Cross selling proposition
Summary of the entity:
Cofares is the leading pharmaceutical wholesaler in Spain. It is a cooperative with 100 % pharmaceutical capital that provides pharmacies with products as well as health related services.
Summary of the challenge:
Cofares wants to create a consumer-facing cross-selling proposition of products to pharmacies (OFs) customers, based on the analysis of the consumers’ purchase ticket.
The objectives of this initiative are:
- Increase the sell out of the OF;
- Influence the consumer’s decision to buy brands, products, etc;
- Increase the number of different references that the customer buys to Cofares;
- Influence the OF’s decision to purchase brands and products;
- Improve the experience of the OF in the ordering process.
All recommended products must be active in Cofares and in stock.
All recommended products must be directed to active OFs.
The solution must not affect the operational performance.
- Clustering of anonymous OFs reflecting similarities between them, based on the universes, categories and products they sell to their customers and the province, habitat and decile (sales interval) of the OF. Identify the clusters of greatest interest.
- Soft assignment of each client OF to the clusters.
- Generation by client OF of a list of recommended products, from among the products sold by the anonymous OFs of its nearest cluster, and which the client OF does not buy from Cofares. Systematic recording of results in BigQuery: OF-clusters and OF-recommended products.
- Results tracking: Systematically measure the volume and % of recommended references purchased, for each client OF.
- Implementation of the consumption modes (APIs or others as appropriate) that allow the algorithm to be used in the corresponding applications (OFs and Farmanager Portal).
- Measure match of recommended products in case 1 vs. case 2, for each client OF
From the pharmacies purchase tickets, the health status of a client can be inferred. Therefore, this information is sensitive data according to the GDPR. Furthermore, if the substitution of the client id in the provided dataset is reversible, the traceability of usage of the client data should be ensured. This latter feature can be achieved using the ProRegister Tool provided in the REACH ToolBox.
Recommendations to the client OF should at least be based on the products commonly sold by the anonymous OFs in their nearest cluster and not purchased by the client.
The degree of overlap of products purchased from Cofares by OF with the products sold by anonymous OFs.