The Art and Science of Analysis, Part 1/
December 10, 2012
Last week my colleague Sayf Sharif posted another rant and playfully suggested it might be the second in a series named, “You’re Probably Doing It Wrong.” This week I’m starting a series of my own: “There Is No Right Way.”
There’s no perfect process, no one true path that guides you from reporting to analysis, from data to insights. Instead it takes both intuition and calculation, knowing when to let your curiosity follow tangents and when to let structured thinking rein it in. You have to use both your right and left brain in the art and science of analysis.
Ask Questions: Be Curious and Contrary
Create your own path to analysis by continually asking questions. Start with questions. End with questions. Be curious. What did your visitors do? How did the data change? Why did that happen? How can we make it better? What will you do next?
Avoid jumping to conclusions – fast thinking which can be overly influenced by personal biases. Slow down and challenge your assumptions. What haven’t you considered? What if you ask the question differently? I’m actually quoting a Buddhist physician here.
Once you think you have some answers, be contrary. Ask, “So What?” Can you attract better-qualified visitors? Can you give them a better experience? Can you improve your bottom line? If the answers don’t help you take action to make things better, keep asking.
Use Tools: Explore Possibilities and Seek Hard Evidence
Digital measurement tools are evolving faster than ever. Universal Analytics is just one example. You have a dashboard? Get ready to have a better one. You have favorite reports? Get ready to have favorite-r ones. The pace of innovation means more data to explore and discover.
Let your exploration of the data suggest new ideas and new questions. Add context to what you find. Your business has its own patterns and is subject to a unique set of conditions that affect your data. And there are external factors that affect everyone’s data. Use it all to tell a compelling story of your customers and your products or services.
At the same time, build a case for your ideas – back your story with hard evidence. Dig into the numbers. Look for significance. Love math.
Tools can limit you or they can be the springboard. It’s your choice: will you be a number-cruncher or analyst?
So What? It’s All Still a Bit Fuzzy
I’ll be writing about detailed examples of “doing analysis” in the upcoming posts of this series. Stay tuned for real-life questions and answers (which of course lead to more questions).
What’s your approach to analysis? Do you find some types of questions especially challenging and how do you tackle them? Please share in the comments.