After an intense period of M&A and expansion across 9 countries, the KPIs about our Debt Collection process were experiencing big fluctuations from week to week.
We started reviewing our data sources and realized that the data representing the company receivables was only covering 2 out of the 9 countries. After conducting several interviews with the Finance team we discovered that they were doing a huge amount of manual work gathering, cleaning, and transforming data from the remaining 7 countries, which led to inaccuracies, sudden changes in KPIs, and a lack of efficiency. We also discovered that only reliable receivables data could be used to increase our financing lines, which at that time consisted of the data from 2 out of 9 countries.
Knowing that the 2 countries that were covered by our data sources were performing well, we decided to design a Data Product that would ingest, clean and transform the data from all the countries (+10 ERPs), starting with the ones with a bigger problem in the debt collection process, this way we would reduce manual work and standardise the information reported. We also introduced quality checks to monitor freshness and accuracy to catch big fluctuations, we designed a set of entities that could represent the Finance business and finally created some final visualisations that would show global KPIs for the less technical profiles.
The result was a Data Product for the Finance team that was improved continuously according to new needs and feedback.
Tech Stack
Impact
Manual work was reduced +5h per week, availability and trust of the data was increased which led to an improvement of +€13M in our financing lines and a reduction of 10% in our Debt KPIs.
Jose Gabriel Martínez Martín