By Raj Badarinath, VP Marketing & Ecosystems, RichRelevance.
Increased competition for web traffic and the rising costs of customer acquisition continue to make e-commerce more challenging. With Code Freeze day soon approaching, marketers are “feeling the chill” to get the most innovative personalisation tactics implemented to ensure a great holiday sales season.
While personalisation is deemed to be the most obvious solution, it works best when a retailer has tonnes of behavioural data for each product.
But how does a retailer personalise for new products or ones that are considered “long-tail”? How do they ensure that this inventory is immediately visible to customers in through search or recommendations? Well, you couldn’t until recently... not without hyper-personalisation.
Many e-commerce traditional recommendation engines rely on consumer behaviour patterns, which are not very useful for new products that have no browsing or purchase histories.
Customers today expect highly personalised experiences, with products and services tailored to their individual affinities. Personalisation, at its core, provides customers what they want, when they want it. With this in mind, hyper-personalisation is knowing and serving your customer as an individual, and not as part of a rough, probabilistic segment. In effect, hyper-personalisation is the intelligence of a platform that creates a “segment of one” for each customer, by understanding their likes, preferences and behaviour, inspiring them to discover more products and content that will speak to their individual needs at the moment of shopping, and providing a useful , pleasant and totally unique shopping experience.
The concept of understanding human language and symbols dates back to the 1950s, with Alan Turing’s paper “Computer Machinery and Intelligence”, where he asked: Can machines think? What he meant was: Can machines think like humans, therefore can machines understand facts and data from human reality?
NLP (Natural Language Processing), a technology that allows for unstructured data to be understood and processed by computers, is a form of artificial intelligence which uses several disciplines, like computational linguistics, to try to make structured sense of unstructured human communication. In the beginning, NLP drew heavily from Linguistics and Logic.
NLP take in text data including all rich feature characteristics (descriptions, styles, ratings & reviews, etc.) associated with a product and it then uses it to create new data science models that have a product graph, which is a techy way to saying that all related products are linked together in a tapestry.
In a sense, NLP enables computers to ‘understand’ many human languages, execute spoken or written orders, infer insights from them and apply them to the future. For example, try searching for a red jumper online. The computer would understand that you want to buy a red jumper, and would show you all the items it knows of corresponding to “red” + ”jumper”. With NLP, the factors would expand to not only red jumpers, but also to other factors that suit your own preferences: for women, organic materials, sustainable fashion, and others.
To display such an array, NLP analyses components of human language words or phrases or descriptions from product catalogues to compare them to its own database of words and recordings, which it categorises and extracts meaning from them using semantic analysis - in effect, understanding them. This means it can make connections humans may not.
NLP to the rescue for retailers
During the holiday season every retailer adds new inventory to attract customers. Traditionally, retailers would approach this by creating manual marketing methods to boost sales, but this approach comes with its own set of drawbacks like loss of relevant product discoverability, unrelated products showing up, resulting in a cluttered and inefficient customer experience.
As customers continue to interact with the brands and selections, the solution gets better and better over time at ensuring the right product recommendations for the given context of each shopper – whether they are online, in store, on mobile, or on email. NLP is a game-changer for retailers/brands that regularly introduce or quickly cycle through new styles or seasonal catalogs, making it possible to personalize and expose shoppers to relevant products through cross-sell offers from the start, as well as increase the relevance of the recommendations they see as they move through the funnel.
Now is the time to get NLP
Days are becoming shorter, and marketers can sense a chill in the air. The season of the dreaded yet critical "code freeze" is rapidly approaching. The shopping holidays are drawing close - and the key issue for retail now is Code Freeze.
Code freezes are a lock down of a website that keeps IT from making any major changes. This hold on change minimises the risk of introducing flaws to a site's code that could lead to errors and glitches, downtime and purchase interruptions. For retailers and e-commerce businesses, code freeze is meant to insulate the company from missing any potential revenue during a critical period of time for the business, and avoiding a nightmare situation when what should be a routine change impacts production, and retailers miss out on major sale events or fails to give the customer the experience they are looking for, or whatever a holiday shopping nightmare would be.
This year code freeze day is just around the corner!
Implementing technologies like NLP must be done before the code freeze as NLP can prove to be a lifeline for marketers during the holidays, providing an outlet for the long tail selection, campaigns, new product lines and generally making the consumer experience much better.
The most critical thing to remember as we head towards Code Freeze Day is this: Holiday shopping is seasonal but loyalty is lasting, so make the right technology investments now to make that a reality.