Unpacking the Realities of Data in Today's Organizations
Weekend insight based on recent posts from Dr. Sebastian Wernicke
LinkedIn is for me a valuable source of wisdom when it comes to data. Maybe you know Dr. Sebastian Wernicke? He posts every sunday interesting insights about Data Strategy and Data Culture. I analyzes his last 25 posts to get the following insights:
It's crucial to understand that data isn't a neutral, objective entity that automatically generates value. Instead, the emphasis lies on recognizing that data always has a human origin, shaped by human decisions, assumptions, and processes. Its interpretation and the resulting benefits are heavily dependent on the context, the skills of those who use it, and the organizational frameworks in place. While data can provide valuable insights, it also carries the risk of misinterpretations and incorrect conclusions if viewed in isolation or without the necessary understanding.
The much-touted concept of "Data Democratization," the widespread access to data for many employees, is not a panacea for better decision-making . While the idea sounds appealing, recipients often lack the necessary analytical skills or contextual knowledge to interpret the data correctly. There's a parallel with the myth of radical transparency; maximum access isn't always the key, but rather thoughtful choices about what to share. This can lead to misguided conclusions and ineffective strategies. Even seasoned analysts require significant investment in understanding datasets to ensure valid interpretations. Data democratization is a "myth created by software companies developing analytical solutions".
An effective "Data Strategy" extends far beyond mere IT infrastructure and should not be mistaken for a purely technical operational plan. Instead, it must focus on where data truly matters to the business. If the answer is "everywhere," it's probably correct but not a strategy. The strategy needs to identify precise pressure points – specific problems where better information drives immediate value. Furthermore, it needs to address what organizational or technological barriers hinder progress, such as organizational silos and hastily grown tech stacks. Finally, it must outline how insights can be concretely translated into action. Too many dashboards are where insights go to die; clear guidelines and the empowerment of teams to act on data are crucial. The goal is to use data deliberately, like a scalpel, and not to analyze everything indiscriminately.
The true value of data is not just in optimizing existing processes ("Data to Insight"), but primarily in fundamentally rethinking and changing business models and organizational structures ("Data to Impact"). "Data to Change" represents the next frontier, where data reshapes our fundamental understanding of the business, influencing operations, risk approach, structures, and core business models. This holistic view means that technical excellence alone is no longer a competitive advantage, but table stakes. The real differentiation lies in having the courage to let data challenge core assumptions about value creation and organizational design, requiring intellectual humility and a willingness to reimagine what's possible.
It's essential to recognize that all data is ultimately created by humans, whether through the design of collection processes, the definition of categorizations, or the programming of sensors. "All data ultimately has a human source—it is not collected, but created". This "human fingerprint" means that the context of data creation is indispensable for correct interpretation. Documentation on collection methods, the human decisions behind the numbers, and known biases is therefore not supplementary but a crucial "user manual" for the data. Pairing data specialists with subject matter experts is crucial for gaining critical insights.
A "data-driven" approach should not be limited to the pure measurement and optimization of key performance indicators (KPIs). It must also create space for discovery, curiosity, and the identification of unexpected patterns and anomalies, which can provide valuable insights for innovations and new business ideas. Breakthroughs often emerge when someone recognizes an unusual pattern or an overlooked anomaly. It's about fostering environments for "curated chaos," intentionally designed to surface unexpected findings, requiring intellectual bravery and patience to detect subtle signals.
A truly data-driven organization is not characterized by individual, spectacular analysis projects, but by everyday decisions based on relevant data. This requires a corporate culture that values transparency and embraces discomfort when data challenges long-held assumptions. It fosters the willingness to question assumptions, even from superiors, with evidence. In such a culture, insights emerge organically from various roles, not just data specialists. The focus shifts from "who has the data?" to ensuring everyone can access and act on it.
Data Governance should not be viewed as a necessary evil for ensuring security and compliance that hinders innovation. Rather, it is an enabler, whose goal is to make data usable and extract the highest value from it, while simultaneously minimizing risks. Successful data governance creates lean frameworks that protect assets and guide data use for growth and innovation, turning data into a competitive edge through superior decision-making and stakeholder trust. The central theme is not "things you cannot do with data," but "here's how to use it in the right way," balancing oversight with speed and agility.
The assumption that "good data" automatically leads to "better decisions" falls short. The human psychology—including biases like confirmation bias and loss aversion—plays a crucial role in the interpretation of data and decision-making. Organizations must therefore promote not only data literacy but also "Decision Literacy" among their employees, i.e., the understanding of how people actually make decisions and which psychological factors are at play. True data mastery requires us to be as fluent in psychology as we are in programming.
Many data problems are, in reality, symptoms of deeper organizational dysfunctions. Data struggles reveal how teams collaborate, how decisions flow, and how leadership shapes priorities . For example, if financial data doesn't harmonize with marketing data, this often indicates a lack of cross-departmental communication. Unclear KPIs can be a sign that difficult alignment discussions are being avoided. Data problems are thus an "X-ray" of the organizational state, and fixing them requires addressing the underlying organizational issues.
The notion that one can "let the data speak for itself" is a dangerous illusion. Data doesn't speak. Humans do. Data requires human interpretation, contextual understanding, and the ability to question assumptions to unlock its true value. As the example of product ratings illustrates, the same data can lead to different conclusions depending on what questions are asked and what context is considered. It is the task of analysts and domain experts to provide context and ask the right questions to enable informed decisions.
While these are my insights I wanted to highlight from Sebastian Wernicke’s LinkedIn posts, he also wrapped up his personal insights in this (german) data-inspired manifest I can also recommend to read: → https://www.datainspired.org/manifest.html
This aligns deeply with what we often observe around which is data problems are rarely technical. They are reflections of cultural, structural, and psychological patterns within an organization. The myth of data neutrality and the blind chase for democratization often lead companies to overlook the design intent behind their data, resulting in noise over insight. True data impact begins when organizations embrace "data to change," where data doesn't just optimize the old but also challenges us to reimagine the new.