Product repositories: how to reconcile them?

The need

Obstacles to product repositories

Our customer, a major player in the consumer credit market, wanted to offer a one-click online financing plan to all used car buyers.

While most partner sites selling used vehicles use the Argus (and sometimes JATO) as their automotive repository, the algorithms used to create our customer's financing plan were based on a different repository: EUROTAX.

In order for the private individual to receive his financing plan in a matter of seconds, it was necessary to establish a unique correspondence between the entries in the repositories, which did not have a common key, and whose differences in vehicle description made this correspondence non-trivial.

Benefits

Thanks to the involvement of business experts (who have in-depth knowledge of automotive repositories), the fields used in the composite key and the arbitration fields were optimally determined.

The rate of uniquematches has risen :

• From 55% in the first approach, which consisted in asking the customer's Data Scientists to code string matching algorithms in Python, algorithms that had been regularly rejected by the business for several months.
  
  
• The composite key approach proposed by Tale of Data involved the business to the tune of 95%.

• With the remaining 5% of multiple matches showing no significant difference in the financing plan generated, the Tale of Data approach was validated within a week by the customer's business teams.