Refining Data Quality to better detect fraud
In short
For the Government Water Plan, a large company in the environmental sector sought to optimize its customer database and to detect non-compliant uses of water. 10h11 provided the appropriate solution to this problem.
01.
LAUNCH
2022
02.
Customer
Water & Environment Company03.
DURATION
6 weeks
04.
SECTOR OF ACTIVITY
Water & energyThe project in detail
This company in the environmental sector called on our services to strengthen the quality of its customer data and identify non-contractual water consumption. Thanks to the integration of several external databases with this company's existing data, we were able to detect inconsistencies and significantly enrich their customer base.
The need
As part of the Water Plan, the objective of this major undertaking was twofold: improve the quality of its customer database and identify possible non-contractual uses of water. To do this, it was essential to gather and centralize information from various external sources, and to compare it carefully with the company's customer data.
The solution
We extracted data from various external databases (BAN, BDNB, INSEE, etc.) and compiled them in a interactive mapping equipped with filters. By juxtaposing this information with business data, a process of automatic matching allowed us to highlight similarities and differences. This led to the isolation of key information to target non-compliant uses of water.
The results
Our approach allowed to enrich the company's customer database by more than 20%. It also highlighted cases of non-contractual water consumption, leading to regulations for some users. As a result, this major environmental company was able to significantly improve the quality of its data and refine its fraud detection methods.
Contact
A project?
Contact us.
Use the form below to send us an email.
Similar customer cases
Des données accessibles pour un transport plus performant et connecté.
Optimizing operational efficiency through the power of digital transformation
Breaking inequalities with data-driven mapping