Using data without context is a bit like going on a blind date knowing only someone’s name
Nowadays there is tremendous pressure for everything to be data-driven. Absolutely every opinion is supported by a number of statistics. Every decision-making must be data-driven as well. Sometimes driven by random data. The key is that there is data. It seems to me that sometimes we fall into a data trap. Everywhere you look, in business, online, or on your step-counter, there are numbers or statistics that promise you clarity and. If you work with data, it’s probably one of your tasks to analyze it and provide insights and hypotheses. These are supposed to be the engines for making decisions. Data-driven decisions.
Using data without context is a bit like going on a blind date knowing only someone’s name.
Using data without context is a bit like going on a blind date knowing only someone’s name. Yes, you know his name is ‘Jack.’ You probably even remember THAT handsome Jack from your college days, which is why the name brings pleasant associations and makes you feel like going. But actually is this the Jack who climbed Mount Everest? Or is it the Jack who’s allergic to cats? Or even worse — the Jack who eats pizza with pineapple? Without the backstory (context), you’re navigating the evening with potential surprises (not always the pleasant kind).
Data vs Information
Imagine receiving a sheet filled with random numbers. Without headings, explanations, or any clue, what is the actual value of these numbers? Not much that I see. But if you were told these numbers represent the monthly sales figures of your favorite store, suddenly they start telling a story. This differentiation — raw numbers -> meaningful insights — is the leap from data to information.
There is also that little goblin in your head that affects how you perceive information — bias.
The power of context
Consider the simple act of weighing yourself. If the scale reads 70kg, that’s a piece of data. But what if you’re a wrestler trying to fit into a specific weight category, or recovering from an illness? The same number takes on a very different significance with added context. Without context, data is like a compass without a magnetic north — it might point you somewhere, but not necessarily in the right direction.
Biases
Besides, there is also that little goblin in your head that affects how you perceive information — bias . Back to my example about the blind date. I mentioned that the name Jack may bring positive associations because you remember somebody with that name. Maybe you believe that one handsome person named Jack means that all individuals named Jack are too. We talk here about representative bias. Do you know the saying ‘You’ll never know how many people you hate until it’s time to choose a name for your child’? This can be representative or availability bias.
Biases can also make your life miserable when you analyze data.
Here are three (although I encourage you to dig deeper):
- If you surveyed only urban populations about transportation habits, you’d likely overestimate the use of public transport compared to rural areas. This is sampling bias.
- A coffee lover (like me) might only read articles praising coffee's health benefits, thereby getting a one-sided perspective, which is a confirmation bias.
- If we only look at successful companies to determine business strategies, we miss the lessons from the many that failed. That is a survivorship bias.
Risk of Misinterpretation
Even the most reliable data can be misread without context. A company might celebrate a sudden increase in website traffic. If nobody realizes this boost is due to a negative viral news piece though, the company might make optimistic business decisions on a faulty premise. It is necessary to keep a holistic view where data isn’t just quantified but qualified.
The Role of Storytelling
Data tells you what is happening, but stories explain why. While data can tell you 200 books were bought, the story behind it might reveal a drive to increase children’s literature, an initiative born out of a local child literacy challenge. Not only does the storytelling provide a background story for your data, but it also gives a human touch to the cold hard facts, making the information more relatable. It bridges the gap between numbers and real-world implications. Thus, while data offers structure and foundation, stories give it meaning and life.
How to Ensure Context?
To avoid the pitfalls of context-less data:
- Always question the source. Question the data. Ask questions to understand. Where did the data come from? What methods were used to gather it? Understand your data.
- Collaborate with diverse teams and reach out to experts. Different perspectives can offer varied contexts.
- Gather, maintain, and consult metadata. This is data about your data. It can tell you when and how a piece of data was collected.
— — -
In our eagerness to rely on data, we must remember that context sharpens our understanding of it. As we lean more heavily on data these days, grasping its context isn’t just advantageous, it’s essential. We must delve deeper than the numbers and uncover the narratives behind them.
If you feel like discussing this further, drop me a line at:
So much fun to talk to other data geeks!
