Author

Guest

Guest contributor

This week’s guest blog post is contributed by Fred Bergklo, International Team Lead at Nosto. Fred shares how to get started with improving your testing efforts: to drive sales and improve performance on your eCommerce site. 

Most retailers are familiar with the concept of A/B testing and its potential to improve engagement, conversion and revenue. However, as the eCommerce industry continues to evolve, testing and optimisation strategies do as well — which means you may be missing out on a world of new and advanced practices that can increase your sales. From the placement and design of product recommendation to banner images on homepages, there are a number of site elements that can be tested to gain deeper insights about your customers and drive better shopping experiences.

1. Test and optimise your homepage banner

Most online stores display a hero banner above-the-fold on both mobile and desktop devices, which is meant to immediately draw in visitors. Given the large amount of space banners take up on the homepage, it’s important to make sure you’re making the best use of this space to minimise bounce rate and entice visitors to click on the banner.

In the example below, clothing brand O’Neills tested two banners: one which showcases general ‘men’s new arrivals’ and the other showcasing attire representing Ireland’s rugby team.

           

Homepage banner tests typically yield conclusive results. However, if your test shows signs of inconclusiveness, there may be a number of granular insights hidden beneath your customer behavioral data that can clue you in as to why there is no clear winner — and help you optimise your tests according to this data.

For example, perhaps the banner promoting Ireland’s rugby team’s apparel increases your click-through rate significantly for site visitors in Ireland. Alternatively, perhaps the banner doesn’t increase click-through rate but it does increase conversion rate of the apparel shown on the banner in terms of shoppers who do click through to browse more.

Understanding these granular customer segments allows you to make more informed decisions based on your customers’ behavior and curare your homepage banner according to the affinities of your different customer segments — leading to more highly personalised experiences for each of your banner variations.

2. Test and optimise product page recommendations

Placement

Product recommendations are often used to tackle multiple goals. In the case of cross-selling and upselling, testing and optimising the actual placement of your recommendations helps ensure that you’re making the most of your both of these strategies.

If the primary goal for your product recommendations is to achieve more upsells over cross-sells, then it makes sense to position your higher-ticket alternative products before your supplementary add-on products. Many retailers, such as ASOS, use a similar setup on their product pages:

However, showcasing upsell products before cross-sell products isn’t always the optimal option across the board, as different verticals yield different shopping journeys.

For example, furniture and home decor retailer Atkin and Thyme showcases cross-sell items — such as chairs that match a table — front and center in order to inspire shoppers to purchase entire sets or collections of furniture:

By testing which product recommendation layout strategy aligns and performs best with your specific goals, you can better identify the difference between shoppers who never purchase add-on products vs. those who do.

Design

Once you’ve optimised your price anchoring strategy and your recommendation placement, the only logical next step is to test and optimise the design of the recommendations themselves. Testing various product recommendation designs is a critical step in using testing to its fullest potential.

To illustrate the impact of this strategy, let’s look at some examples of product recommendation designs you can deploy in your testing strategy.

Sport apparel retailer Gymshark showcases a single product recommendation at a time, giving each recommendation the biggest possible on-screen real estate. At the same time, the set up also alludes to a carousel of other product recommendations that can be viewed with just a flick of a finger:

Outdoor apparel retailer Fjällsport includes two on-screen recommendations while alluding to more recommendations using a fading carousel design:

By testing product recommendation designs like the ones shown above, you can achieve a higher click-through rate, a lower product detail page bounce rate, or more cross-sells that can help you ultimately increase revenue potential.

3. Test and optimise the flow of your page narrative

It’s widely known that online shoppers rarely navigate to the bottom of a website — even on mobile devices where real estate is limited. This adds a lot of pressure on website designers and business owners to optimise the flow of their page narratives. For example, should a ‘Best Seller’ product recommendation be deployed immediately below your hero banner or would category promo boxes work better?

Testing and adjusting elements to achieve an optimal page flow experience is a simple, but highly effective testing strategy. By changing the placement of your personalised onsite content, you can easily determine which layout performs better than any of the others.

4. Test and optimise your product recommendation algorithms

Finally, there are a number of product recommendation algorithms available to yield the most relevant results to your shoppers. To really optimise the performance of your product recommendation algorithms, the best possible strategy is to test how different recommendation algorithms perform according to your specific business needs.

To illustrate how you can test and optimise product recommendation algorithm, we’ll look at three algorithms that Nosto’s clients are often curious about optimising.

Bought together algorithm

The bought together algorithm is a very simple model that recommends products that shoppers actually buy together regularly. In order for it to work efficiently, there needs to be a fair amount of sales volume, as the algorithm solely takes into account products that have actually been sold in the same order.

As we see in the example below, products that are most commonly bought together can vary greatly due to the fact shoppers don’t typically purchase multiple versions of the same product:

Viewed together algorithm

The viewed together algorithm is similar in that it recommends products that shoppers explore within the same site visit. This model doesn’t reflect actual buying behavior, but browsing behavior instead. Typically, the end result is a list of very similar products, so it’s a commonly used option for product page cross-sellers when a customer is arguably still contemplating different product options. As we see in the example below, products that are most commonly viewed together are often very similar:

Score-based algorithm

The Relationship score-based algorithm strikes a balance between these two above as it tracks both actions by giving a relatively low relation score for products that have been viewed together, and a larger score for products that are actually bought together. As we see in the example below, this model ends up displaying both different versions of the same product and completely different products altogether:

A rule of thumb for choosing the right default configuration depends on the use case and your store. Sites with smaller sales volumes are likely better off using viewed together and relationship score-based algorithm, while sites with high sales volumes can get more out of the bought together algorithm.

That said, cart page recommendations are likely to work better when dictated by the bought together algorithm, while product page recommendations are more likely to perform under the viewed together rules. Like everything else on a website, this can change according to the vertical, store, or even segment. That’s what makes this a great test to run.

Eager for More Onsite Testing Strategy Techniques to Optimise and Improve Personalisation and Performance?
When optimising your online store, the testing possibilities can be endless. The most important rule to remember is that you should always choose tests that align best with your business goals and always continue to optimise these tests to improve your results. If you’re curious to learn more testing strategies and how you can get started with testing and optimisation on your eCommerce site, get in touch with the Nosto team for a first-hand look.

Subscribe to our newsletter

Get updates, news, industry events and tips to increase conversion rate.

Top