FinancialConsolidation.com
HomeGlossary

Data transformation

Data transformation is the process of converting data into the structure, format, or form required for further processing and analysis.

Domluvit ukázku Reporting.czDomluvit ukázku Reporting.cz

Data transformation is the process of adjusting, converting, and structuring data into a form suitable for further processing, analysis, and use in reporting or business intelligence (BI). It includes changes to data formats, structures, values, and relationships between data elements so that the data meets the requirements of the target system or analytical process.

Data transformation is a key part of data integration and of ETL and ELT processes. It makes it possible to harmonize data from different sources, remove inconsistencies, and prepare information for effective use in data warehouses, analytical tools, and reporting systems.

What does data transformation include?

Data transformation can include various operations depending on the nature of the data and the processing requirements:

  • Data type conversion – changing the format or type of values, for example converting a text date into a date data type.
  • Data standardization – harmonizing formats, names, units, or the way values are recorded.
  • Data cleansing – removing errors, unwanted characters, duplicates, or inconsistent values.
  • Data filtering – selecting relevant records based on defined conditions.
  • Data aggregation – grouping and summarizing values, for example calculating total revenue for each month.
  • Calculated and derived values – creating new attributes or metrics based on existing data.
  • Data joining – combining data from different tables or sources using shared identifiers.
  • Restructuring data – changing the data structure to match the requirements of the target system, such as a data model for reporting.

How does data transformation work?

Data transformation usually takes place as part of data processing between extracting data from source systems and using it in the target environment.

  • Acquiring input data – data is loaded from databases, ERP, CRM, files, or other information systems.
  • Data analysis and profiling – the structure, data types, values, and any gaps in the source data are identified.
  • Defining transformation rules – the operations to be performed on the data are specified, such as calculations, filtering, or format harmonization.
  • Running the transformation – data is adjusted according to the defined rules and prepared for further use.
  • Validating the results – the accuracy, completeness, and consistency of the transformed data are checked.
  • Storing or exposing the data – the transformed data is stored in the target system or made available through a data layer.
  • Regular updates – transformation processes run again at the required frequency and update the data when source systems change.

The result is a data structure that meets the requirements of analytical and business processes and makes working with data more efficient.

How to automate data transformation

Data transformation can be automated using ETL and ELT tools, data integration platforms, data warehouses, or analytical systems. Automated transformation processes make it possible to process data regularly according to defined rules without manual intervention.

In ETL, data is transformed before it is stored in the target repository. In the ELT approach, data is first loaded into the target environment and transformed afterward, often directly in the data warehouse.

Automating data transformation delivers in particular:

  • Repeatable data processing – transformation rules are applied consistently every time the process runs.
  • Less manual work – there is no need to adjust data by hand and prepare inputs for reporting.
  • Greater data consistency – standardized transformations harmonize data from different systems.
  • More efficient data preparation – data is automatically structured and prepared for analytical use.
  • Easier reporting updates – regular transformation processes keep current data available.
  • Better quality control – validation rules help catch errors while the data is being processed.

Data virtualization can be an alternative or complementary approach to data preparation. It makes it possible to transform and harmonize data as it is exposed through a virtual data layer, without having to physically store all transformed results in a separate data repository. The specific options depend on the capabilities of the virtualization platform and the performance requirements.

Data transformation and ETL

Data transformation and ETL are closely related, but they are different concepts. Data transformation refers to the process of adjusting and converting data itself, while ETL is a broader process that includes extracting data from sources, transforming it, and then loading it into the target repository.

Transformation is therefore one of the key parts of ETL. It can include cleansing, standardization, calculations, aggregations, or joining data. Similar transformation operations are also used in ELT processes, where they take place after the data has been loaded into the target environment.

Data transformation is essential for preparing high-quality data in a modern BI architecture because it turns raw data into a structure that supports reporting, analytics, and data-driven decision-making.

Related terms

← Back to the glossary

REPORTING.CZ, s.r.o.
Budova Myslbek
Ovocný trh 1096/8
Praha 1, 110 00

IČ: 24248991
DIČ: CZ24248991

The company is incorporated in the Commercial Register administered by the Municipal Court in Prague, Section C, File 197204.

Data Protection (GDPR)

Information Security Policy ISMSGlossary