Why Data Driven Decision Making is Your Path To Business Success

Worker looking at his smartphone with KPIs while working on dashboards on the computer

Big data, analytics, business intelligence…

We read about it everywhere. Once we implement the right tools, so they say, and figure out how to analyze the data correctly, that knowledge will translate to better business decisions. It sounds great in theory. But in practice, even if you have the best data in the world, decisions are sometimes made in spite of that data, or with what is often described as going with a gut feeling. Before we go over the whys and hows, let’s clarify what we are talking about.

At The Origin, Data Science

Data driven decision making is a way of working that values the business decisions backed up by verified and analyzed data. Such governance is possible the quality of data gathered is ensured. It used to be a long and difficult process to collect, extract, format and analyze the data requiring full-time experts, which of course impacted the time required to take action and the quality of these decisions. But today, the development and democratization of business intelligence software allows anyone without a heavy technical background to analyze and gain insights from their data, requiring much lower support from the IT department to produce the reports that will later need to be analyzed, accelerating then the decision process. From there, was born data science: when hacking skills, statistics and expertise meet. This fairly new profession involves sifting large amounts of raw data to make intelligent decisions.

The gold that data scientists mine comes in two varieties: qualitative and quantitative, and both are critical when it comes to decision making. Qualitative analysis uses data that is not defined by numbers, e.g. interviews, videos, anecdotes. Qualitative data analysis is observed rather than measured. The data must be coded so that items may be grouped together intelligently. Quantitative data analysis is based on numbers and statistics. The median, standard deviation, and other descriptive stats play a major role. This type of analysis is measured, not observed. Both qualitative and quantitative data should be analyzed to make smart business decisions.

Why Is Data Driven Decision Making So Important?

Data driven business decisions make or break companies. Such governance is undertaken in order to be more competitive.

MIT Sloan School of Management professors Andrew McAfee and Erik Brynjolfsson explain in a Wall Street Journal article that they performed a study with the MIT Center for Digital Business. They found that among the companies surveyed, the ones that were mostly data-driven had 4% higher productivity and 6% higher profits than the average.

Companies that approach decision making collaboratively tend to treat information as a real asset more than in companies with other approaches. That way, they tend to identify business opportunities and predict future trends more easily, and generate more revenue with data. 

Examples of Data Analysis Leading To Data Driven Decision Making and Business Success

One of the best examples of data driven decision making falls to Google. Startups are famous for disbanding hierarchies and Google was curious whether having managers actually mattered. To answer the question, data scientists at Google looked at performance reviews and employee surveys from the managers’ subordinates (qualitative data). The analysts plotted the information on a graph and determined that managers were generally perceived as good. They went a step further and split the data into the top and bottom quartiles, then ran regressions. These tests showed large differences between the best and worst managers in terms of team productivity, employee happiness, and employee turnover. So good managers make Google more money and create happier employees, but what makes a good manager at Google?

Again, the analysts reviewed data from the “Great Manager Award” scores, in which employees could nominate managers who did an exceptional job. The employees had to provide examples explaining exactly what made the manager so great. Managers from the top and bottom quartiles were also interviewed to round out the data set. Google’s analysis found the top eight behaviors that make a great manager at Google and three that don’t. They revised their management training, incorporating the new findings, continuing the Great Manager Award and implementing a twice yearly feedback survey.

Walmart used a similar process when it came to emergency merchandise in preparation for Hurricane Frances in 2004. Executives wanted to know the types of merchandise they should stock before the storm. Their analysts mined records of past purchases from other Walmart stores under similar conditions, sorting a terabyte of customer history to decide which goods to send to Florida (quantitative data).  It turns out that, in times of natural disasters, Americans turn to strawberry Pop-Tarts and beer. Linda M. Dillion, Walmart’s CIO at the time, explained that, “by predicting what’s going to happen, instead of waiting for it to happen… trucks filled with toaster pastries and six-packs were soon speeding down Interstate 95 toward Walmarts in the path of Frances. Most of the products that were stocked for the storm sold quickly.”

Walmart’s analysts not only kept Floridians pleasantly buzzed on beer and Pop-Tarts during the storm, but also created profits by anticipating demand.

Why Business Leaders Often Don’t Make Data Driven Business Decisions

image depicting data driven decision making with gearwheels, nuts and bolts in action in the brain

How many times in your life have you prepared for a meeting, had the facts and figures ready to go, and in the end the decision goes the complete opposite direction? It probably felt like the decision had been made before the meeting even began. If this sounds familiar, you are not alone. We aren’t just talking about a startup full of newbies who think going with their gut is more critical than KPIs; we are talking about huge companies. Rob Enderle, a former IBM employee and Research Fellow for Forrester wrote a fabulous article which documents the shortcomings of executives at IBM and Microsoft. While the article is chock full of examples, perhaps the most egregious is IBM’s partial sale of its ROLM division to Siemens. Enderle and team produced an internal report that proved selling to Siemens would be a catastrophic failure. It turned out that the decision had been made before the research came out. In fact, executives forgot the research had been commissioned at all. Their gut decision ended up costing the company over one billion dollars.

publication from BI-Survey show us that 58% of the companies they surveyed said that they base at least half of their regular business decisions on gut feel or experience, instead of being data and information-driven. On average, they realized that the companies would use only 50% of the information available when it came to decision-making.

As business intelligence providers ourselves, we create an online data analysis tool that enables clients to get the most out of their data, visualize it in a meaningful way and easily share these generated insights in stunning real time dashboards to make better business decisions faster. However, the insights we provide are completely useless if, at the end of the day, these reports are ignored by the actual decision makers. This conundrum prompted us to take a deep look: why are business leaders not using data driven decision making? And what should you be aware of to make sure your decisions are based on numbers, not feelings?

Quality of the data

First and foremost, the main reason usually invoked is the data quality. Data quality is the condition of a set of qualitative of quantitative variables, that should be “fit for [its] intended uses in operations, decision-making and planning”, according to Wikipedia. A good data quality management (from the acquisition to the the maintenance, from the disposition to the distribution processes in place within an organization) is also key in the future use of such data. Collecting and gathering is only good if well managed and exploited afterwards, otherwise the assets’ potential remains untouched, and useless.

Overreliance on past experience

Overreliance on past experience can kill any business. If you are always looking behind you, there is a real chance of missing what is in front.  So often, business leaders are hired because of their previous experiences, but environments and markets change and the same tricks may not work next time. One of the most cited examples of this is Dick Fuld, who saved Lehman after the LTCM crisis. Ten years later he pulled out the same bag of tricks and, as the Wall Street Journal Reports, “the experience he was relying on was not the same as this massive housing-driven collapse.” The recent crisis was much more complex. Environments and markets constantly change and, in order to be a successful manager, one must combine past experiences with current data.

Going with their gut and cooking the data

While some managers naturally go with their instinct, there is a significant portion who first trust their gut, then persuade their researchers or an external consultancy to produce reports that confirm the decision that they already made. According to the Enderle article mentioned above, this was commonplace at Microsoft. Researchers were tasked with providing reports that lent credibility to the executives’ decisions.

Cognitive biases

Cognitive biases are tendencies to make decisions based on limited information, or on lessons from past experiences that may not be relevant to the current situation. Cognitive bias occurs every day, in some way, in every decision we make. These biases can influence business leaders to ignore solid data and go with their assumptions, instead. Here are a few examples of cognitive biases commonly seen:

  • Confirmation bias –  Business leaders tend to favor information that confirms the beliefs they already have, right or wrong.
  • Cognitive inertia – The inability to adapt to new environmental conditions and stick to old beliefs despite data proving otherwise.
  • Group Think – The desire to be part of the group by siding with the majority, regardless of evidence or motives to support.
  • Optimism Bias – Making decisions based on the belief that the future will be much better than the past.

Managers need to recognize that we are biased in every situation. There is no such thing as objectivity. The good news is that there are ways to overcome biased behavior.

9 Tips And Takeaways For An Enhanced Data Driven Decision Making Strategy

1) Guard against your biases

Much of the mental work we do is unconscious, which makes it difficult to verify the logic we use when we make a decision. We can even be guilty of seeing the data we wish was there instead of what’s really in front of us. This is one of the ways a good team can help. Running your decisions by a competent party who doesn’t share (or even know) your biases is an invaluable step.

Working with a team who knows the data you are working with opens the door to helpful and insightful feedback. Democratizing data empowers all people, regardless of their technical skills, to access it and help make informed decisions. Often this is done through innovative dashboard software, visualizing once complicated tables and graphs in such ways that more people can initiate good data driven business decisions.

With more people understanding the data at play, you’ll have an opportunity to receive more credible feedback. The proof is in the numbers. A 2010 McKinsey study of more than 1,000 major business investments showed that when organizations worked at reducing the effect of bias in their decision-making processes, they achieved returns up to 7% higher. When it comes to data driven decision making, reducing bias and letting numbers speak for themselves make all the difference.

Tips for overcoming a biased behavior

  • Simple Awareness – Everyone is biased, but being aware that bias exists can affect your decision making can help limit their impact.
  • Collaboration – Your colleagues can help keep you in check, since it is easier to see biases in others than in yourself. Bounce decisions off other people and be aware of biased behavior in the boardroom.
  • Seeking out Conflicting Information –  Ask the right questions to yourself and others to recognize your biases and remove them from your decision process.

By eliminating bias, you open yourself up to discovering more opportunities. Getting rid of preconceived notions and really studying the data can alert you to insights that can truly change your bottom line. Remember, business intelligence shouldn’t only be about avoiding losses, but winning gains.

2) Define objectives

To get the most out of your data teams, companies should define their objectives before beginning their analysis. Set a strategy to avoid following the hype instead of the needs of your business and define clear Key Performance Indicators (KPIs).

3) Gather data now

Gathering the right data is as crucial as asking the right questions. For smaller businesses or start-ups, data collection should begin on day one. Jack Dorsey, co-creator and founder of Twitter, shared this learning with TechCrunch. “For the first two years of Twitter’s life, we were flying blind… we’re basing everything on intuition instead of having a good balance between intuition and data… so the first thing I wrote for Square is an admin dashboard. We have a very strong discipline to log everything and measure everything”. That being said, and done, implementing a dashboard culture in your company is a key component to manage properly the tidal waves of data you will collect.

4) Find the unresolved questions

Once your strategy and goals are set, you will then need to find the questions in need of an answer, so that you reach these goals. Asking the right data analysis questionshelps teams focus on the right data, saving time and money. In the examples earlier in this article, both Walmart and Google had very specific questions, which greatly improved the results. That way, you can focus on the data you really need, and from bluntly collecting everything “just in case” you can move to “collecting this to answer that”.

5) Find the data needed to solve these questions

Among the data you have gathered, try to focus on your ideal data, that will help you answer the unresolved questions defined at the previous stage. Once it is identified, check if you already have this data collected internally, or if you need to set up a way to collect it or acquire it externally.

6) Analyze and understand

That may seem obvious, but we have to mention it: after setting the frame of all the questions to answer and the data collection, you then need to read through it to extract  meaningful insights that will lead you to make data driven business decisions.

7) Don’t be afraid to revisit and reevaluate

Our brains leap to conclusions and are reluctant to consider alternatives; we are particularly bad at revisiting our first assessments. A friend who is a graphic designer once told me that he would often find himself stuck towards the end of a project. He was committed to the direction he had chosen and did not want to scrap it. He was invested, for the wrong reasons. Without fail, when this happened he would have to start all over again to see the misstep that got him stuck. Invariably, the end product was light-years better reworked than if he had cobbled together a solution from the first draft.

Verifying data and ensuring you are tracking the right metrics can help you step out of your decision patterns. Relying on team members to have a perspective and to share it can help you see the biases. But do not be afraid to step back and to rethink your decisions. It might feel like defeat for a moment, but to succeed, it’s a necessary step. Understanding where we might have gone wrong and addressing it right away will produce more positive results than if we are to wait and see what happens. The cost of waiting to see what happens is well documented…

8) Present the data in a meaningful way

Digging and gleaning insights is nice, but managing to tell your discoveries and convey your message is better. You have to make sure that your acumen doesn’t remain untapped and dusty, and that it will be used for future decision making. With the help of a great data visualization software, you don’t need to be an IT crack to build and customize powerful business dashboards that will tell your data story and assist you, your team, and your management to make the right data driven business decisions.

9) Set measurable goals for decision making

After you have your question, your data, your insights, then comes the hard part: decision making. You need to to apply the findings you got to  the business decisions, but also ensure that your decisions are aligned with the company’s mission and vision, even if the data are contradictory. Set measurable goals to be sure that you are on the right track… and turn data into action!

Now that you have all the keys in hand to make the best data driven business decisions, start chasing them! Remember the most important points to avoid the wrong behaviour towards data – ignoring it, not relying enough on it and going with your guts and biases, or not implementing the right data culture in your company.

Source: https://www.datapine.com/blog/data-driven-decision-making-in-businesses/