Data quality is not only crucial in factoring. Regardless of the systems or processing procedures involved – whether traditional rule-based or AI-driven – those who tolerate poor data quality should not complain about modest results. If we reverse the ‘garbage in, garbage out’ rule to ‘quality in, quality out’, factoring and supply chain finance providers in particular will benefit significantly – in several ways. In this short article, we would like to discuss what is generally meant by ‘data quality’, the advantages of good data quality, what to look out for in your own data management, and why a good database will be of crucial importance, especially in AI-driven analysis.

 

Let’s take a few steps back: What exactly is meant by data quality? Essentially, it refers to the following factors: the completeness of the data, its accuracy, timeliness, consistency and traceability.^1,2 The following overview lists some details.

 

• Completeness: All relevant information about the claim must be available, such as debtor data (legal form, address, credit rating), payment terms, proof of delivery, invoice references, etc.
• Accuracy: This means that there are no invoicing errors, the amounts/VAT/payment terms are correct and there are no duplicates.
• Up-to-date: All creditworthiness information and limits should be up to date..
• Consistency: Is the data consistent across different systems (core banking, risk systems, reporting)?
• Traceability: Data origin, change history and processing steps are transparently documented and can be reconstructed at any time.^1,2

So far, so good. What advantages does good data quality offer from the perspective of factoring or supply chain finance providers? Specifically, these include improved risk management (lower default risks, better liquidity forecasts), greater operational efficiency (less manual follow-up, better automation) and, last but not least, fewer conflicts with regulatory requirements (compliance, DORA, etc.). In short, data quality has a direct impact on a company’s own risk, profitability, regulatory security and competitiveness.

If we look at risk analysis by a factoring provider, for example, poor data quality can have serious consequences – ranging from financial ruin to criminal prosecution. Furthermore, in the financial sector, high-quality data is the basis for security strategies and effective compliance.³ If that’s not enough of an argument, according to Gartner, companies lose up to several million US dollars annually due to poor data quality.⁴ So what can be done to ensure the best possible data quality within an entire company? What would be the path from a purely reactive approach (firefighting) to proactive data management?

One possible solution is to establish a ‘Data Centre of Excellence’. This is a centralised function that defines company-wide standards, governance and best practices for data management. Such a ‘Data Centre of Excellence’ acts as a kind of regulatory authority for all data-related initiatives and ensures that data is treated as a strategic asset. It is therefore a process that helps companies move from fragmented data management to a structured, scalable approach. This is what enables advanced analytics, AI and digital transformation.⁵ In this context, data is also referred to as the ‘single source of truth’ (SSOT).

For SMEs, which have to manage with more limited resources, careful planning and implementation of a ‘Data Centre of Excellence’ is all the more important. First, clear objectives must be formulated: What are the short- and long-term goals in terms of data quality, how will their achievement be measured in concrete terms, and are certain milestones planned? Next, the roles must be assigned internally: Who will lead the project, who will be responsible for data analysis? The next step is to define the data standards (What do we want? What do we need to comply with?) and data-related internal processes must be defined. Ultimately, it is a matter of finding suitable tools for collaboration between individual team members and for the actual data management, including analysis and visualisation.⁶ There are a number of tool operators who also offer their services to smaller organisations with correspondingly lower data volumes.

In this context, let us now consider the ever-increasing use of AI. Whether data is available and its quality play an increasingly important role in AI-controlled systems. This is because when AI agents are used, they act largely autonomously on the basis of the data and instructions they receive. In such a scenario, there is deliberately no longer any human control authority to detect errors, inconsistencies or deviations after a decision has been made. Consequently, poor data quality in an agent-based model can even lead to systemic risk, as errors are amplified by machines.⁷

 

One thing is certain: the future of data management lies in ‘autonomous data quality management,’ where AI-driven tools proactively manage cleansing, enrichment, and monitoring with minimal manual oversight. These systems use machine learning (ML) to identify complex patterns that could indicate fraud or data corruption before they impact business results.⁸ Such automated ‘decision engines’ require high-quality, consistent and rich data – otherwise these technologies will not work reliably. In other words, if you rely on low-quality data, a model’s attention will inevitably be drawn to the wrong characteristics. That is why the so-called data-centric approach prioritises quality over quantity.⁹ This brings us back to the concept of data as SSOT, which is essentially about decoupling data and logic.

Ultimately, technical tools and frameworks are only as effective as the culture that supports them. Data quality is a corporate discipline and not just an IT task. It is therefore crucial to establish a data-first culture within the company. However, the transition to such a culture can meet with internal resistance. It is therefore necessary to clearly communicate the benefits and offer training to involve all employees in the change process.¹⁰ It is important to implement data processes that are accessible and understandable to everyone involved. Otherwise, this will lead to frustration among employees and the entire system will quickly falter.¹¹

 

Furthermore, the success of data-driven initiatives must also be measurable and adaptable as needed. It should be possible to identify obstacles and implement improvements when using data across different teams.¹² Otherwise, there is a risk that, after an initial phase of euphoria, the employees responsible will quickly lose track of the big picture and subsequently lose interest in driving the project forward. Last but not least, collaboration between individual departments should be encouraged. When teams work together on shared data, it promotes a sense of collective responsibility and reinforces the cultural importance of data orientation. Introducing and maintaining a data-oriented culture should therefore be viewed as a marathon, not a sprint.¹²

Data quality is playing an increasingly important role and is even one of the decisive factors in the success or failure of AI-driven systems. It is therefore advisable for forward-looking companies to take a systematic approach to their own data quality management and invest in clear standards and tools. A sustainable data-oriented culture provides the basis for coping with change and promoting innovation. Companies that follow this approach will not only survive in a data-rich future, but also thrive.¹²

Sources:

¹ Monte Carlo Data, 2025: The 6 Data Quality Dimensions (Plus 1 You Can’t Ignore) With Examples

² IBM: Was ist Datenqualität?

³ Data Ladder, 2026: Why Data Quality Is the Foundation of Effective Compliance in Security and Investigations?

⁴ Gartner: Data Quality: Best Practices for Accurate Insights

⁵ lake FS, 2026: Building a Data Center of Excellence for Modern Data Teams

⁶ Revolent: 5 steps to building your Data Center of Excellence

⁷ The Forrester Wave™: Data Quality Solutions, 2026: How to read it and why it matters now

⁸ delpha, 2025: The Ultimate Guide to Data Quality and AI in CRM: Unlocking Maximum Performance

⁹ Human First, 2021: The Importance of High-Quality Training Data

¹⁰ statworx, 2024: Wie Führungskräfte die Datenkultur im Unternehmen stärken können

¹¹ datacoves: 3 Core Pillars to Achieve a Data-Driven Culture

¹² James Fee, 2024: Sustaining a Data-First Culture: Turning Principles into Long-Term Success