Marketers, Pick Personalized Recommendations Over Content-Based Recommendations Anyday
While a lot has been written about the importance of personalization, a recent report by Retention Science suggests that only about 50% of online retailers in the US offer personalized product recommendations. Source –MarketingProfs
Though the importance of product recommendations is reflected in the fact that 76.4% of sites in the IR Top 500 offer recommendations, there is a clear disparity between the number of retailers offering content-based recmommendations and the ones offering *truly personalized* recommendations.
So whats the difference and why does it matter?
Content-based recommendations traditionally employ collaborative filtering to show products recommendations based on similar product attributes. It maps products to visitor segments and shows similar products seen by a particular segment of visitors.
Personalized recommendations show product suggestions based on a single visitor’s preferences. A personalized recommendation engine will identify visitor interactions like search keywords, products viewed, social signals and more, to identify preferences and show 100% personalized recommendations.
Here’s a quick look at the key differences between content-based & personalized recommendations:
| Content Recommendations | Personalized Recommendations |
|---|---|
| Content recs are static in nature - All visitors are shown the same set of products. A content based engine essentially follows a ‘one-size-fits-all’ concept. | Personalized recs are dynamic - Visitors are shown product suggestions tailored to their preferences in real-time. |
| Content recommendations tend to the show the same products repeatedly. They don’t help merchandizers solve the problem of long tail product discovery. | Personalized recs show products based on the likes and dislikes of the visitors. They are more adept at showing a larger percentage of the catalog and improve product discovery significantly. |
| Content based recommendations depend on a large catalog size to show accurate recommendations. Therefore, in case of a retailers with a small catalog size, the recommendations may be irrelevant. | Personalized recs on the other hand shows personalized recommendations even for relatively smaller catalogs. |
Conclusion: With Personalized recommendations merchandizers can show product suggestions in real-time that will engage visitors on the site and are known to account for almost 30% of all conversions.
A real world case study: Impact of personalized recommendations on conversions
Unbxd helped Puravidabracelets.com personalize their homepage, product pages and cart page with its personalized recommended widgets.
With the personalized ‘Hand Picked Just For You’ widget on the homepage and product pages, Puravidabracelets.com sees a 9% conversion rate – This means out of 100 visitors who view these recommendations, 9 visitors go ahead and buy a product.
Offering online shoppers personalized product recommendations help ecommerce sites boost their average order values and conversions.
Is your product recommendations helping you enhance your site’s bottom line? Do let me know in the comments below.