not access the public domain of training data. Once again, individual customer context is extremely important as this context drives the engineering that generates prompts specific to an individual customer’s particular context at that point in their journey.” 

AI systems require large amounts of high-quality data to learn and make accurate predictions, Carvalho says, and as such, some of the biggest challenges currently being faced are the quality of data, algorithm bias and data privacy. These can hamper the ability to quickly get actionable results based on the data. “Given that machine learning models are written by people and trained on socially generated data, it is important to compare and validate training data samples for accurate representation. Regarding data privacy, adhering to data protection policies is critical. Giving clients a view of how their data is used and the ability to opt in and out of certain instances is non-negotiable.” 

Asked about how to keep a “human element” in such a technologically heavy scenario, Carvalho says: “Companies should enrich the quantitative data with qualitative data such as client surveys and feedback through various channels. A feedback loop to clients also goes a long way in maintaining the human element. In the new digital marketplace, consumers expect intuitive user experiences and personalisation.” 

 A hyper-personalised future

Molebaloa believes that moving forward, businesses should look at implementing technology that enables the business to relate to customers across multiple channels via an omnichannel strategy. “Different customers have different channels of contact that they prefer – whether it is email, WhatsApp or online – and so businesses need to cater for the needs of all customers. In the end, you need to be able to reach all the different customer touchpoints if you wish to meet their needs. Personalisation will be followed by hyper-personalisation, with the latter achieving its full potential as digital transformation across industries progresses, and legacy systems are eliminated and replaced with AI and predictive analytics.” 

Renzon agrees on the importance of channels such as WhatsApp and other super apps, but believes that the key to hyper-personalisation lies in context. “Implementing predictive analytics, AI and intelligent loyalty programmes without context will always lead to consumer disappointment, and ultimately to companies going back to square one.” 

Carvalho agrees, suggesting that the world is entering an era of real-time, contextual rewards. “Technology will enable the best reward at the right moment, creating seamless and engaging experiences that not only reward customers based on their changing preferences, but also keep them loyal. 

“What it boils down to is that companies who aren’t utilising AI and predictive analytics should start sooner rather than later, as the opportunity to maximise client satisfaction and drive client lifetime value – and thus to differentiate your business while maximising customer loyalty – is huge,” Carvalho concludes.