Consumers face so much choice in online shopping that they can be paralysed into indecision.
AI allows brands to personalise the choices offered to consumers, converting indecision into sales.

The digital environment provides endless aisles and product variety to consumers, 24 hours a day, providing an overwhelming number of options. It is no surprise that shoppers prefer some of those choices to be eliminated for them, so they can choose between a manageable selection. In November 2017 a Demandware Report revealed that personalised product recommendations now drive up to 26% of e-commerce revenue.

Consumers like the personal touch, and recommendation engines can play the role of digital shopping assistants. Implementing personalisation for individual users requires the analysis of vast amounts of data, and the ability to deploy the results in real time. This demands a more sophisticated system than traditional fixed decision trees. This is where Artificial Intelligence (AI) techniques – and in particular, neural networks – offer key advances.

The most commonly known applications of AI, in this context, are the look-a-like product recommendations based on a user’s previous purchases or purchases from other users who bought similar items. These are used for prospecting, retargeting, cross-sell and up-sell.

Another set of AI techniques seek active responses from users to understand their personality and match those to products or services that they are most likely to purchase. For instance, Zenith is continuing its ground-breaking work in AI with a machine-learning application for a global client that is driving product conversions on a retail website. The AI-powered tool is a product-recommendation app that features the brand’s range of fragrances together with competitive products. Zenith’s algorithm continually assesses all consumer responses to a series of simple questions that lead to the fragrance recommendation. The algorithm determines how successful each recommendation is at converting and adjusts future recommendations accordingly. This means that when each consumer uses the app, it becomes ever more successful at driving conversions.

Soon, brands should – and will – extend the concept of personalisation beyond product selection and across the entire consumer experience, to maximise brand engagement and the desired response. Successful brands will design and execute with a customer context in mind for all touchpoints.

Machine-learning algorithms are being used in attribution models to optimise every paid digital touchpoint at the individual level. They ‘learn’ from historic data to understand what combination of touchpoints is most likely to generate a conversion. Similarly, AI techniques have enabled classification of individual components in any piece of content; text, pictures or videos. There is a great opportunity to link these applications of AI together with a recommendation engine.

An enhanced attribution capability that can model every touchpoint at the level of each creative component can provide greater insight into what works within the customer journey. The model can be further refined by including additional first-party or third-party audience data and actual shopping behaviour. The combination of these machine-learning applications with recommendation engines can ensure that the right content – creative and products or services offered – appears at each stage of the journey, and that each piece of content is as effective as possible at producing the desired consumer response at that stage. Such recommendation engines will retain freedom of choice for consumers but empower brands and marketers with techniques to structure these choices for greatest relevance.

Recommendation engines are already demonstrating their ability to drive better results across KPIs (e.g. for increased propensity to buy or higher average order values). By combining these systems with the knowledge of user journeys and content analysis, brands can create strategies across paid and owned marketing efforts that are more effective in sending users down the path of maximizing returns. AI and machine-learning can help brands unlock opportunities for personalised experiences that drive positive consumer decisions.

Related Insights Posts