Data Quality

The amount of data worldwide is growing explosively. In 2025, it will be around 175 zettabytes. A large part of the data is generated in companies. Data about customers, costs, competitors. Whoever uses the information in the smartest way will be ahead of the competition. However, the prerequisite is that the data quality is right.

Definition

What Is Data Quality?

Data quality describes how well data can be used for an intended purpose. The quality criteria therefore differ depending on the use case. The most common data quality criteria include completeness, uniqueness, correctness, timeliness and consistency of the data.

Why Is Data Quality Important?
Effects of Poor Data Quality

9.7 billion US dollars in losses are incurred by companies every year due to poor data quality and overall productivity is reduced by up to 20 percent, according to analyses by the market research institute Gartner.

Organizations need to adopt strategic data quality management and embrace data cleaning if the growing wealth of data is to fuel, not burden, their business.

High Cost

They should pay attention to a consistent data basis already at the creation stage. If problems with the data quality are only noticed later in the inspection process, the costs for remediation can be up to ten times higher than with an incoming inspection. Companies face the highest costs with poor data quality when data is already in the target system. For example, if customers have read incorrect or misleading product information in an online shop, they may order the wrong size of goods. Returns are made and complaints to customer service increase.

Poor Decision Quality

Future-oriented companies are increasingly making their decisions based on data. Incorrect or faulty data sets lead to inaccurate or incorrect analyses. If company decisions are based on analyses with a poor database and poor data quality, major financial damage can result.

Reliance on data
Poor data quality also has the effect of reducing the company's use of systems and data, including less accurate data maintenance, because there is simply a lack of confidence in the accuracy of the data. This often results in "shadow systems" or "personal data silos" that are used instead of the official data. This creates inconsistencies, compliance issues, and unreconciled metrics.

Lack of Compliance

Missing data for meaningful evaluations is annoying. However, if legal requirements for deletion or documentation are not adhered to when storing data, this can also have painful legal consequences. The legal framework for data integrity and data management is becoming increasingly complex. Most recently, the GDPR, among other things, has created new requirements. Companies should implement data governance in order to be able to use data in a legally compliant manner.

Loss of Sales

Customer master data and behavioral data play an important role in marketing and sales. Companies are giving away revenue if they do not collect this data or do not collect it consistently, do not check their CRM data quality or do not use data for segmentation or personalization of marketing measures.

Improve Data Quality – How Do You Go About It?

According to Forbes analyses, 84 percent of CEOs fear that the data quality in their company is poor. To increase data quality, companies should implement strategic data quality management.

It should start with an inventory of current data assets and data management. The company can then develop a data strategy that is derived from the global business strategy. What measurable goals are to be achieved with the help of specific data? When there is clarity about the intended use, data quality can be defined and thus achieved in the first place.

Any existing data initiatives of individual departments should be consolidated and aligned with each other, e.g. to simplify access, break down technical silos or minimise maintenance costs.

Measuring Data Quality – What Is “Good Data”?

What characterizes high data quality depends on the intended use. Companies should determine what quality criteria their data assets should meet in order to measure their data quality and assess the success of measures. The most commonly used dimensions for data quality include:

Correctness: Do the data match reality?
Consistency: Is the data consistent within itself and with other data sets?
Accuracy: Is the data available in the defined accuracy?
Actuality: Does the data correspond to the current state of the depicted reality?
Completeness: Are all attributes defined as necessary included in the data?
Freedom from redundancy: Is the data free from duplicates?
Reliability: Is the origin of the data traceable?
Accessibility: Is the data retrievable?
Relevance: Does the information content of the data meet an information need?
Process quality: Is a data maintenance process defined and efficient?

Data Quality Management

41 percent of the most successful companies consider good data quality management to be a decisive competitive advantage, according to market research by Gartner. They see ensuring high data quality as a strategically important corporate task: it allows IT costs to be reduced, projects to be completed more quickly and market changes to be responded to more flexibly. Incorrect decisions due to inadequate data prevent the measures.

Once companies have defined key figures for measuring data quality, they need to introduce data quality management. Roles, responsibilities, standards and processes must be established to ensure that data is collected in the defined quality, that it is available where it is needed and that its quality is maintained over time. Depending on the initial situation in the company, it may be necessary to adapt technical systems, set up interfaces or migrate data.

In medium-sized and larger companies, it is also advisable to establish a central data governance committee due to the increasingly complex data structures. In this committee, experts from different departments regularly discuss and decide on data models, data sovereignty and data maintenance processes. This ensures that data quality increases continuously and keeps pace with changes in the company.

While the revision of technical and procedural data quality measures quickly achieves visible success, the introduction of a data governance body has a more long-term effect - but is essential to ensure success.

Frequently Asked Questions Around Data Quality

What Are Examples of Poor Data Quality?

What is considered good or bad data quality depends on the purpose of the data. Poor data quality always exists, for example, when data is inconsistent, inaccurate, outdated, redundant and incomplete. If an online shop is linked to an ERP system, for example, it must be ensured that the product prices are transmitted in a format that can be read by the online shop. Otherwise, incorrect information can be displayed in the shop and lead to returns and loss of reputation.

What Does Data Integrity Mean?

Data integrity refers to the consistency, correctness, and accuracy of data throughout its lifecycle. Data integrity is an important factor with regard to compliance and legal requirements. Technical measures, processes, and responsibilities should be put in place to ensure the integrity of all data processed in organizations.

What Are the Leading Data Quality Tools?

The market for data quality software is constantly on the move. Competition is fierce, and vendors compete for customers with innovative features. Leading solutions come from SAP and Informatica, but depending on the size of the company and the industry, other focal points are relevant. For an overview of current data quality tools, including filtering options and ratings, see Gartner's vendor comparison. However, even the best technology alone cannot increase data quality. Organization, processes, and technology must mesh and follow a holistic data strategy.

Faster to Data Excellence

We offer various consulting and implementation services to support companies on their way to becoming a data-driven company.

Data Quality Healthcheck

We check the essential data domains for relevant quality criteria, carry out a benchmarking and give you recommendations for the further procedure.

Data Quality Consulting

Benefit from individual consulting and implementation services - company-wide or very specifically focused on concrete issues.

Implementation of Data Quality Tools and Rules

Using our proven methodology, we implement the processes and software solutions for data governance and data quality that your organization needs.