Communicating Data: 3 Lessons Every Leader Must Know

Data is vital for business success, yet many leaders struggle to use it effectively. Often, they are misled by how data is presented. This article is designed for leaders who aren't trained analysts and for junior analysts. Over the past two decades, I've observed common patterns in how data is communicated and misused. This article aims to help you avoid these pitfalls and equip you with tools to recognize potential traps in data communication. Here are three key insights to enhance your data-driven messaging.

Causality Matters

Data are king, but they can also mislead without proper context. I remember a particular MBA program advertisement. It claimed that over 50 of its graduates secured positions at FAANG companies. This statistic, while impressive, fails to paint a helpful picture. Without knowing the total number of graduates, we can't even estimate if an MBA helps to land a better job. I call this pattern the "data as an illustration". This illustration does not bring value in the sense of a data-driven culture. It is a noise.

To ensure that the data you are presented with adds real value, ask yourself the following questions. They help you to diagnose whether the data is relevant and contextually appropriate:

• What is the goal, and how does this data relate to achieving it?
• Does this data illustrate a potential outcome or a guaranteed outcome?

These questions provoke thought, ensuring that statistics start meaningful discussions. This prevents data from being just decorations.

Beyond Confirmation Bias

The second pattern refers to cases when data builds a false or incomplete context. I recall a post from the European Commission that showcased selective data. The intention was good, but data usage was at the kindergarten level. This kind of blunder can invalidate the whole message. Consider the post yourself.

I labeled this pattern "prove my point." As a leader, it's crucial to ensure that the information you rely on is a holistic and fair representation. To do this, ask these critical questions when reviewing data:

• "What other factors might influence the outcome?"
• "Is the measuring to the outcome and the impact of presented factors on the outcome valid?"

Such questions ensure a comprehensive view of the data and limit confirmation bias.

Conquering Information Overload

The trained analysts are aware of the above pitfalls. So, they present a holistic view of the data and explain the causal relationships. Sometimes, they create information overload. Showing all data is equally bad as cherry-picking; both lead to diminishing data value. A pivotal moment for me was finding a resolution to this conflict. The solution is to concentrate on cause-and-effect relationships.

The business value is understanding what factors impact the business outcome and how much. If you, as a business leader, find yourself in an overwhelming discussion, don't give up. You can help the analyst and your team by asking guiding questions:

• "Which factors most significantly impact our outcome?"
• Can you quantify the impact of the factor, at least approximately?
• Which of these factors can we effectively control?"

These questions quickly shift the conversation's focus to actionable insights. "What aspects of the business can be influenced or controlled, and to what extent do they impact the business outcome?" A valuable conceptual tool for analysts to identify causal relationships is the "value driver tree."

Takeaway

Recognizing which failure mode occurs in a particular team helped me improve their data-driven culture. Your ability to determine what blocks the capturing value from data helps you, too.

Share your stories about failures in communicating with data.

#ImpactfulAnalytics #ImpactfulAI #PracticalHints #ThoughtLeadershipForAnalytics #AIandML #datadrivenculture