Share this content

Welcome to the era of hyper-personalisation

4th Jan 2018
Share this content

Something major is about to disrupt the retail and services industry, warns graph database pioneer Emil Eifrem.

‘Customers who bought that also looked at this’ style recommendations have been a boon to the online retailer.

The problem: ‘Customers who bought that also looked at this’ style recommendations may soon be too basic and stop being of interest to the consumer. Digital consumers now actually expect personal recommendations based on their individual preferences, shopping history, interests and social context to be taken as read.

Retail is getting more and more competitive and these simple personalisation predictions about what else customers might want to buy need to become far more sophisticated. To survive in this more challenging context, and meet and surpass your customers’ expectations, retailers and service providers need to offer something much more advanced – more along the lines of, ‘You bought this item or service today and this last week, you’ve also looked at all these items and services, but today you are looking at these other things – so how about this?’ level of suggestion.

A much richer customer support scenario

The market needs to provide savvy consumers with intelligent, highly context-sensitive prompts. However, the reality is that these hyper-personal recommendations can only be generated with the assistance of technology as a way to embed far more intelligence and data-fed capability into your recommendation engine.

AI (Artificial Intelligence) and data-driven, real-time smart software is what is required – but the key enabling database technology which will allow these next generation recommendations to be drawn is graph database technology.

eBay’s AI-based ShopBot is a prime example of the type of this graph-powered hyper personalisation that brands need to move to in action.  

ShopBot begins with qualifying questions, and moves on to make suggestions on relevant product examples customers can select. The functionality allows the buyer to send the app a photo with a direction like, ‘I like these winter gloves, can you find similar for me?’ or ‘I am looking for a pair of blue Patagonia ski boots costing less than £100; please find me those’ – and it will figure out and display back similar products for you, and speedily as well.

To accomplish this requires a combination of ML (Machine Learning), accurate predictive analytics, a distributed, real-time storage and processing engine, with NLP (Natural Language Processing). But ideally it also needs a graph database to process all the real-time data connections required.

That’s because graph technology helps to refine the search against inventory with context – a way of representing connections inside your data sources based on shopper intent. This is the basis for the system building up its internal profile of the customer on the fly, and working with that portrait as the main way of generating its hyper-personal and relevant suggestions.

What’s more, that context is stored, so that ShopBot will recall this information for future interactions. So, when a shopper searches for ‘ski trousers’ for example, it knows what details to ask next like type, style, brand, budget or size – and as it builds this library of information by rapid traversal of the underlying graph database, the application is able to quickly identify specific product recommendations.  

Context is king

Tapping into human intent like this and delivering highly responsive, accurate help is what next generation, hyper-personalised recommendation engines have to look like.

That’s because without graph software at the centre, you can’t easily offer consumers the hyper-personalised hints. The traditional way of storing data is ‘store and retrieve’, but that doesn’t give you much in terms of context and connections. SQL queries are also complicated, and can’t deliver the information in real time – and for customer search to deliver useful recommendations, real-time contextual information has to be accessible.

That’s why relational isn’t the best way to build Recommendations 2.0 engines. The same caveat also applies to other Big Data technologies, which stumble at managing data connections, while these are what graph was actually designed to manage.

The message is clear: anyone with a web store needs to emulate eBay and use practical, applied AI to secure that successful retail future – and native graph is the most practical way of getting there.

The author is CEO of Neo4j, the world’s leading graph database



Replies (0)

Please login or register to join the discussion.

There are currently no replies, be the first to post a reply.