How to Approach an Analysis

Whether I am in the office or at a conference, I have noticed that the phrase “I think we need to do an analysis,” invokes absolute fear in many people. Sometimes it’s the fear of not knowing analysis techniques, sometimes it’s the fear of what the data will reveal and often it is because they just don’t know where to start.

The mindset for solving analytical problems is outlined in the following steps:

1. Determine the list of questions you are truly trying to answer.
2. Check the data.
3. Determine which analysis methods should be used.
4. Conduct the analysis and interpret the information.
5. Tell the story.

Determine the List of Questions You Are Truly Trying to Answer

Don’t try to answer everything about everything. The simpler you can keep the problem definition, the better chance you have of remaining focused on the goals that are of value to the business.

Check the Data

First, something must have originally triggered the impression that an analysis was needed. Find out what this was and make sure that the data behind it is accurate. Very few people perform this check and nothing is more embarrassing than spending valuable resources on something unnecessary … except having to explain that to the company’s executives!

The second area of data to check is the raw data that you will use for analysis. This is where the old saying of “garbage in; garbage out” comes into play. For large data sets, the best approach is to start by performing a graphical analysis.

A graphical analysis is nothing more than several graphical views into the data at a high level in order to see whether anything looks odd before you dive into more sophisticated analyses.

Determine Which Analysis Methods Should Be Used

Thousands of pages could be dedicated to the analysis methods available to the reader, but the important guideline is to keep it as simple as possible. Higher end methods are really cool to math geeks like me, but they are of no use if you can’t explain what it means to the executives and their business.

Conduct the Analysis and Interpret the Information

I have paired these two tasks for a very important reason. I have met some people who can “run the numbers” but have little chance of figuring out what the analysis is trying to tell them. The value to the business is not in the numbers; it’s in the message behind the numbers.

Tell the Story

There is one final step in this process and that is to figure out how to convey the message of the analysis. The interpretation of the data has to be presented in a format that the audience will understand. This is known as “telling the story.”

The ability to tell the story in a clear and concise fashion is a skill that few people possess. Most analyses take a fair amount of work to complete but fall short in conveying the results to the audience.

For me, the audience has been as high as the Board of Directors and the company founder. From what I have experienced, the higher up you have to present your analysis, the less time you have to convey your message. This is where my previous recommendation of being clear and concise really comes into play. At these levels, you have 10-15 minutes to convey your message!

In summary, it is important to have an accurate analysis. Beyond the analysis, it is important to take the time to “tell the story” so that your audience can understand it… and for goodness sake, make sure your story actually answers the original business questions!

Taken from Tracey Smith’s book “Data Driven Decision Making for Small Businesses: Unleashing the Power of Information to Drive Business Growth.” Her book will show business owners the practical uses of data analysis and how it can drive value to the bottom line. Those interested in being on her distribution list for book updates can email publications@numericalinsights.com

Copyright © 2012 Tracey Smith