When Implementing ‘Searchandising’ Best Practices, Test, Test and Test Again

Posted by Shaun Ryan, April 26th, 2011 at 1:11 pm PST
Categories: Site Search, Testing | No Comments »
 
   
 

At SLI, we spend a lot of time talking with our clients about ways they can improve performance of their site search and navigation, and options to consider for merchandising results pages. We’ve also discussed many best practices approaches in this newsletter – everything from refinement options, display options (e.g. ‘grid view vs. list view’), and advanced features like Rich Auto Complete and Mouse-over Pop-ups.

With so many features and options to consider, how do you know what you should implement, and if they’ll truly make a difference in the experience you deliver to your customers? The answer is pretty simple. You need to TEST.

If that sounds frightening, it shouldn’t. There are a few tools out there you can use (including our own Conversion Optimizer) which make A/B and multivariate testing of site search, navigation, promotions and page layouts pretty painless. But rather than focus on HOW to test (if you are an SLI customer and want guidance here please talk to your Customer Success Manager), we’re going to discuss WHAT to test. The options are wide-ranging, and there are many factors you should consider.

First off, how should you present search results on your page? Do you display items in a grid view or a list view? Should you offer your customers the option of choosing what view they like? There are guidelines you can follow – for example, if you have a lot of products that match particular keywords, a grid view allows more items to be shown above the fold. It also allows for less product detail to be included, avoiding clutter. The only way to tell what method will work best with your customers is to perform some tests.

Another option to test might be how best to display color options in search results. Should you take Jelly Belly’s approach, and offer a clickable color swatch along the left side of the page, or should you provide color options underneath each item in the search results, like Dover Saddlery does? There are many good ways to display color options and many factors to consider and test in implementing the solution that works best for you – like what products you sell, how many of them offer different color options and how they display.

How about video, and user ratings and reviews? These are popular features on e-commerce sites, and we’ve made recommendations to include this content in search results. How you include them and even IF you should include them are questions best answered by doing some tests.

Is your search box placement motivating enough of your users to use it? Research shows that people who search are 2-3 times more likely to convert, so would you see even more conversions if more of your visitors used your search box, and would a different placement, or enhancements to the search box itself, drive more usage?

Let’s take banners. We often recommend that retailers use banners on ‘no results’ pages to drive people towards popular items. Banners are also good ways to promote special offers you’re making, like discounts on shipping, brands on sale or other promotions. Where you show the banners, how big you make them and what content you include are all variables you should test.

There are many more things you can test, such as what refinement options you offer and how you present them (e.g. pull-down menu vs. clickable links), what size images you should include in search results, and whether you should add a “click to buy” button right on each search result. If you have any questions about what to test or how to conduct your tests, our team of experts is here to help – just contact us at sales@sli-systems.com.

   

Leave a Reply