For retailers and e-tailers gearing up for the 2020 holiday season, we are seeing across the board, the dramatic conversion of shoppers into buyers as a result of investments in e-commerce technology. While physical locations are opening up, socially distanced shopping during the holiday season will be particularly challenging. Many of us, particularly those vulnerable populations, have found convenience and safety in digital commerce. Researchers forecast that this consumer preference for online shopping is here to stay, and especially critical for places experiencing a second lockdown.
We put together a list of technology solutions to ensure your online store is optimized to provide positive experiences for customers and a supportive alternative to in-person shopping.
For many years shoppers have suffered at the hands of poor on-site search solutions and their inability to deliver accurate search results. However, that should not be the case anymore as data science has come to the rescue of shoppers. Now, using AI tools we can apply image analysis to identify and tag just about every feature that can be recognised on a product.
Let’s consider fashion, specifically dresses, not only can we identify a feature, such as ‘sleeve type” but also the specific value of that feature such as ‘juliette.” At Velou we can identify 29 additional sleeve types using AI.
It’s easy to see how this ability to enrich data using image analysis allows us to create dozens of new tags for a product, which in turn makes it easier to find that product when searching.
For just as many years, those same search solution have been relying on key word matching of search terms with product descriptions to find products, which is fine if shoppers are using very basic search terms such as “dress” or “red dress.” Anything more complex often results in “no products found that match your search” or ‘here are 10,000 products that (don’t) match your search.” I’m sure you will agree that these are poor user experiences. Again, data science has come to the rescue and using Natural Language Processing techniques that consider the semantic intent, or the interpretation, of the searched term we are able to consider what the shopper ‘might’ be looking for. Then, we can broaden the search criteria to deliver better search results.
If we consider the above example, the Velou solution would know that a shopper searching for a “puffed sleeve dress” would also be interested in seeing products with ‘juliette’ or ‘raglan’ sleeves. They’d be shown both options.
So what happens if a shopper finds a product, lets stick with the dress example, but isn’t 100% sure about that actual dress? Perhaps the pattern is too bold or the sleeve length is too long. Wouldn’t it be great to show other products that are ‘similar to’ that dress?
At Velou, we use data science and the power of image analysis to drive this product comparison and we’re able to show shoppers products that look very similar to the one that was found.
Facets have been a long standing part of the on-site product discovery process and shoppers are used to clicking boxes to help refine their product searches. But most sites simply throw up a long list of facets and expect the shopper to click through each of them in an effort to discover the product they so desire.
There is a better option. To apply what’s called dynamic faceting, which means the list of facets change automatically in line with the search criteria. For example, “red dress” shows a long list of facets whilst “red striped dress with long sleeves” will reduce the amount of facets and only leave the ones that still have options. Using this approach, shoppers can find the product they want easier and quicker.