Skip to content
Home » How To Personalise Customer Experience

How To Personalise Customer Experience

We all know how good it feels when a company KNOWS us. Not the “How did you my get phone number!?” knows, but more the feeling when we’re in a supermarket and they’ve placed exactly the brand of Milk and Eggs we like right next to each other. It’s as if their entire supermarket was setup just for us. What happens when we get this feeling in a supermarket, in a restaurant, in a clothing store? We become loyal to the business, right?

That, my friends, is the power of personalized customer experience.

Now imagine if the smaller businesses had access to the same resources; they would get a fairer chance at competing against the big players. Personalized Customer Experience is one of the many resources major retailers have under their belt that allows them to maintain their dominance over smaller businesses. The worst part about all of this – small businesses don’t even know these technologies exist.

In this blogpost, we will cover what Personalized Customer Experience is and how it has become a must-have in order to compete in today’s cutthroat retail market. (Warning: Expect multiple shameless plugs ahead)

According to the famous ChatGPT, Personalization Intelligence refers to “the ability of a system or technology to analyze a user’s behavior and preferences and then use that information to deliver customized experiences, messages, and recommendations to the user”.

Personalization intelligence has been proven to be valuable for businesses looking to improve their customer engagement, conversion rates, revenue, customer retention, and customer experience. Using various techniques under the personalization umbrella, businesses have successfully built a seamless buying experience for their customers; we cover two of those techniques today.

That’s what this post is about. Before you read on, check out our interactive personalisation intelligence sandbox and see if you can figure out the benefits of Personalisation Intelligence on your own.

The Milk-Eggs example is a classic use case of a data mining (shall we say personalisation intelligence) technique called Market Basket Analysis. It’s used to identify items that people buy frequently together. Most retailers understand and apply the concept of selling items that go well together as cross-selling. Market Basket Analysis is a super-charged version if it. While most businesses use Market Basket Analysis to identify bundle-able items (Combo Menus, Gift Baskets etc.) for cross-selling, there are some really creative use cases as well.

For example, one retailer, knowing that Shampoo and Conditioner are frequently bought together, decided to put the Shampoo on the first shelf right as customers walked into the store and the Conditioner on the shelf right before the exit. The result? Customers browsed through the entire store before picking up the two items, and picked up a few other items on their way.

We, at Xabit, have also used Market Basket Analysis to help our clients identify cross-selling and up-selling opportunities, optimize product placement, and improve inventory management. Eventually, it helped us help them gain insights into customer behavior and preferences, and led to higher sales.

When we buy an item on any e-commerce site these days, there’s always a section that says “Recommended For You”, which shows products that we’ve not yet bought. Or, if you’re a music lover, you know Spotify’s “Discover Weekly”. Both of the above are examples of recommendation systems. Their use is to help customers discover products that they’ve not yet bought.

Recommender systems are extremely useful for companies that have the long-tail problem. Now you might be wondering what in the world a long-tail problem is. Basically, say you’re a retailer that has a 1000 different products on offer. You do some analysis and you find that a majority of your sales come from only 200 of the items. You dig some more and you find that you can’t get rid of the other 800 products because there are a small niche of customers that buy those 800 items but not the 200 that you sell the most. You have what is called a long-tail problem and are left with two choices:

  • Remove the 800 items from your product list, foregoing the x% of revenues they bring; which is a good decision if the x% is small and don’t bring in a lot of margin as well but bad business if they do.
  • Find a way to get some of the customers that only buy the 200 items to also buy a few of the items in the 800 list and vice-versa. This is where Recommendation Systems come in. In a very basic sense, they help you identify products customers have not yet bought (for many reasons) but are likely to love and “recommend” these products to your customers.

The beauty of Market Basket Analysis and Recommender Systems is that they can easily scale to millions of products and customers. Take Spotify for example, it has 100 million tracks and 500 million users; yet a simple recommendation engine is one of the most loved feature in any app out there.

Even with Xabit’s clients, the recommendation system has turned out to be a hit. We have integrated it in our clients’ e-commerce platforms. Our proprietary algorithm goes through individual customer’s buying history and recommends products specific to their liking. And guess what? The customers are actually trying out our recommendations, thus generating higher sales for our clients.

That’s the power of personalised customer experience.

Keep Data. Decisions. Repeat-ing,
Bikranta