As reported by Marketing Sherpa, the folks at Barilliance mined data from 1.5 billion shopping sessions over the first three months of 2015 to determine which product recommendation options generated the most revenue. On the sites studied, 11.5% of revenue was attributed to product recommendations.
Since not every recommendation tool was applied on all sites, we measured the “effectiveness rating” as a way of looking at revenue relative to usage. The higher the effectiveness rating, the better the performance.
- "Visitors who viewed this product ultimately bought" (effectiveness rating 3.42)
While this option generated the second biggest boost in revenue (25%) it was used by only 7.3% of the sites, suggesting that it may be most effective at turning visitors into customers and demonstrating the value of social proof.
- "Visitors who viewed this product also viewed" (effectiveness rating 1.19)
Over 68% of the revenue was generated by this option. It was also the most common recommendation option, used by more than 57% of sites in the study.
- “Trending in category” (effectiveness rating 0.84)
10% of revenue gain is attributed to this option. 12% of sites use it. It appears only on category pages and is another example of the power of social proof.
- “Customers who bought this product also bought” (effectiveness rating 0.67)
While less than 12% of stores used this option, it generated almost 8% of the revenue. There’s a fairly subtle difference between this and "Visitors who viewed this product ultimately bought" but the benefits are far less.
- “Recently viewed” (effectiveness rating 0.57)
About 12% of revenue, used by 21.5% of the stores. “Recently viewed” recommendations appear site-wide.
- “You might also like” (effectiveness rating 0.38)
Generating over 16% of the revenue, it was used by over 42% of the sites. The second most commonly used option but among the least effective.
“Top Sellers” (effectiveness rating 0.34)
Top Sellers generated 8.5% of the revenue but, as it was used by 25.3% of the stores, it’s one of the least effective types of recommendation options.
"Visitors who viewed this product ultimately bought" recommendations generated 25% of all revenue even though it was used by only 7% of sites, so it received the highest effectiveness rating. If you offer visitors only one type of product recommendations this may be the way to go.
Selecting the right mix of recommendation options, displaying them in the right place and at the right time is key. It depends on the number of categories and the kinds of products and, most importantly the needs and desires of your customers.
The ultimate goal of adding personalized recommendations to your store is to enhance the customer experience. Measuring the actual performance on your site will reveal how well you’ve succeeded.
Join The Conversation
More On The Blog
Are Spree sites meeting Google’s page speed standards?
A study of 680 sites built on the Spree OS platform reveals nearly all are mobile friendly, but only a handful achieved a mobile speed score high enough to be considered by Google as performing well.
Barry Harrison — Sep 12, 2016
Machine Learning and eCommerce Talk at SolidusConf
I recently spoke at SolidusConf about Machine Learning and eCommerce
David Jones — Jun 14, 2016
Spree vs Solidus: Which is Right For You?
Spree vs Solidus? We'll help you decide which eCommerce platform is best for your online store.
David Jones — Apr 26, 2016