Product descriptions are a great way to maximize conversions because what a customer reads about the product influences whether or not she’ll buy. A compelling product description describes not only in what the product does but also what the product does for her.
Fashion brand LK Bennett strikes a great balance between listing the product details and describing the great things about buying this dress (highlights mine) –
Kara is a modern evening dress – a contemporary riff on T-shirt silhouettes. Knee length, with short sleeves, it is rendered in a rich green velvet. Its bodycon fit featuring ruching detail at the side seam, for a flattering finish.
The product description also contains valuable marketing signals. Marketing professionals already use segmentation and business rules to help improve newsletter personalization, but marketers have also become increasingly interested in the ability for artificial intelligence to revolutionize their profession, and one of the hottest areas is called “Natural Language Processing,” which entails teaching computers to understand the nuances of human language.
In the jargon, text is known as ‘unstructured’ data because it doesn’t come in numerical values and therefore is difficult to convert into tabular data that can be stored in a marketing database. Yet growing computing power and advances in programming have enabled data scientists to transform product description text into a goldmine for marketers.
In the field of Machine Learning, Natural Language Processing, often shortened to NLP, is gaining widescale attention from marketers. As an example of best practice, Spotify uses NLP to systematically scan a listener’s listening history – scoring each song with a variety of attributes and using that data to store/update a user’s musical tastes. Spotify then uses this semantic profile to ‘surprise and delight’ listeners with a stream of fresh and relevant music that is ranked according to that listener’s unique and evolving musical tastes. Applying this technology to email merchandising is the next step in NLP’s evolution. To date, NLP has been used only by global retailers, but in today’s retail environment every retailer with many products and many customers are seeking novel ways to use NLP to surprise and delight customers with relevant products and boost marketing ROI.
Retailers and e-commerce companies can use NLP by doing what Spotify, Netflix, Google, and Amazon are already doing very successfully: proactively recognize each customer, understand their unique tastes and preferences, and make individualized recommendations.
Consumer engagement data is being produced and captured at an exponentially increasing rate, and NLP is a critical strategy to help serve customers better. Retailers are using NLP to infer a customer’s tastes and make recommendations based on those tastes because retailers that cannot understand each customer’s unique tastes and preferences risk falling behind.
Consumers expect a more individualised retail experience.
Retailing is going through a period of intense change, and NLP is one of the most potent tools a retailer can use to stay ahead of Amazon and other global retailers.
NLP can help with email personalization and enables you to add a human touch to boost your site’s shopping experience. NLP can help smash the glass ceiling of rules-based personalization and provides the ability to talk with customers as if there was an attentive and helpful personal shopper on the other side of the screen.
Dress credit: LK Bennett
Cover image credit: do androids read robot book? http://d4n13l3.deviantart.com/