Ease of access to Artificial Intelligence (AI) and Machine Learning (ML) platforms is making it easier than ever for brands to gain an exponentially greater understanding of the needs, wants and behaviours of their customers – giving them unprecedented insight into just what makes the users of their products or services, tick.

These technologies are even more useful when optimising brand loyalty or rewards programs – the customer has already indicated an interest in, or preference for, the brand, so it’s a question of retaining them and delivering more, and better, opportunities for them to interact with the brand.

Artificial Intelligence

AI is a field of computer science associated with the concept of ‘teaching’ machines to ‘think like humans’ in a bid to get them to learn, solve problems, reason, plan and identify patterns, but using much larger data sets and delivering quicker responses than any human could. ML, on the other hand, is the subset of AI concerned with creating algorithms that can recognize those patterns in the swathes of data and drawing useful conclusions.

Ultimately, the aim is greater personalisation, which will boost the customer’s experience – in the hope of upping conversion rates. Amazon.com is the textbook case study, with each site visit, product view, purchase, rating, review, shortlisting and referral delivering insights into customer behaviour, resulting in an eerily-accurate set of suggestions for future purchases. One type of software that analyses the information and delivers the recommendations is part of the Amazon Web Services (AWS) suite.

Tracking Loyalty

South African company Synthesis, a subsidiary of JSE Listed Fintech firm Capital Appreciation, is EMEA’s top-rated AWS provider, facilitating the digital loyalty intersection between companies like Standard Bank, Discovery Vitality and Absa, and their clients. These kinds of loyalty programs require systems which can interrogate large volumes of data, quickly, from the data points produced by the participation of hundreds of thousands, and even millions of users. AI and ML make managing all that data, much simpler – and then coupling those systems to could computing allows companies to cater for quick and cost-effective scalability as their user base grows. “AWS cloud software powers some of the biggest recommendation engines in the world – Amazon.com being one example, and Netflix, another,” says Synthesis MD Michael Shapiro. “There are many real benefits offered by AWS, but a major one is the ability to provide personal recommendations to users, based on their interaction with the relevant platforms”. Amazon’s SageMaker and PinPoint are the two software components at the heart of this kind of recommendation service, with the former giving developers the ability to wrangle algorithms in the ML workflow to track data and the latter using that data to communicate offerings to customers with push messages – the recommendations themselves.

Enabling this level of functionality allows companies to segment customers and create profiles, allowing them to gain the information hey need to personalize incentives and keep customers engaged. The ability to create ‘customer cohorts’ based on purchasing trends opens up the opportunity to offer personalised incentives that speak directly to their needs, rather than offering a broad spectrum of offers that may actually alienate potential customers.

Synthesis use the AWS platform to provide the scalability required by Discovery’s Vitality platform that has helped make it a global success. Outside of South Africa, Vitality is a bolt-on facility currently operated in 15 countries by seven global insurers to add value to their clients, from North America to Japan and Germany. “Korean insurers AIA launched Vitality as ‘free-to-air’ product for several months, meaning that anyone could sign up and reap the program’s benefits,” says Shapiro. “AIA were subsequently able to gauge the behaviour of the consumers who signed up, and then offer an insurance product based on the real data they’d collected via ML. The benefit to consumers was competitive pricing, based on their actual behaviour – and for the company, a working knowledge of their client base to help them manage their risk and provide more intelligent products. This kind of thing is impossible to construct without using cloud-based systems”.

As presented by Standard Bank at the AWS Cape Town Summit in July 2019, when they were looking to find a way to use multiple data points to react in real time to events meaningful to their customers, Synthesis were able to step in with a solution that pulled together historical data and opt-in customer information to provide more targeted solutions. “If, for example, a Standard Bank customer has linked their social media account to their customer profile, and they tweet about a burst geyser at home, the system is now able to send them an offer for short-term insurance, on their communication channel of choice,” says Shapiro. “This allows the bank to offer genuine solutions to their customers, in real time, in the way they want to be contacted – delivering a much better and more authentic customer experience”. Providing this kind of experience is only possible from an economic perspective with the introduction of cloud services. Allowing Standard Bank to elastically expand without having to buy infrastructure and and host it, in-house.

South African Solutions

Dashpay is a payment technology solutions provider which offers electronic Point of Sale (POS) devices that are able to facilitate ‘universal acquiring’ – the use of a single device to be used by multiple parties to offer an array of unique and differentiated products and services. One of the company’s B2B solutions effectively allows them to target credit card users as though they were a part of a loyalty program – even without signing up for a specific loyalty card – based on their purchase behaviour. They currently offer this kind of service to companies like Illy, Lavazza, Harley Davidson and SASOL, locally.

Dashpay’s Mike Rowley says that the company is working on an AI system that will track historic spend data – including what, when, where, how and how often consumers spend on particular products – to allow companies to deliver highly-targeted offers. “Look at the fuel space, for example. People often fill up at a petrol station because it’s the nearest or most convenient one, rather than being driven by any particular brand loyalty,” he says. “While many petrol brands have partnered with existing loyalty programs, what we’re looking to build, using AI, is a platform that understands when a consumer will need to fill up their tank, and is then able to deliver a targeted offer that gives them a specific benefit in going to a particular brand’s forecourt – like a deal on products from the shop attached to the petrol station. This allows brands to drive revenue streams from outside their core offering, as one example”.

There are B2B applications too – Dashpay is able to deliver incentives to customers who regularly buy coffee from one of their clients, Lavazza. “We’re now able to incentivize Customer A to go to Store B to buy Lavazza coffee – and when they buy their 9^th cup, we can give them an instant digital voucher to have a 10^th, for free,” he says. “This gives a business like Lavazza real insight into where, how and how often the end consumer is buying their product – invaluable information that they never had access to before”.

Shapiro says that, ultimately, using AI to streamline loyalty & rewards programs has benefits for both businesses and consumers. “For the business, the loyalty program becomes an unprecedented source of rich insights about their customers, allowing them to better serve their needs. And as customers, the appeal of a personalized deal is compelling in that it better meets our needs, rather than a one-size-fits-all solution”. Adding a loyalty program offers far more insight into what – and how – customers are buying, as opposed to one-off purchases by transactional customers. “The loyalty program becomes the lynchpin holding everything together, giving the business invaluable data to help them discern which specials they should be running, when, and what kind of offering their customers are seeking out”.

Rewards Review

• U.S. consumers hold 3.8 billion memberships in customer loyalty programs (18 per household).
• 81% of consumers agree that loyalty programs make them more likely to continue doing business with a brand.
• Redeemers are twice as satisfied with loyalty programs as non-redeemers.
• 86% of consumers who like a loyalty program will shop more, and of those, 58% will shop 15% or more with their retailer/brand of choice.
• 73% of members are more likely to recommend and say good things about brands with good loyalty programs.

(From Macorr.com)

*A version of this article appeared in the 2019 Sunday Times Loyalty & Rewards supplement.