Data Integrity

One click and the report is generated: data analysis simplifies many strategic business decisions. In theory. In practice, surveys show that one in five executives has doubts about the trustworthiness of the available data. If companies want to motivate their employees to act based on data rather than intuition, they must invest in data integrity.

Definition: What Is Data Integrity?

A uniform definition of what is meant by data integrity has not yet been adopted. In some cases, integrity is equated with the correctness of data. In others, the definition is broader. For a common understanding, the following definition of data integrity has proven itself in corporate practice.

Data integrity refers to the reliability of data throughout its lifecycle. In addition to correctness, completeness and consistency, data security is a central aspect of data integrity. To ensure it, companies implement processes, rules and technologies.

Two basic types of data integrity can be distinguished: physical and logical data integrity.

Physical data integrity refers to the protection of data assets from physical impact factors. Examples: fires or floods that destroy servers, power outages that make data access impossible, or hacker attacks that steal data. Hardware degradation or user error can also compromise physical data integrity.

Logical data integrity refers to the operation of relational databases. Rules ensure that only desired changes are made, data is not stored twice, and certain formats are adhered to. Volume limits are also used to ensure logical data integrity in entities and data domains.

Why Is Data Integrity Important?

Sales figures for individual products do not match actual sales because data has not been fully transferred from one system to another. Incorrect data from a market analysis leads to expansion in ultimately empty markets. These are just two examples of how poor data integrity can lead to serious business mistakes and massive financial damage.

According to a 2020 market study, 60 percent of companies in Germany (and the UK) place data at the center of their decision-making processes. In the USA, the figure is as high as 77 percent.

However, data does not only play a central role in major directional decisions. In the operational business, HR, controlling and marketing regularly take data as a basis for next steps. For the simple business user, the correctness of the data is no longer comprehensible. He must be able to rely on data integrity.

Risks for Data Integrity

Data integrity is influenced by various factors. To ensure the reliability of their data in the long term, companies should define technical and procedural measures against negative influencing factors in a global data integrity concept.

User errors

If users enter data in the wrong field or in different formats, it can cause inconsistencies in data sets. Analysis tools can then no longer detect redundancies or evaluate data incorrectly. Errors of this kind can be avoided: through good training, clever user guidance or technical limitations.

Transfer errors

Users may accidentally delete records that reference other tables in the database or choose incorrect locations when migrating and copying records. Such errors are not always discovered immediately, leaving applications with incomplete data sets. Again, training, user guidance and technical solutions reduce the possibility of errors.

Hacker attacks

Hackers can inject malware into systems, altering or deleting data in databases or preventing access. Tampering with hardware can cause computers or servers to crash repeatedly, restricting the execution of programs and thus limiting or preventing access to data files.

These risks for data integrity can be minimized if companies implement clear processes to ensure data integrity as part of global data governance.

ALCOA Principle of Data Integrity

The ALCOA principle has proven itself in the handling of data. It ensures data integrity and is considered a good documentation practice in regulated industries. The acronym ALCOA stands for:

Attributable: Can the action be assigned to a person or system?
Legible: Is the data readable?
Contemporaneous: Was the time of the action documented?
Original: Was the data stored in the original or in the form of a certified copy?
Accurate: Was the data stored without errors or were corrections traceably marked?

In the meantime, the principle has been extended to ALCOA-Plus. According to this, data must additionally be stored in a complete, consistent and durable manner and be permanently available during it's lifetime.

Data Integrity - Checklist: The Best Approach to Establishing Data Integrity

The larger and more complex the IT infrastructure, the more challenging the task of ensuring data integrity throughout the entire data lifecycle. Medium-sized and larger enterprises in particular benefit from a central data governance framework in which they define technologies, processes and rules for ensuring enterprise-wide data integrity.

To check the status quo of data integrity, companies should clarify the following aspects in particular:

How is data currently collected and protected from forgery and loss?
Is access to data only possible for employees with a legitimate interest?
Are regular internal audits conducted to check data integrity?
How is data backed up and archived?
What tools and processes are used for data integration?
What backup processes and systems are used?
How are processes and interfaces validated?

Parsionate supports companies in analyzing their systems and processes within the framework of "Data Management Benchmarking". With the help of analysis tools and best practices, we identify fields of action and derive concrete measures to optimize data integrity.

Frequently Asked Questions Around Data Integrity

: Data quality management tools can check whether data contain prescribed attributes, have consistent formats, or have redundancies and adjust data records accordingly. Employees should pay attention to the validity of data as early as the data entry stage. Regular security checks can indicate unauthorized or conspicuous data changes so administrators can take corrective action.
: Data quality and data security are facets of data integrity. Data quality refers to the business requirements for data, such as completeness, timeliness, and traceability. Data security refers to the protection of data from mishandling.
: In the manufacture of medicines, it is essential that companies prove product quality on the basis of test results so that no ineffective preparations are marketed or medicines with a high risk of side effects reach the market. Independent regulatory agencies verify the efficacy and safety of medicines through comprehensive data analysis. If there are doubts about the integrity of the data, the medications are not approved in order to protect public health.

Quality despite quantity

Poor data quality often goes undetected until an incorrect database leads to a serious wrong decision. We explain how you can protect yourself from such mishaps and ensure high data quality - and why it is worth systematically improving it.

Learn more