Practical Ecommerce

How to Personalize Shopping on an Ecommerce Site

Personalizing an ecommerce site refers to the process of altering its appearance or navigation to the interests of visitors. Personalization can greatly increase conversion rates and improve loyalty. Many large ecommerce sites use personalization. Multiple vendors offer personalization platforms for merchants of all sizes.

I recently attended the annual National Retail Federation conference in New York City. Personalization was a popular topic among attendees, speakers, and exhibitors. All said that personalizing the customer experience would likely improve conversion rates. An ecommerce merchant and a visitor to the merchant’s site both benefit from personalization if it is implemented correctly.

In this article, I’ll review what three large ecommerce companies do with personalization. I’ll then offer a phased approach for all merchants to implement it.

1. Nordstrom

Nordstrom is known for exceptional customer service. It offers a personalized experience to its shoppers on According to published reports, it is also working with vendors to identify the shopper in the physical store and offer relevant suggestions and promotions. For example, Nordstrom recently tested a mobile app that allows shoppers to be identified when they walk into a store. The shopper’s profile is shared with the in-store sales team, enabling them to personalize the shopper’s visit.

Regarding its ecommerce site, Nordstrom has created a separate employee group to focus on cutting edge personalization technology to further improve the shopping experience of their visitors. Nordstrom executives reportedly believe a more relevant experience will ultimately be worth more than the miles, points, or coupons offered by a traditional loyalty program.

2. Netflix

If I had to list leaders in ecommerce personalization space, Netflix would surely be on that list. It first started using personalized recommendations based on a shopper’s viewing history and ratings. It then introduced profiles to allow an individual to keep her viewing history and recommendations separate from others in the family. Netflix then followed profiles by a utility called “Max” that recommends movies based on the genre a customer prefers to watch in conjunction with viewing history and ratings.

Netflix home page.

Netflix, the movie site, is a pioneer in personalizing product offers to its customers.

Max has taken personalization to a new level by offering an option to pick movies automatically for a viewer without the need for any additional input.

Netflix does this by using predictive models that are further refined depending on if the user does not end up watching the movie or gives it a low rating. Ecommerce retailers can likely learn a few things from Netflix by studying how it combines product ratings, browsing history, order history, and customer persona to personalize content for each subscriber. As a Netflix customer, I look forward to its next advancement in personalization.

3. Staples

Staples apparently got serious about personalization last October when it purchased Runa, a personalization company. Runa uses algorithms to attract shoppers and keep them engaged until they complete the checkout process. Staples reportedly plans to use this capability to display real-time personalized offers to shoppers while they are on the site. Shoppers can also receive free shipping offers based on their order histories, products in their current order, and other criteria. Staples hopes to reduce the cart abandonment rate by using the personalization features offered by Runa.

Integrating Runa into Staples is an example of how a third-party personalization platform can integrate with an existing commerce environment to make the entire offering more appealing to shoppers.

Personalization for Small Retailers

Smaller retailers can implement personalization, too, by following these guidelines.

Hosted ecommerce platforms can offer personalization features. Otherwise, merchants can evaluate third-party solutions like RichRelevance, Retention Science, and 4-Tell. Regardless of the platform, it is important launch personalization in phases to help understand its impact on different areas of the business and also to give an opportunity to optimize one area before applying it to the entire business. In “Using Omni-channel Personalization for Ecommerce, my previous article, I explained that personalization is based on multiple rules and it typically takes a while to get these rules right. So, consider dividing personalization implementation into five phases, as follows.

  • Phase 1: Focus first on navigational personalization. This is the most basic form of personalization. Shoppers can be shown different content on a page based on their activity — i.e., clicks. This also includes redirecting the shopper to a landing page based on the referral site.
  • Phase 2: Build customer-specific personalization. Most sites use customer preferences and order history to personalize the experience. It’s common to group similar customers in a segment to enable personalization without impacting site performance or overloading the back-end systems.
  • Phase 3: Personalize using third-party data. Shopper location, social network feeds, and other third-party data, like weather, can further improve personalization. With this kind of personalization, it’s necessary to correlate data from internal and external sources. This is best handled by deploying a personalization engine.
  • Phase 4: Enable predictive personalization. Once a business has basic personalization in place and has a good understanding of its customers, predictive personalization can be enabled. This personalizes the site to meet the customer’s needs even before a customer has performed any activity on the site. For example, Amazon recently announced that it could ship orders before a customer places them, taking predictive personalization to the next level.
  • Phase 5: Support segment of one. Tracking individual customers and personalizing their experiences is called catering to a segment of one. This is the holy grail of personalization where a business is able to analyze all customer attributes along with third party data in real-time to increase relevance and improve conversion. Each customer sees his own version of the site to make his experience more effective and pleasant.

These phases will work for most retailers that are starting out on personalization. But these phases are not set in stone. They may require reordering depending on the nature of the business or customer base.

Gagan Mehra

Gagan Mehra

Bio   •   RSS Feed


Sign up for our email newsletter

  1. Wendi Makuch January 30, 2014 Reply

    This is a terrific article because most personalization strategies focus on the largest retailers with the deepest pockets. Thank you for giving such great suggestions for how the SMB retailers can phase in personalization too!

  2. Gagan Mehra January 30, 2014 Reply

    Thanks Wendi

  3. Radu Vrabie February 4, 2014 Reply

    Of course.
    About third party data, with our tool, Marketizator you can also use weather, customer journey, name, previous bought articles, the ammount of spending and many other spending. Based on our experience tih 200+ eshops, what matters the most is the strategy behind any personalization and always test it.
    Tx, Gagan for sharing ;)

  4. Jerry Jao February 7, 2014 Reply

    Great tips and appreciate the examples – esp. the Netflix one.

  5. Gagan Mehra February 7, 2014 Reply

    Thanks Jerry

  6. Trevor Mitchels February 14, 2014 Reply

    Okay, you have no idea how grateful am I. These are killer tips from the big players for a small e-trader like me just migrating from one service to another ( ->

    I’ve ticked off 1, 2 and 3, but have to work on 4 and 5 it seems :)

    • Andy Hunt April 14, 2014 Reply

      @Trevor and Wendi —

      Personalization is absolutely the future of eCom and all websites. Dan Siroker from Optimizely talks about this all the time and I wouldn’t be surprised if they launched an SMB focused product.

      That said, you have to be really careful on whether the ROI is there and if it’s worth your time. Simple A/B testing is usually enough to get a lift for eComms under $500K in revenue, and even then, investing the time to do a lot of personalization may not make sense. When our team is working with clients we usually find that there’s still opportunity for A/B tests before you personalize to get the real lift.

      Anyway, I hope this helps!

  7. Gagan Mehra February 14, 2014 Reply

    Good to hear that this article was helpful, Trevor.

    All the best for #4 and #5 :-)