his was an amazing year for advancement in technologies all around – Amazon launched its first few self service retail stores, self driving cars started negotiating routes with other cars on the road, telematics based pay per mile insurance picked up a lot of steam. A lot of technology work that had been in research labs for a long time made their debut this year.
Keeping pace with the evolving technologies, search solutions made some incredible advances of their own, particularly around eCommerce. If you are looking at rolling out a new eCommerce search solution in the new year, you should pick a solution which can evolve at the same pace as machine learning and AI solutions.
The eCommerce search solution of 2017 should:
1. Improve customer experience on its own
2. Learn from ‘after purchase’ experience of customers
3. Help merchandizers improve product placement
4. See Products from customer’s perspective using images.
A few years ago, it was extremely costly to implement introspection solution in search engines that let them figure out what customers do not want and adapt from there. This year, one such implementation, visual navigation started becoming mainstream:
Instead of overwhelming customers with a long list of filters and facets on the left hand side, search engines started figuring out what really matters to customers from that list. This knowledge was then used for providing selective visual filters to help customers find the right product easily. eCommerce websites with large catalogs have seen as high as 70% more engagement with visual filters as compared to standard filters.
After purchase customer feedback:
Most search engines have incorporated feedback loops to learn from user’s interactions – clicks and browse. This year, this went one step further – search engines extended their learning by extracting insights from product reviews left by customers after using the product. This helped them get a view of what customers liked about a product, what other products the customer used it with and what specifically the product was used for. Powered by these new insights, search engines can now cater to customers who are looking for a solution to their need instead of a product.
We have always looked at eCommerce search engines as sales agent for the website. Besides selling products, sales agents also bring feedback from customer’s perspective to the product team.
Imagine you run a brick and mortar store that sells active wear. Customers buy your products for working out in gym, yoga, running, etc. Over a period of time, you start noticing that a lot of customers start asking your sales folks if they have any ‘Zumba pants’. After figuring out that it’s a new trend amongst the customers, the sales team starts branding your yoga pants as Zumba pants because they are designed for similar activity.
In the online world, this feedback has now started flowing back from the search engine back to the merchandizing teams in form new trends. Some search engines take it further by automatically tagging products that the customers end up buying after searching for new trends. But it also opens up new organic SEO and branding opportunities for the merchandizing teams.
Product Image Analytics:
The advancement in image recognition algorithms have provided much needed vision capabilities to search engines. While a lot of search engines implemented image search this year, the real value came in from the search engine being able to ‘see’ a product and relate it with other products. This enabled them to figure out what’s unique about a product or what other products look similar to a given product. This information has opened up new opportunities for cross sell and up sell.
Search technologies made a lot of advances in 2016 riding on the back of machine learning. The new year promises to build on top of that to impact online sales dramatically. Make sure your choice of search solution sets you up to reap the rewards of these advances.