Blog 4: Mastering the Data Dance: The Potential of Clean Core and AI
Today my colleague @Wolfgang_Epting published Blog 4 about the role of data in our Blog Series. Therefore I interviewed him to get more background information.
Timothy Kaufmann: Wolfgang, why are you so enthusiastic about Data and Analytics and where do you see the big benefit of SAP Business Technology Platform?
Wolfgang Epting: To put it simply: “There Is No AI Without Data.” The recent technological revolution concerning generative artificial intelligence has exponentially increased the significance of data beyond previous levels. Generative AI has entered the boardrooms, and companies are more than ever challenged to provide data that is of high quality and ethically sound for model training. The SAP Business Technology Platform offers an integrated, coordinated Data & Analytics Stack that effectively addresses these new challenges. Customers are placing their trust in SAP, even for artificial intelligence, which further fuels my long-standing enthusiasm for data.
Timothy Kaufmann: Why does the Clean Core paradigm encompass not just applications and processes, but also data?
Wolfgang Epting: Amid diverse challenges, poly- and permanent crises, and economic uncertainty, companies need to act more agile and become more resilient. The speed of innovation needed for this is constantly increasing and will never be as slow as it is at this moment. In this context, data should not act as an inhibitor. Instead, in line with the SAP Clean Core paradigm, it must be an enabler for stable processes, rapid decision-making, and trustworthy input for generative artificial intelligence. Therefore, companies need to accord due attention to the subject of data and must treat it as an asset from which data products are generated to create added business value.
Timothy Kaufmann: You mention a really important point. Asset management for many companies is really important. Why is data often not handled in the same way and what should companies change that?
Wolfgang Epting: The awareness that data provides business value and therefore, like any other asset, needs to be strategically managed is often not sufficiently present in Line of Business departments. Data and its quality are seen as something that falls under IT's responsibility. Changing this perception and making clear that data ownership needs to be firmly anchored in business is a change that, as we all know, is never easy. To transform into a data-driven enterprise, a significant cultural change needs to occur, and this can be initiated by a business outcome driven data strategy. A considerable amount of enablement, education, and persuasive efforts will be necessary to make employees data-literate.
Timothy Kaufmann: Why should data be managed holistically across systems?
Wolfgang Epting: Maintaining high data quality in one system and neglecting it in another is not advisable. Data represents objects of the real world and are merely stored in systems, often in silos, which contradict each other or contain multiple entries for the same entity. The frequently postulated single source of truth is indispensable for establishing trustworthiness, without which analytical and AI application scenarios are worthless. Data requires a holistic treatment, as their quality in transactional systems is crucial for the smooth running of business processes, and equally vital for all downstream applications. Cleaning suboptimal data before it enters the analytical world is always reactive and will never achieve a state where data quality can no longer erode. A proactive 'First Time Right' approach should be pursued to prevent the generation of erroneous data in the first place. One cannot aim to innovate in the field of generative artificial intelligence while simultaneously managing the foundation, namely the data, without modern tools with high automation.
Timothy Kaufmann: What does generative AI mean for data strategies and data management?
Wolfgang Epting: Data quality and data fairness are the limiting factors for the adoption of generative artificial intelligence. This presents entirely new challenges that go far beyond existing tasks. In the future, it is expected that topics such as ethics will be fully mastered, not only in algorithms but also in data. Illustrative examples include the elimination of systematic bias or distortion and ensuring adequate representation of the target population within the dataset.
Timothy Kaufmann: What are the advantages of an integrated approach with the SAP Business Technology Platform?
Wolfgang Epting: We are experiencing the rise of best-of-breed approaches, such as the Modern Data Stack, coming with the hope of building a data management platform quickly and easily to meet requirements and implement modern approaches like Data Mesh. We have demonstrated that with our Business Data Fabric approach, we operate much more cost-effectively. A Data fabric architecture streamlines operations and facilitates digital transformation by connecting various data sources for seamless information flow, boosting accessibility and integration. It breaks down data silos, enabling real-time analytics and decision-making while supporting data governance. SAP Datasphere’s unified approach simplifies data management across diverse environments and preserves critical business semantics with an industry-leading semantic layer and knowledge graph.
Timothy Kaufmann: In your opinion, what does the future of data management look like?
Wolfgang Epting: The advent of Generative Artificial Intelligence (GenAI) will significantly influence future data management, emphasizing the importance of data quality, fairness, and ethical considerations, including privacy, transparency, and consent. GenAI demands real-time data handling, enhanced data security measures, and robust data integration strategies. It also requires sophisticated metadata management strategies to effectively navigate and manage the vast amount of data that GenAI can generate, necessitating a re-evaluation of traditional data management strategies.
- SAP Managed Tags:
