Get Started

Personalize Customers’ Shopping Experience With Market Basket Analysis

Use Cases & Projects, Dataiku Product, Scaling AI Benoit Rojare

We all have our shopping habits, whether it’s buying this brand of tomato sauce that we particularly appreciate or this shape of pasta that the kids ask for all the time. If both products were offered in the same pack, we would most likely buy it. Turns out, we might not be the only ones in that scenario, and many other people might be doing the same combinations.

It’s been years since retailers understood this — a very visible consequence of it is the organization of supermarkets: sauces in the same aisle as pastas, brooms next to dustpans, etc. With the shift of shops moving to online, the way of shopping changed, but some habits remained and new ones emerged. With data from online shoppers, retailers end up with many new ways to sell better. For instance, I might be looking to buy a power strip online. After finding it and putting it in my basket, the website offers me to buy other items like an extension cable. Bingo, I almost forgot I needed one too. Anyone who has shopped online has probably already experienced it: Personalized offers became the norm. 

Unsurprisingly, retailers really understand the power behind personalization: 80% of companies report seeing an uplift since implementing personalization, which includes recommending relevant products to users, states an article from Capgemini. In fact, personalization is not just something that retailers appreciate to improve revenue, it is also appreciated by consumers themselves: According to Accenture, 91% of consumers are more likely to shop with brands who recognize, remember, and provide relevant offers and recommendations. This resonates well with the physical world where we naturally tend to prefer shops where we can get individual attention. Buying shoes in a supermarket is usually less pleasant than in a local shoe shop where we can get the advice and the help of a sales person. 

online shopping

Studying the Online Shopping Basket

We can find market basket analysis on top of the list of techniques an online shop can do to personalize its shopping experience, as it can be easily implemented. It aims to find significant correlations with products that are often shopped together by online consumers. This analysis is based on large datasets that compile the history of purchases. By doing so, brands are able to recommend the best product associations to the consumer. 

Once identified, these associations of products can either be used along the shopper journey on the website with recommendations for extra items or in a physical shop by optimizing product placement or proposing bundles of products, for instance.

This topic is critical for all retailers, so that both the shopping experience and the revenues can be improved. This is why Dataiku created an Industry Solution dedicated to the topic. 

Let’s have a look at what you will find in this solution, but first here is a reminder of what an Industry Solution actually is.

How Can Dataiku's Industry Solutions Help You Reach Full Potential?

Industry Solutions are Dataiku add-ons accelerating the journey to realize advanced or foundational industry-specific use cases within your organization. They are an operational shortcut to achieve real-world business value. Taking advantage of Dataiku’s core features, they are built to be fully customizable and entirely editable.

They come with:

  • A user-friendly interface that enables fine tuning to match specific business requirements
  • Ready-to-use dashboards that can be customized
  • Documentation and training materials

Dataiku industry specialists develop solutions for every vertical, among which:

As a result, business professionals experience a boost in AI productivity and can rationalize their resources.

How Does It Work in Practice?

The Market Basket Analysis solution provides a turnkey project to accelerate the development of analytics tailored to your data and business structure. It includes a Dataiku application that eases the data upload and adjustment of the project to your own needs.

 

Learn more about Dataiku's Market Basket Analysis solution in the video above.

With this solution, merchandising analysts, digital media managers, or marketing departments can access:

  • Business rules: For each customer, find the best business rules to trigger repeat purchase and cross-sell scenarios.
  • Item frequency analysis: Analyze the proportion of each given product (also known as support or frequency) and display the most popular ones. 
  • Rules browser dashboard: Analyze the association rules learned across four main metric filters (frequency, confidence, lift, conviction).

From a user perspective, the solution is made of the following components:

1. Dataiku Application: 

To easily upload your own data and adjust the project to your needs.

market basket analysis

2. Structured Flow:

To understand the project organization, navigate data transformation steps, and tailor anything to your needs.

structured flow Dataiku

3. Prebuilt Dashboards:

To get a quick and shareable view of transaction data and understand your items’ support before and after selection, with prebuilt graphs.

prebuilt dashboards

4. Interactive Dashboard for Analysis:

To explore your sales history, uncover business patterns, and identify the most frequent items purchased.

item frequency analysis

5. Interactive Dashboard for Recommendation:

To extract new business opportunities thanks to the rules browser. Search items to browse rules linked to them and define your metric thresholds and orders to filter the results.

dashboard for recommendation

6. Automated Pipeline:

If needed, automate the refresh of the flow and configure its cadence according to your needs. Feed the output data into your marketing suite to launch enhanced marketing initiatives.

automation in Dataiku

Start implementing your Market Basket Analysis system in minutes, with these simple requirements:

  • Datasets: history of clients purchases
  • Dataiku version: 9.0 or later

You May Also Like

Mitigating Data Bias by Implementing Responsible AI Practices

Read More

Graph Neural Networks: Graph Classification (Part III)

Read More

Smart Manufacturing: How We’ll Get There and How We Won’t

Read More

Data Transformation: How to Transform Data More Efficiently

Read More