How Leaders Blend Data And Intuition To Make Better Decisions

During the rise of digital transformation over the past two decades, the promise of data has loomed large. No doubt, data is essential for understanding your customers, growing your business, and measuring success, but it’s not the only thing you need. Good decisions require both data and intuition.

Many people have come to the erroneous belief that data is king and intuition is the jester. At times it’s seemed that the two are engaged in a tug of war, assuring that neither can reign with the presence of the other.

This couldn’t be further from the truth. Intuition also plays a role in all good decisions. When data and intuition are partnered, they create a cycle of feedback that refines and strengthens mental models. Intuition can lead to the right question to ask of the data, with the resulting story informing intuition. Intuition can warn us when data is incomplete or has quality challenges. While data can help us recognize when we are operating from biases or the circumstances have shifted.

This matters in an age of growing uncertainty, with new business challenges around every corner. Data can give a firm understanding of the past, but when we get too caught up in preciseness—in accuracy, in crafting the perfect data model—we can miss out on what is happening right in front of us. Intuition can help us quickly make sense of directionality, which can be as impactful to decision making as any quantitative figure. When used appropriately, intuition and data can be your two main allies in achieving victory against uncertainty.

Decision making in the real world

We spoke with Michael Nolting, the senior director of Digital Services and Data Analytics at Volkswagen, and Michael Sasaki, former vice president of Global Head of Customer Success and Support at Mitek, to learn how their companies balance data with intuition to make decisions and drive business results.

Tableau: How are decisions made at your company?

Nolting: We worked really hard over the last years to make our car production data-driven [at Volkswagen]. We created a platform called Snowpark, which harvested all the data we had from our test drives and customers. We analyzed if there was a gap in terms of the car usage.

If we understand how real customers use our cars, we can build cars according to their needs and deliver better products—as well as minimize the overall cost.

We make decisions at Volkswagen based on gut [feelings] and data. Data is preferred and can be used to gradually optimize something. Your gut is needed for exploration, when you make hard decisions based on not enough data (due to a lack of data, too many input dimensions, too low effect size, or too much context knowledge needed). Core business has to be moved as far as possible into the data zone.

For risk-taking, you need a hierarchy based on the amount of risk to take. C-level leaders have to take risks.

Data from our MOIA fleet (a shared mobility solution in Hamburg and Hannover), has been democratized. It can be accessed by anyone at Volkswagen with an account.

Our goal is to internally democratize all our data. We’re currently building a huge data warehouse in my department, where we want to enable every business [user] to import and analyze data. We make every business [user] a data engineer/data scientist.

Sasaki: Making decisions [at Mitek] requires alignment among stakeholders. Ultimately, there are final decision makers, and they are usually the functional experts who end up making the decision. But we do spend a lot of time meeting and making sure we all have the same information and are looking at the same data, understand the data, and agree on the definitions.

Tableau: How do you balance data, intuition, and experience when making decisions?

Nolting: Intuition is needed for heavy-loaded questions when people finally have to take risks and there is not enough data available due to the high complexity of the model/question.

We’re still in the gut zone with a share of our core business and want to move it step by step into the data zone to become a data-driven company. Nevertheless, innovation projects or exploring new business opportunities will always remain partly in the gut zone. What is the challenge with the gut zone, if your core business is still there? In the gut zone, if you want to answer a question, which has a high risk (read: millions of dollar you could lose) you need managers of the company who are willing to take the risk. According to this we, of course, have a hierarchy. Based on the estimated risk in euros, we have different management levels, who can take the risks. If the risk is about millions, C-level steps in.

Sasaki: They’re all intertwined in my mind.

Data is super important. With data, you start to see a hybrid of data informing your gut. You’re making decisions based on customer data. And that is that experience that you have working with the data, and seeing the outcomes that you’ve driven with customers really helps get you to the right place. That experience is super important working with the data.

So I wouldn’t say it’s one or the other. It’s a hybrid of both right now. And both are super important. The gut is driven by the data.

Tableau: When do you know you have enough data to make a decision?

Nolting: You cannot say, “Do we have enough data?” or “Do we not have enough data?” This is more about connecting the right systems and having good data. The question is always between quality and quantity.

When companies undergo a data transformation, the big issue is data quality in the beginning. You have to really look into the data if you can work with it or not. For certain dashboards, you need high quality sales data. You need data stewards.

For big effect sizes, you need a small amount of data (e.g., from small car fleets). We wanted to find out how our commercial customers like [parcel shipping company] DPD use their cars in comparison to the drivers of our shared mobility solution, MOIA. This data can be collected from a test fleet. If we want to measure small-effect sizes, we take data from our big fleet.

We also use Tableau dashboards to help prioritize which components are produced based on the shortage of components we have. One dashboard predicts the orders of the components that we need. It’s really complex—there are billions of combinations. And then we do the calculation and order the components when we have a shortage. This results in an optimal production process.

Sasaki: Five to ten years ago, there was a lack of data. And now there’s so much data. Trying to figure out what data is important is really the key and the challenge. Because you can look at data to justify almost every single decision you want to make. And that’s a pitfall you can fall into, where you have the decision you want to make, and you look for the data to justify it, so that the data are really revealing the path you need to follow.

So the question is, when do you know you have enough data to make the decision?

I would say, well, here’s my customer success experience with customer-related decisions. You can take a look at customer bright spots to see what data was present to drive the desired outcome you delivered in the past. So we look a lot at outcomes that were driven, and then what data was really important that really drove that decision. So we’ll identify those and really pick that apart.

We also lean a lot on our data analyst team. At Mitek, there are a lot of different types of data team setups. There’s decentralized, where there’s a data analyst in different functions—one in marketing, one in finance, one in customer success. You can have a centralized function where that’s all just one team. But data analysts work on any requests that come in, regardless of what function that comes in from.

I created and built out a data analyst role on the customer success team. That was super important for a couple reasons. I believe a data analyst needs to be an expert in data analysis, but also a functional expert in what they are analyzing the data for. Having a data analyst on the customer success team is valuable for understanding the customer data. I lean on my data analysts when they have time to help me decide when we have enough data to make a decision. And it’s a balancing act between being inaccurate and being inactive.

Which is more costly—making the wrong decision, or not taking any action at all? I don’t know if you ever feel like you have enough data, but you get to a point where you’re comfortable enough that you can make a call based on the data.

Tableau: It’s easy to look at data and forget the numbers represent real, human customers. How can we defend against this mistake?

Sasaki: I’m customer facing; I’m responsible for the customer and the revenue. The product development team has its own goals, and it’s not always about the human necessarily, or maybe they don’t understand that, and it’s not their fault. It’s my responsibility as a leader on the customer-facing side, to put a face to that number, that data point.

There are certain things leaders can do to try to put a human face on data. We’ve launched a lot of programs at our company. One is a lunch-and-learn. We’ll bring in a customer and buy lunch for the entire company. Now engineers can hear from the customer and they can relate the metrics they’re looking at and driving toward to a human being, to a purpose.

Tableau: How can early career folks begin to “train” their gut?

Nolting: Young people must learn to have failures and take the risk of making decisions. This is a cultural thing that German companies struggle with. You can train your gut only by gaining experiences and making mistakes—and then you can step up to take the risk of harder decisions in the future. At Volkswagen, we have created an environment of psychological safety, where failures are accepted. To achieve this, you need to have the right enterprise and data culture.

Sasaki: [At Mitek,] we start with experience with data. Leaders on my team have turned the customer success managers into data analysts. Our data analysts have provided the tools in Tableau to turn customer success managers into data analysts. Now, if you look at the views in Tableau, across the company, 70% of the views are from my customer success managers.

You can’t be afraid of the data. You have to take every opportunity as an experience and get as many experiences with data as you can, whether positive or negative. That’s going to be really valuable for trusting your gut. Just get in there, understand the data, play around with it, ask questions, and get as many experiences—positive or negative—that you can. And that will really train your gut.

If you have data, you can’t argue against it. There’s no better way to work with other functions and other leaders and other team members than to have them to have the data. When you bring the data to the conversation, you can align really quickly. You can make decisions; you can even persuade customers. It’s going to be a data-driven meeting, it’s going to be a data-driven discussion. Meetings and decisions happen a lot quicker because they’re just more informed with data.”

Are you ready to lead with data?

Data-driven leaders are better equipped to adapt to change, and they understand the nuances of decision making in a fast-moving business landscape. They know that data, augmented by experience and intuition, is fundamental to success across their organizations. Visit Tableau for Executives to learn more about how data is influencing a new breed of business leaders, and how Tableau can power your data transformation.