In 2014, shoppers spent 101.9 billion online during the holiday season, and this year, that number is expected to grow. Avoiding the crowds and the ability to shop at your convenience draws many to do their shopping online, and retailers need to make sure their eCommerce experience is ready for the influx of shoppers. Creating product recommendations that are timely and relevant not only makes the shopping experience easier for your customers, it also helps improve average order value and overall revenue.
Why recommending the ‘most popular’ items won’t cut it.
It’s no longer enough to recommend the most popular items to your customers, especially during the holiday season when many people are buying items for others. While recommending the most popular item has its place, with the vast amount of customer data available to today’s eCommerce marketers, product recommendations can go far beyond the most popular item to help you make the most of this holiday season:
1) Recommendation and Temporal Diversity
Your customers are individuals who should get individual recommendations. By recommending the same products to all customers, you’ll be sending irrelevant recommendations that will either be ignored, or worse, cause customers to become disengaged with your brand. It’s also important to remember to diversity the recommendations you’re sending to each customer; that is, customers should not be receiving the same recommendations week after week.
2) Make recommendations based on recent behavior
When a frequent shopper comes to your site to purchase holiday gifts, make sure you’re making recommendations to them based on their recent behaviors. Even though you may have information on a user that spans multiple months (or years), using their most recent behavior will make your recommendations more relevant – especially when a user has been browsing items that they wouldn’t usually look at.
3) Exploratory recommendations
The holiday season is a great time get your customers to explore items that they don’t normally see. Making random recommendations helps customers explore your entire inventory. However, be very careful with these as being random with all of your recommendations can backfire and lead customers to believe you aren’t relevant. Keep them at a minimum and be sure to do A/B testing to see if they’re working for your brand.
4) Knowledge-based recommendations
Using what you know about a customer is vital to relevant recommendations. For example, when someone is looking at a shoe on your website, you don’t want to recommend the same shoe in different sizes. Using knowledge-based rules regarding a customer’s behavior, demographic, gender, etc. helps make recommendations more relevant and engaging.
5) Recommendation privacy
Companies should tread lightly when it comes to recommendations, especially around the holiday season. This is notably true for eCommerce sites where multiple people in one family may share the same account. Amazon Prime, for example, is usually shared within one household. To help protect each user’s privacy, Amazon allows you to turn off recent browsing history / recommendations. While it may seem counterintuitive to allow customers to turn off recommendations, doing so gives shoppers a sense of security when it comes to keeping their holiday gifts a secret when shopping on your website.
With the mass amount of customer data available to digital marketers, as well as the various third-party platforms that allow companies of any size to apply data science to their marketing, creating relevant and timely product recommendations this holiday season is a must. Relevant product recommendations will help not only improve your customer’ experience, but also increase your average order value and overall revenue.