Classifying products based on the analysis of cash receipts

For profit margin and price-image management 

This article presents an overview of an analysis carried out by the data scientists from Mercio. Our experts rely on the latest technologies and develop their own tools to help retailers to answer complex and interdisciplinary issues.

Through the analysis detailed here, we wanted to offer some credible and concrete ways to solve one of the recurring problems encountered by our customers: How to create a classification system for their products that is not based only on obvious characteristics but rather takes into account customers’ expectations, all while managing tens or even hundreds of thousands of items? How can our clients use these results to improve their pricing and merchandising strategy? How can they price new products for which there is no sales history or comparison with competitors? Mercio teams found an innovative approach to answer these questions.


The first layer of the algorithm classifies the products and assigns them a score based on whether they are purchased:

  • Always with at least one other product (score of 0)
  • Always alone (score of 1)
  • In between (from 0 to 1)

The second layer goes one step further: it ranks products based on their tendencies to be most often purchased:

  • With a wide range of other products (lower)
  • With a limited set of products (higher)

The third layer combines these two indicators to define the following three categories:

  • « Driver” products: These products are bought sometimes alone, but more frequently with complementary products. These are the products that bring customers to the store. Their prices are more likely to be compared and not finding the product on the shelves can lead customers to leave the store without any purchase. Examples from drug stores: baby diapers or shower gel.
  • “Complementary” products: These products are rarely bought alone, they are more likely to be purchased with Driver products: either because they are purchased on impulse or because their usage is complementary.For instance, sponges are more likely to be purchased with shower gel.
  • “Independent” products: They are bought alone most of the time. In drug stores this is the case with exclusive beauty or health products for instance.

An almost instant return on investment

During this analysis, most of our data scientists’ time has been spent conceptualizing the relevant indicators and then analyzing the results. The cutting-edge tools they used – and in some cases developed -, made data loading, visualization and computation a lot faster, taking at most a few minutes, even on massive datasets. Thus they were able in this way to draw actionable insights that can be leveraged through several optimization strategies with an almost immediate return on investment:

  • Increase the value of the average basket
    By increasing the margin of the complementary products, for example with a maximum + 20% price cap compared to the competitors, a brand can retrieve several hundreds of thousands of euros of additional margin without damaging their price-image.
  • Improve the price-image to develop the volumes
    By lowering the price of sensitive products and by increasing the price of complementary products, the brand increases its attractiveness without damaging its total margin.
  • Sell additional services
    Having identified how the consumer of «independent products» behaves, our client can focus on how to increase its basket value, services and warranties associated with these products, rather than other products.

As an extension of this analysis, we can envision several ways to push optimization further:

  • Replicate the method at the a store level, for each individual drugstore – we are likely to find various customer behaviors between suburban and downtown stores, for example.
  • Identify and classify «Sensitive product + complementary products» groups within a category and cross-category to improve merchandising.
  • Cross-reference these insights combined with data from the retailer’s e-commerce website to further refine results. Website data would for instance tell which products are more likely to be added in a basket first, which ones are rather seldom added first and finally which ones are often bought alone.

Towards predictive pricing for new products and marketplaces

This experiment shows how data science can bring valuable, immediately actionable insights to optimize retail pricing. Combined with a flexible pricing engine able to translate these insights into price rules, data science allows a predictive rather than reactive approach to pricing. By identifying the relevant product attributes related to a specific buying behavior, and classifying products according to those results, we can set prices for new products or for marketplace products without having to rely on sales history or competitive alignment rules.

Going further with Mercio: 

Mercio’s price optimization software is designed to bridge the gap between marketing insights and pricing strategy. This means that each retailer can express their own business logic in a flexible and scalable solution, while benefiting from a solution developed according to the best pricing practices in retail. Discover the presentation of the Mercio platform in the « Our solutions » section. 

Are you looking for content about pricing? Leave your email so we can send you eBooks!
Alexandre Point
Alexandre leitet unsere Entwicklungsteams und stellt sicher, dass Mercio der innovativste Editor für das Pricing im Großhandelsvertrieb bleibt. Als Absolvent der Ecole Polytechnique und der Stanford University sowie als erfahrener Unternehmer verfügt er sowohl über einen exzellenten Unternehmergeist als auch über eine unstillbare Neugier, die es uns ermöglicht, bei der Innovation für unsere Einzelhandelskunden noch weiterzugehen.