Data Ignorance can sound like a dirty word, and in truth, it can be detrimental to an organization. However, the causes of Data Ignorance, are not all bad, and often times, are rooted in good intentions, but with the damage that can be done from being ignorant to data, it is especially necessary that companies tackle this problem before it compromises their company.
The number one cause of data ignorance is Intuition or experience. This is where data ignorance may not necessarily start out as a bad thing. Many companies are built from years of experience in an industry. The CEO spends years developing their idea while working in the same industry, understanding the nuances of the business, the customers, and the pitfalls. Then, they spend years building their business, bringing on top talent from similar companies, expanding their experience and understanding of the business, the landscape, and the processes. Over time this industry tribal knowledge becomes so ingrained in the business that it becomes seen as fact.
On the surface this method of starting a business is not wrong and is probably necessary due to a lack of data and a need to make gut decisions. These are where the CEO of a moving and storage company makes the call to send out a larger truck, because of the years he spent in the moving van, and remembers that customers in the summer usually have more things than expected. It’s the gut instincts that are bets the CEO and leaders of early companies need to make, to get themselves to milestones that seem unattainable.
However, as this tribal knowledge continues to take root and expands through the business, it causes a company to ignore what’s happening and trust what they remember happened. This is when a bond rating agency continues to give AAA ratings to companies based on past performance, without noticing the metrics changing beneath them.
This is one of the hardest causes of ignorance because it is rooted in good intentions and is backed by successful decisions and accomplishments throughout the early years of the company. The best method for curbing this decision is to question the age-old assumptions. This is obviously easier said than done. When attempting to question the assumptions, this type of data ignorance fires back with statements like, “I’ve seen this for 20 years”, “this worked last year”, and “Why don’t you trust what we know”.
We have found that the best way to combat this type of data ignorance is through systematic back-testing of data. Testing the assumptions in real-time can be challenging, because you are going up against the gut instinct bets that others have already made. By testing the age-old theories on the ages-old scenarios, you can begin to see how often these assumptions held. As you begin to see things deviate you begin to provide new insights that can inform those “gut” instincts and begin to merge intuition with data. This creates a synergy between the experience and data allowing you to shift from data ignorance to data informed decision making.
The second major cause of data ignorance is Fear of what the data will say. This is where data ignorance becomes a direct choice and can derail even the strongest of businesses. Fearing what the data will tell you leads a company to not want to understand the data. This is when companies begin to rest on the metrics that they know, that they like, that they feel like paint the nice picture. If you have a fear that your LTV might be lower than you think, it’s easier to not measure it than to have to tackle fixing it.
This type of data ignorance can cause teams to focus on arbitrary or vanity metrics, attempting to impact areas of the business that aren’t the true levers. A logistics company we worked with had significant issues with their transportation costs, they didn’t monitor them on a consistent basis, because it would take a significant amount of time and resources to develop a new routing system. As a fast growing company, the logistics company was focused on many other aspects of their business, and without sufficient visibility into these transportation costs, the Company defaulted to data-ignorance on one of the key drivers of their business.
When you’re a kid and you fear the monster under the bed, you go and get your Dad to look with you. You know that you have a plan, if there is a monster, you have help to tackle the fear! You need to create a plan that will help you tackle the problem if your fear is realized. Most companies that fear what their data will tell them don’t have a systematic approach to tackling problems that arise in their metrics. Do you have a system in place to get the right data to analyze a problem? Do you have a plan in place to determine possible fixes? Do you have a method to decide from the fixes the first tested course of action? Finally, do you have way to measure if your test has fixed the problem? If you have these steps in place, tackling problems becomes streamlined and a lot less scary. It’s easier to lift up the covers, if you know how you will take on the monster.
The logistics company we worked with, had some of these pieces in place, but ultimately needed to develop a systematic approach to overcoming these problems. Before finding out the cause of the poor pick-up and delivery costs, the team determined a plan of action that would involve systematic testing of solutions, utilizing low-tech test fixes to iterate quickly, and creating reporting dashboards to monitor progress along the way. With these steps in place they were able to pull back the curtain and make a drastic impact in this aspect of their business driving their costs down by nearly 40%! Now, with a plan in place, they are prepared to tackle new challenges head-on.
The third cause of data ignorance is Accessibility to the data. This happens at companies of all sizes, the right people can’t get access to the right data in a timely manner to make the right decisions, so they become ignorant to what the data is telling them. This type of ignorance happens at small companies when they don’t have the sophistication of collecting the right data, or the systematic approach of getting the data to the right people. This happens in large companies just as often, but for a different reason. Large organizations have the sophisticated systems to collect and store any data that could be necessary, but don’t create ample access points to the people that need to make the decisions. The separation between the decision makers and the data managers becomes so large that leaders become ignorant to the data and are forced to fly blind in their decision making.
Accessibility to data includes a number of processes and critical points, a failure to employ any of them can lead to data ignorance. Accessibility starts with collecting the right data and storing it efficiently. It doesn’t end their though, you also need a systematic reporting of the data to business leaders as well as ample data interpretation to pull value out of the data that is collected, stored, and presented.
A large money transfer business we worked with ran into a number of issues with access and interpretation. The data collection and storage was handled nicely, but creating access points to so many stakeholders and decision makers was not handled correctly. Often times the teams managing data reporting and modeling would take many weeks to get the appropriate access to the databases and then would provide in-depth analysis to the decision makers, after they had already made an implemented their solution, ignorant of what the data said.
Solving data accessibility issues that lead to data ignorance is all in setting up the right system. However, setting up a data analytics system doesn’t have to come with a large price tag! The points necessary for setting up a strong analytics system that prevents data ignorance are, collection and storage, aggregation, reporting, and presentation and findings. Each of these steps can be accomplished through a number of different software’s that do parts or all of these. Your choice of software will depend on a number of factors, but ultimately the package you create needs to accomplish these goals.
With many of our companies we have utilized a combination of Alteryx and Tableau to do much of the heavy lifting for aggregation and reporting and presentation. Our team has also leveraged R and Python for pure analysis to provide the deep data insights that present themselves from reporting. Other tools to investigate for your solution should include Looker, Domo, and Power BI. Each tool has it’s perks and pitfalls, but all can propel your company forward and get you to a new steady state in your data program.
Is your company suffering from Data Ignorance? Can you curb this problem before it becomes detrimental? What processes are needed to grow your data program to meet the current landscape? How can we help to supercharge your data program?