Duplicate-free
Data
Clean a dataset with rules you can verify.
A practical need.
A clear response.
Focused data preparation to standardise formats and flag likely duplicates. Ambiguous cases are set aside for review, without silent automatic merging.
How it works.
Explore each step to see what is included.
Step 01 / 04
The source file
An authorised CSV file and an anonymised sample establish the starting point.
Step 02 / 04
Normalisation
Formats are standardised using the agreed cleaning rules.
Step 03 / 04
The duplicates
Likely matches are flagged. Ambiguous cases are separated for your review.
Step 04 / 04
The clean output
The cleaned file comes with the reusable script and transformation log.
What you receive
- A dataset review and approved cleaning rules.
- A reusable normalisation and duplicate-detection script.
- A cleaned file with a list of cases requiring review.
- A transformation log and operating instructions.
The starting scope
One CSV file of up to 50,000 rows, up to ten columns and five cleaning or matching rules.
Who is it for?
For preparing a contact list, inventory or reference list before import.
Outside this scope
External data enrichment, business decisions on ambiguous matches, sensitive data and continuous cleanup of a connected system.
Before we start
What should I prepare?
An anonymised sample, an authorised full dataset and the required business rules.
What does the starting price include?
One CSV file of up to 50,000 rows, up to ten columns and five cleaning or matching rules. One feedback cycle on the agreed deliverables is included.
What if I need a wider scope?
We define any changes in a separate proposal before work begins. No paid extension is added without your agreement.
Do you work remotely?
Yes. Discussions and reviews can take place remotely in French or English, for clients in France, the European Union and internationally.
Other angles
for your project.
A clear need.
Let’s talk.
A few lines are enough to begin. We will define the right scope together.