Testing for Success: Optimising Site Search for Conversions

By Mark Brixton | 14 Mar 2012

Mark Brixton discusses why testing and conversion optimisation are crucial for not only an optimised online experience, but to ensure efficient site search and navigation functionality.

With various search engines, such as Google and Yahoo, offering consumers instantaneous ways in which to navigate online, it’s easy to understand how they have the same expectations of the search function on the websites they visit.

In terms of optimising the online customer experience, testing has been revealed to be a major contributor to enhancing this offering. In the same way retailers experiment with different aspects of their websites, such as page loading times, placement of information and calls to action, site search and navigation must also be tested and tweaked accordingly.

To Test or Not to Test?

For retailers who aren’t sure about investing in testing their site search or navigation, it’s getting harder and harder to ignore the data that says otherwise:

  • Avail Intelligence’s 2010 Searchandising and Recommendation Report revealed that when users search for a product, site traffic typically splits up with 10% utilising the search box, and 90% using category filters.
  • MarketingSherpa research indicated that people who use site search are more likely to convert at two to three times the rate of those who don’t use it.

In addition to this, if these facets are tested and optimised, businesses are provided with the perfect channel through which to gather invaluable data about on-site browsers and their behaviour. With site search users being objective focused – that is, looking for something specific – potential customers are telling online businesses exactly what they want in their own language. It simply doesn’t get any clearer than that, and conversions are far more likely if the results are relevant.

Tweaking for Success

The location of the site search box itself affects customer behaviour and of course, can be examined and studied in order to discover its prime position. A popular intimate apparel online retailer found that although customers who used its site search function were more likely to convert, this tool was not being used effectively due to the difficulty in locating the search box on the website.

After the implementation of SLI System’s Learning Search solution and placing the search box in a more prominent position, this change increased the use of the retailer’s site search by 600%, as well as contributing to a 70% lift in revenue from customers using the function.

Tools of the Trade

There are many tools available that will eliminate the guesswork about the best ways to optimise and refine site search and navigation, without adding to the workload. By employing these tools, variations in site design, formatting and layout can be tested, while also viewing how such changes impact visitor behavior. For example, different options in the layout of a website’s search results can be tested, in order to decide on the best sized product thumbnail images to show, or whether a grid or list view is most effective.

When looking for a solution to assess these types of deliverables, companies do not need to waste time on building individual tests. Software, such as the Conversion Optimizer, allows users to define the features and options to test and determine how many site visitors will view each page variation. Retailers can select the part of the conversion funnel they want to optimise – from counts of customers who see particular pages (e.g. product pages), to the total value of products added to shopping carts or the revenue from completed carts.

Retailers should also consider solutions that allow for easy stoppage of entire tests and automated stoppage of any variations that are performing poorly. In addition these solutions should incorporate access to reporting that is both easy to understand and presents actionable data. Utilising tools of this calibre will allow total control over the tests being performed, without the headache of the complex configuration steps required by other testing systems.

0 Comment

Leave a Reply

Your email address will not be published. Required fields are marked *