Denominators are often more interesting than numerators.
You might not have known that. You might not care. You might be wondering what I’m doing talking about third-grade mathematics when you’re here to read about data analysis. (I’m going somewhere with this, I promise.)
So why would denominators be more interesting than numerators? (They really can be.)
If you’ve ever done any marketing work, you’ve probably heard the phrase “Context is king.” It’s actually a reaction to a remark Bill Gates made a few decades back — “Content is king” — but I’m not going to drag you through a history of marketing strategy either since that’s probably even further from what you want to hear about than the math lesson I started on.
But context is king — or at least, it should be — in the data world. Bear with me.
Finding Meaning in Data
At Datassist, we’ve spent a lot of time studying financial inclusion with different partners. We collect data and learn things like, a typical group will get a 3% return on investment, but a group with a really great management processes is getting a 15% ROI. Cool, right?
Having that information might be interesting, but it’s not really meaningful. Why? Because we don’t really know what those numbers mean. There is no context. Without the denominator, we have no clue what those percentages represent — and that context could make a significant difference in how much value we place on those figures.
The numerator generally refers to the things we’re counting — money earned, people helped, crimes committed. But the denominator is the context for those figures. And the context is usually just as meaningful as the count.
Is that 3% return on investment meaningful to the group members we’re studying? It depends on the context.
When we add a denominator to our data, we can see what amount those returns are on. Say that 3% ROI is equal to $75. It’s a lot easier to determine how meaningful that figure is to our groups. Maybe it’s a lot of money. Maybe it’s not. In that case, 15% is a $375 return. How much difference does that $300 make?
Putting Numbers in Context
The importance of denominators to go with our numerators isn’t restricted to financial data. Context matters in all kinds of data analysis. Robert Kosara does a great job using crime statistics to explain how raw numbers (without context) can be incredibly misleading.
Think about it. Knowing that 100 crimes had occurred in your city might make you feel unsafe. And if you lived in a city with a population of 1,000… well, you would probably be justified in taking some extra precautions. On the other hand, if 100 crimes had occurred in your city of 10 million, it would mean you lived in a pretty safe place. Understanding the context makes all the difference, and is crucial if you hope to avoid making statistical mistakes with your data.
Saving $500 on a deal sounds good. Saving $500 on a $1000 purchase is fantastic, whereas saving $500 on a multi-million dollar purchase is barely noticeable.
Improving the lives of 50 people with your organization’s efforts is cool. If your project only involves 60 people, you’ve got a winning formula on your hands. If your project is a national push… well, helping those people is still nice, but you need to consider if you’re spending your resources wisely.
See how much more interesting the denominators make the story? Numerators are important, of course — but the denominators provide the context you need to truly understand the trend you’re seeing or the story you’re going to tell.
Let’s Talk Denominators (and Numerators)
Are you struggling to understand the story your data is telling? Are you a journalist writing about a story that’s only got numerators? Let the skilled statisticians and data analysts at Datassist help tell your story. If you’d like to know more about how we can help you, check out some testimonials from our happy partners, or get in touch with us today.