How Artificial Intelligence Can Optimise Customer LoyaltyHow Artificial Intelligence Can Optimise Customer Loyalty
How Artificial Intelligence Can Optimise Customer Loyalty
December 18, 2018
by Sal Visca
Every successful marketing department needs to ask themselves at least two critical questions. Firstly, “what do our customers really want?”; and secondly, “which of our products or services best meets an individual customer’s needs?”.
To answer these questions, marketers can use artificial intelligence (AI) to create incredibly personalised customer experiences and targeted offers that go beyond selling products and services and develop the holy-grail of long-term loyalty and customer lifetime value.
It improves human to computer communications
Natural language processing (NLP) and machine learning (ML) are capabilities within the broader area of AI that can be used to improve product discoverability, overall customer experience and better targeting for merchants.
Speech recognition, NLP along with Natural Language Understanding (NLU), Natural Language Generation and Text-to-Speech are all capabilities that allow people to use spoken and written phrases to interact with computer systems – familiar examples include Siri, Google Assistant and Alexa. It allows consumers to directly communicate requests rather than having to communicate via a user interface on a device (e.g. mobile phone, tablet, laptop etc.).
Fundamentally, NLP translates the spoken word to capture the intent of the user, which when used alongside the user’s context, can make it seem as though the intelligence is more than just “artificial”. For example, if you were to ask Siri “how cold is it outside?” the technology would use your GPS location to determine the temperature outside and advise accordingly.
This could be enhanced further by syncing with your calendar for example, which would be able to determine that you’re flying to another city and proactively provide information about the weather there. Or remind you to pick up your dry cleaning if your warmer coat is ready for collection, which your autonomous vehicle can plot a route via on your way to the airport. Then AI can open up opportunities to inject commerce into these experiences, like offering a coupon or promotion that can be used at for your dry cleaning or at the airport lounge.
Additional cognitive methods such as Deep Learning and Predictive Modelling can take this even further. While advanced supervised and unsupervised ML techniques can be used to create advanced models and algorithms that will adaptively learn and get better at predicting outcomes.
As a result, commerce systems enhanced by ML and AI can intelligently cross-sell and up-sell compelling offers, which increases conversion rates, total order value and ultimately customer loyalty and customer lifetime value.
Related: The AI Customer Revolution Is Here
It can help you identify what each customer cares about
Marketers are already taking advantage of big data and predictive analytics for personalisation – aggregating purchase behaviour to identify patterns and anomalies. The logic follows that the bigger the data sets and more powerful algorithms used, the higher the probability the recommendations will be appreciated by a greater percentage of customers. For example, if a segment of people bought ski boots and table tennis bats, those that buy either will have been promoted the other. But with machine and deep learning personalisation it becomes more individualised in real-time.
Personalisation engines combine individual behaviour with macro information – tracking their behaviour across every touchpoint with brands to suggest products and services based on past interests. This brings together known customer information with intent information based on ML analysis, which incorporates things such as time spent browsing, clicks, scrolling and inactivity to determine preferences and interest levels.
The more an individual browses and purchases with a brand the better the personalisation becomes as each offer learns from the last. This is because different algorithms generate custom recommendations and configure a chain of “next best offers” to ultimately achieve sales. Consumers may be more open to providing insights into personal interests or considerations for purchasing if these systems do actually adapt, learn and find products, services and offers that happily surprise them.
It enables you to be dynamic with your pricing
One of the biggest challenges pricing managers face is daily dynamic fluctuations, a common example being Uber during high-demand hours. For e-commerce companies, platforms can help manage this supply and demand with ML that can segment customer segments. It gives the business the power to better pinpoint the price customers are willing to pay at the individual level, at any time. It will also examine customer intent information and ascertain what price point the customer would be willing to pay, pushing the customer over the purchasing threshold. However, multiple pieces of data need to come together including historical customer interactions and purchases, aggregate browsing behaviours and company information such as product information and pricing options and inventory and supply.
As a result, companies are empowered to ‘fine-tune’ pricing as customers interact with them, leading to the determination of more appropriate pricing. This dynamic real-time pricing optimisation coupled with deep personalisation should be the objective for all brands looking to create strong customer loyalty while balancing increased revenues.
Perhaps an overlooked benefit of AI is how it changes the narrative in e-commerce and customer service away from clunky chatbots and towards more intuitive and natural systems. AI/ML technologies unlock hidden patterns in user behaviour to “learn” intent and the next steps to take, or which offers have the best chances of success. While leveraging NLP, the customer also receives a better experience and the business gains valuable insight. It’s a win-win for both the customer and brands.
Sal Visca is the Chief Technology Officer for Elastic Path