Data mining – how to analyze customers’ market baskets to increase sales

Data mining is a set of techniques for automated discovery of statistical dependencies, patterns, similarities or trends in very large databases. Of course, for those involved in sales techniques and correcting sales results, data mining is an extremely valuable issue. Due to ease way of obtaining large online data, data mining (here web mining) has become an interesting issue for e-commerce, especially for exploring purchasing trends by analyzing customers’ market baskets. This involves finding a relationship between bought products by imaging and discovering links using simple rules, known as the rules of association. Market basket analysis can effectively present the offer of products, create more effective promotions and develop more efficient marketing campaigns – in brief helps increase profits and improve competitiveness.

data mining

Construction of association rules

While analyzing a market basket we are looking for associations. This is done by combining elements together and finding the relationship between them, at the same time discovering the shopping habits of customers. This is done with special algorithms that will be presented in the next post. The result of the analysis are rules of association in the form : A → B, ie. IF [body] then [head]. The body and head can be freely extended and consist of a conjunction of many conditions and create a more complicated rule.

But let’s start with a simple example: [frozen pizza ] → [ketchup]. This rule assumes that if a customer bought a frozen pizza there is a substantial likelihood that it will buy a ketchup. Of course, it is not necessarily true one hundred percent of the time. The fact that these two products are strongly linked decide specific indicators which define quality rules.

Measures of quality rules

The quality of the rules, that is how much they are true and valuable from the point of the business view, outlines several measures: support, confidence and lift.

  • Support is the percentage of transactions containing the rule in all transactions in a retail outlet. In our example support determines the probability of buying both frozen pizza and ketchup by a randomly selected customer in a particular store.
  • Confidence is the probability that the B product will appear when the A product also will be present. In our case it is likely that the customer will buy ketchup, once he bought frozen pizza.
  • Lift informs what is the impact of the sale of the product A on sale of the product B. If the lift of value is 1, then we say that the products do not affect each other, if it is less than 1, we deal with products antagonistic, and if is greater than 1 both products are complementary.

Interesting is the rule, where both the support and confidence are greater than some minimal values fixed by an expert from the field  – we say then, that the rule of association is strong.

Let’s take another example in which we enumerate the support and confidence. Suppose that the sales report in the supermarket shows that last Thursday at the 1,000 customers, 200 bought loaves of bread, and from those who bought loaves of bread, 50 bought a butter. Here we see a rule that if someone buys bread, also buys butter, which support is 5% (50/1000), and the confidence is 25% (50/200).

Market basket analysis in e-shop

Initially, the analysis of market basked included transactions in supermarkets. Now, in an increasingly growing importance of e-commerce, analysis of market baskets in e-stores seems even more relevant, because Internet is the perfect place to get to know the users habits. Transactional data collected in an automated way from online stores, provide reliable information about the customer and about product he buys. Additionally utility developed on the basis of the rules and the effectiveness of taken actions can be easily estimate, eg. by calculating the ROI.

Most common use of associacion analysis is finding products which are complement to each other and are therefore the most commonly consumed together. It positively affects sales – knowing that the customer buys product A, you can show him the products B, C and D, which are usually purchased together with A. The online bookstore Amazon can be used as the example here, where while we search a particular title, we get together a list of books with similar themes. These include items that other customers bought together with the title (or viewed products before purchase) and those that are recommended by other customers. But for the purpose of market basket analysis in e-commerce you can use also other types of association rules:

  • Negative association rules discover links between the products present in the basket and those which are not. On this basis, you can find significant customers’ buying behavior like that when someone buys a Coca-Cola does not buy Pepsi, or if someone buys a juice he will not buy bottled water. Negative rules are equally useful in the preparation of online store offer as positive rules (presenting products sold together), because thanks to it you can dynamically modify the appearance of the page, so that the customers do not see the content they do not want to buy.
  • Cyclic association rules take into account the repeatability/cyclical nature of purchases over time and thereby customize the appearance of the website to the customer. He gets the advertising message while his increased activity when – according to the results of analysis – usually buy specific products or services.
  • Inter-transactional association rules do not look for dependencies in a single transaction but examine the relationship between purchases made in a certain period of time. These rules take into account the context of shopping (eg. time, place, customers), which is ignored in the intra-transactional rules. They aim to find sequences of events.
  • Ratio association rules – besides information about purchased products they also contain information about the amount of money spent for various goods and services. As a result, it is possible to predict the value of the transaction and the purchase of specific items from the e-store’s offer.

The most important are the relevant data

Often jokingly data mining is characterized not as “the search for rules in the database”, but as “mining for dollars”, because given rules help increase profits – surely this is in the case of market basket analysis. It allows you to optimize marketing campaigns by identifying the probability of purchase of individual products, develop effective methods for cross-selling (selling a product or service to the customer associated with the purchase of another) and up-selling (selling more expensive products versions). It is necessary to know customers’ buying habits if you want to construct an offer consistent with their expectations. The key is having the right data – in their acquisition and analysis helps Synerise platform.

With Synerise you can collect data from various points of contact and gather them in one place. Important in market basket analysis transactional data you will gain from both: online and local stores, thanks to integration with the POS. After finding high quality rules you can easily reach with a suitable offer each customer with dynamic content available on the site (in the form of banners and pop-ups) and e-mail. You can also profile the message based on customer segments. And if, despite a personalized offer, your customers will add products to the market basket, but do not decide to buy them, you can send them an e-mail with abandoned basket and a discount to encourage them to complete the transaction – it will take place in an automated way.