Imagine there are tons of new arrivals on your favourite ecom shop, but all you want in the world is a cotton parka. It’s not much to ask right? Yet many ecommerce site search engines fail to deliver on that basic level of precision.
So, what if you just personalise search suggestions with the use of machine learning? Analyse what users are searching and offer users a way to narrow down their choices inside a particular category. Like our example. Just a quick concept of how things could be and to give you some ideas 🙂
We believe search is super important, because usually numbers don’t lie:
- Search users are much more likely to convert. In fact, on average they are 7-10 times more likely to convert than regular webpage visitors.
- Search users can make up about 10-40% of all the users on your eCommerce site.
- Which means that 40% to 80% of your online revenue is probably generated by these search users. Mind blowing right!
- Still 70% of (desktop) eCommerce search implementations are unable to return relevant results for product-type synonyms.
- What’s even worse is that on average, 10%-25% of all searches end on a 0-results page because most often, search engines aren’t able to process language – this means that shoppers have to search using the exact same jargon as the site does. When shoppers end up on a 0-results page, you’re essentially telling them you don’t carry the product they want to buy, which means they’ll purchase the product elsewhere.
If you’d like to know more you can take a look at this article – top 25 ecommerce site search best practices
Brought to you by: