Confession of a Data Analyst at Triangle AMA Analytics Camp

by Phil Buckley on October 20, 2010

in Analytics

Post image for Confession of a Data Analyst at Triangle AMA Analytics Camp

After a leisurely 5 minute lunch break, we’re back at our seats for the “lunch and learn” session.

Ok, here we go. Michael Rappa, Founding Director at Institute for Advanced Analytics at NCSU is speaking softly over the crunch of potato chips. Not surprisingly he looks very professorial.

Starts by segmenting with questions about who is a marketer and who is an analytics geek.

Confessions of a Data Analyst is the title of his talk.

The world has changed over the last 30 years in many ways, but the amount of data available is one very simple way to measure it.

Who even knew about a terabyte, never mind a petabyte back in the old days (pre-2000).

Data Quality

Back then, people actively went out and collected data… ON PURPOSE. We also have the advantage of being exposed to both numeric and text data. When we collected specific data, we had more control over it, now-a-days we live in a torrent of data and can’t always understand all of the data without a significant amount of work.

Greater Urgency for Resuts

You no longer have weeks or months to analyze data. Often times, we now require data to be analyzed instantly.

Data Disconnect

There is a growing chasm between marketing and data analysis. Self deception is easy, and when you can use numbers, it’s even more persuasive.

Work to use data to show you something you don;t yet know or believe. That’s where the power comes from on the data analyst side.

Assumptions

In data analysis, you’re always making assumptions. Most assumptions are about relationships between data points. Assumptions are ok, but always be aware of what assumptions you’re making.

Analytical Complexity

Complexity does not always mean greater insight. The goal should always be towards simplicity. If you can simplify down to the level of a simple visual, that’s the best. Most humans can consume visual data much easier. Good analysts can always visualize their data.

Privacy

Privacy isn’t that important to people, unless you ask them. Sometimes you can know too much about someone. Just because you know exactly what they are at this moment, doesn’t mean you understand them as a person.

Humility

To be a data analyst is to be humble. You don’t want to be a cocky data analyst, because that’s when you start missing things.

Michael thinks that within 5 years, all marketers will need to be 1/2 data analysts as well if they are to thrive.

{ 1 comment… read it below or add one }

Morgan Siem October 23, 2010 at 9:29 pm

Two great points here – (1) that most humans can consumer visual data much easier, so good analysts can visualize their data and (2) that marketers need to become 1/2 data analysts if they are to thrive. I fully agree. Analyzing the data help you to understand the audience you’re marketing to and the impact of your efforts.

Reply

Leave a Comment

Want to add a picture to your comments here on Media Two Point {OH}? Upload your picture at Gravatar to make it happen.

Your comment may be nofollow free.

Previous post:

Next post: