Prepare your Data to Ensure its Quality and Impact
DATA
Lorena Aguiar Franjoux
1/6/2026


Why preparation is a key step
Data preparation involves cleaning, restructuring, and formatting data to ensure its quality. It is an essential step before any analysis or reporting. It allows you to:
Make data reliable and consistent
Merge multiple heterogeneous sources
Provide a comprehensive and usable view of the business
Without this step, dashboards are unreliable, decisions are risky, and confidence in the data can be called into question.
Barriers and costs
Preparation is often seen as a chore. And yet, not doing it can be costly:
Poor data quality leads to errors, wasted time, and bad decisions
Teams sometimes spend more than 70% of their time searching for, correcting, or cross-checking data
Organizations often don't have the skills needed to clean or transform data
It is therefore essential to give greater recognition to this stage and to incorporate it into the schedules and budgets of data projects.
Involve the business owners, not just the technical staff
Data preparation should not be reserved for IT experts. Business teams understand the meaning of data:
They know if a figure “doesn't make sense” (e.g., 80 hours of volunteer work in one day?).
They can spot duplicates or outliers.
They know what to highlight for their needs or those of funders.
Collaboration between those who manage data and those who use it on a daily basis is the best way to avoid misunderstandings.
A matter of governance
Preparation is also based on collective decisions:
What format should be used for dates (day/month/year)?
How should place names be spelled? What about legal status?
What data should be kept in the event of a duplicate entry?
For example, if you combine data from two different tools, you need to decide with the teams involved which version is the right one.
This consistency makes it possible to create a single database that is reliable and understandable to all, to limit gray areas, and to reduce the development of uncontrolled parallel tools (shadow IT).
Break down silos to increase efficiency
Involving end users in data preparation has several advantages:
Tools are designed faster and better suited to their intended uses
Reduction in back-and-forth communication between business and technical teams
Limited use of isolated files created in a hurry, which are non-compliant and insecure
Collaboration is a key factor in shortening lead times, ensuring reliable results, and reinforcing teams' autonomy in their daily tasks.
In summary
Clean up data
Eliminate duplicates, correct errors
Restructure and harmonize
Ensure interoperability between different sources
Involve business units
Ensure data is understood and relevant
Establish common rules
Align teams on format and processing choices
Document, maintain traceability, facilitate reuse
Data preparation is a demanding but essential step. It determines the quality of everything that follows: visualization, analysis, decision-making. It should not be underestimated or left to a single person.

