The Future of Digital Retail: Interconnected Customer Data
Graph database pioneer Emil Eifrem explains why firms needs to embrace highly-personalised, data-driven digital brand experiences
Retailers are continually looking to help shoppers search for items that perfectly fit their needs. The problem 99% of the market now faces: doing it in a way that rivals the world’s biggest virtual store, Amazon.
Amazon has taught us the monetary value of being able to predict what else customers might want to buy, by analysing online sales data. It’s a lesson that any brand wishing to survive needs to learn, and apply, so as to provide consumers with intelligent, highly context-sensitive prompts.
Such hyper-personal recommendations can only be generated with the assistance of technology, as that’s the only way to embed more intelligence into a recommendation engine. eBay App for Google Assistant is a great example of just such AI powered recommendation technology. And it’s a prime example of the type of graph-powered hyper personalisation that brands need to offer.
The power of spotting connections
The system, which runs off voice commands on Google Assistant, works by asking qualifying questions so it can quickly serve up relevant product examples to choose from. Under the hood, accomplishing this requires a combination of NL (Natural Language) processing, ML (Machine Learning), accurate predictive analytics, a distributed, real-time storage and processing engine, underpinned by a graph database to make all the real-time data connections required.
The eBay App for Google Assistant,is a chatbot powered by knowledge graphs that supports conversational commerce – and is designed to provide a seamless, personalised shopping experience, enabling you to check out the prices of products you request and find the best deal, simply by asking. And as eBay’s Chief Product Officer has publicly stated, existing product searches and recommendation engines were unable to provide or infer the right contextual information within a shopping request, prompting his team to develop a bridge between regular search and natural language search using a graph database to process all the real-time data connections required.
Another example is London-based augmented reality e-marketing agency Quander. It’s all about mining users’ physical presence and movements in video and augmented reality experiences to offer more personalised ‘experiences' and better recommendations and get a deeper understanding of how customers interact with a brand. Again, the firm is using a graph platform for clients’ data collection to do everything from track how many times an event was visited, how many social media messages were opened, in order to identify deep patterns and analyse them to predict real-life outcomes. Finally, the data is used to make recommendations in context so tracking and curating individual experiences.
Graphs can help fend off the Amazon threat
Why do both of these hyper personalised recommendations systems use graph technology? Because it’s perfect for helping refine a search against inventory with context – a way of representing connections inside your data sources, based on shopper intent, while preserving the paths that they are likely to take. This allows the system to build up its internal profile of the customer and working with that portrait is the main way of generating its hyper-personal and relevant suggestions.
I predict we’ll see a lot more of these tech-boosted recommendations coming as brands have huge volumes of data, but are struggling to find the best way to use it to help their customers. If you are going to succeed in personalisation, you have to be able to leverage data connections and join the dots between the relationships – so why not look to graph technology to give you the full picture?
The author is CEO of Neo4j, the world’s leading graph database, Neo4j