Product bundling & Promotions

Service Background

It is necessary to innovate the value proposal aimed at the point of sale, a personalized and tailor-made proposal, which will allow us to increase the penetration of our portfolio.

Based on the current situation, there is an opportunity for growth in coverage on different customer segments

Service OBJECTIVE

• Developed a system capable of showing the strongest relationships between one product and another, that is, we will be able to see which products customers prefer to buy together.

• The product bundling should be developed based on the results of the market basket analysis.

• The results of the market basket analysis can be taken into account for:

  • Design of catalogs and brochures

  • Structure of the product portfolio for each segment

  • Targeting of the product portfolio

More accurate results can be obtained if we take into account other dimensions in the analysis of market basket, such as: channel, segment, geographic location, etc.

Service SCOPE

As first steps, we will thus leverage the data gathered that includes profiling based on:

1. Offer several products as an integrated product (combo)

2. Apply the association rules technique to design product bundles.

Data mining technique that is used to discover relationships between the data collected in a certain set. This technique is frequently used in the retail sector.

Service APPROACH

The services involves 9 steps:

1. Data cleaning & Data filtering

2. Apply Association Rules

3. Directed Data Mining

4. Delivery of Results to Businesses

Step 1 – Data cleaning & Data filtering

• Keep only variables with complete data

• Filter out extraneous variables (or variables that have ranges outside the desired parameters)

• Replace N/A with imputed values, where possible

Step 2 – APPLY ASSOCIATION RULES

1. Read Transaction/Basket data and Product data

2. Using "A priori" algorithm, build association rule set

- min. set size = 1

- min rule confidence = 10%

- min support is controlled by Double Input Quickform node in %

3. Translate Antecedent collections into product name concatenations

4. Translate Consequent Item ID into Consequent Product Name

5. Calculate price stats and rule revenue

6. Write association rule set to file

Step 3 – Delivery of Results to Businesses