1. Type ahead and autocompletes
The moment that somebody interacts with your search bar is the time to get personal with them. The image below is of our client, Mole Valley Farmers and as soon as we activate the search bar we can start to drive people towards product discovery. We can start to show them trending products, popular suggestions etc and if they've been on the site before we can show them recent searches. So again, we can be very personable to the actual individual. It’s a similar thing with e-commerce search autocomplete. But, here the search bar uses each keystroke to bring suggestions and products back to the client. The idea here is we're matching the keywords of the products in people's inventory, we're putting it visibly to them. This is key, as we know from the stats that we generate people that look at the search suggestions. These users typically have 2-3x higher conversion rates and people that click on the actual images have 5-7x the conversion rate.
Now, you may have users to your website from lots of different locations, even regional variations within the UK. In this case, the way that you attribute your data can actually have a quite different effect to how your users actually search. For example, a jewellery store that uses the search term hairband but not headband may get zero results. It's all about understanding your search data, examining where people are having a good experience and where they could be getting zero results. If you can't enrich the search with e-commerce search synonyms, they're likely to get a zero results page. In this case they're likely to go back to Google and off to a competitor.
3. Keyword terms
Look at how people are searching on your site and look at the value of how that leads through to revenue. The words that people use on your search are very often the same types of words that they'll use on Google. For example, instead of laptop brands people may search keywords like silver laptops, pink netbooks, quad core laptops. So if you know that these users are converting very well, you can divert that headline spend into the more conversational keyword terms. You know that once you get them to your site they will most likely convert. SEO is another thing to think about here as you may not have a silver laptop or a pink netbook collection. In this case, it may well be worth building some SEO around that and getting it into your site. Because again, these are words that people are searching for on your website, they're more than likely to be searching for it on Google. So utilise these ecommerce keyword terms, utilise the value of them for your own SEO and PPC benefit.
4. Avoid a zero results
This is key and should be avoided when possible. Below we’ve got an example from Missy Empire. Here, if a user generates zero search results, it has personalised the language that it uses based on its audience. But, rather than avoid getting them to a zero results page, where users are much more likely to bounce, we're able to re-merchandise them. So, showcase popular products, popular searches, trying to re-engage them and direct them to products that they may like.
5. Natural language processing
What we mean by natural language processing is like searching for "men's vans under £50", we're looking at very specific options here, in regards to gender, brand, price range, etc. There's a huge market in this, we know that with Alexa and Siri voice search has gotten much more prominent. So one of the things that we have developed in Klevu is utilising Google Voice Search. Our client Cox & Cox have a microphone option in their search bar and utilise natural language processing to make sure they understand each keyword. By understanding the keyword term, understanding natural language processing, e-commerce stores can directly go through a number of different aspects of the key word and instantly drive clients to the product.
6. Search Filters
We’ve all used filters before. But those little drop-down menus? They’re old news. AI-driven dynamic filters automatically create the most relevant categories and display them to the customer, without leaving the search bar. Smart filters allow shoppers to intuitively discover their next buy alongside the search bar without experiencing the annoyance of pop-ups or drop-downs layered over the products during navigation. So, rather than having a separate box that hides the product listings behind it during navigation (a typical drop-down menu), products are always visually on display during search. Not only does the AI driving search learn what customers want, but it also learns how to filter and categorise the results based on a range of metrics, becoming more personalised as you search.
Overall, these 6 steps are designed to markedly improve your customers ecommerce search experience on your website. It’s important to remember that search often represents a great opportunity to increase conversion rates, improve customer experiences, and control levels of exposure for specific products. While search may seem like a small part of your ecommerce website, when implemented effectively, it can improve your site experience and provide you with a ton of data about your users and how they navigate your website.
Klevu’s AI and NLP-powered discovery suite includes on-site Smart Search, Smart Category Merchandising, Personalisation and Recommendations enabling ecommerce websites to deliver shopping experiences powered by real-time buyer intent.