Performance marketing insights, discussions and expertise

5 Mantras for a Measurement-Driven Organization

By on March 29, 2016

insider-spotlight

The days of Mad Men are gone.  The stereotype of a successful marketer has changed, and unfortunately we no longer have time for those 2 martini lunches. Marketing is becoming as much science as it is art.

I’m a big fan of mantras. I think they go a long way when it comes to cultural transformations, if you repeat them often and make them part of your regular dialogue.

I also believe less is more, so I chose to focus on the 5 most important mantras when it comes to bringing focus in your organization on becoming measurement-driven.

 

It’s good to be wrong!

Nope, that’s not a typo. I love it when the data proves me wrong, because that’s when the value of the data becomes really evident. Without the data, I may have gone down another path, and it’s easy to calculate the opportunity cost of not having that insight. Celebrate your mistakes and the fact that the data you collect minimizes their impact. Share your data – it shouldn’t be a secret weapon!

 

Keep it simple!

Albert Einstein once said “If you can’t explain it simply, you don’t understand it well enough.” Big data makes it so easy to drown in statistical significance stats and complex calculations. Be sure you always boil your findings down to actionable recommendations. Use visualizations whenever you can; the worst thing you can do is project your spreadsheet and your mathematical modeling (it’s worse than death by Powerpoint.) Start with why you care, and what you can do with the information – don’t lead with the method to your madness.

 

Just the facts, please!

This one is really hard for me. It’s hard to set your opinion aside, but I’ve found that being objective and sharing a perspective based on facts both quantifiable and measurable allows the data to do the talking. Beware! The data must be credible, so be prepared to share any assumptions behind your calculations. Once you lose the trust of your colleagues, it’s hard to get it back.

 

Always Challenge Assumptions!

A good friend of mine introduced me to the “FOG filter” years ago. I love it and I use it all the time – at work, at home, even over cocktails with friends. The FOG filter gives everyone in your organization permission to challenge assumptions.  When a statement is made – no matter who makes it – one can ask if it is “Fact, Opinion, or Guess?” Facts are based on empirical data, opinions are interpretations sometimes based on data, and guesses are basically hunches or unanswered questions. Separating the facts from the opinions and guesses can really bring focus to your organization, cut down on those heated debates, and give the team some really good things to test!

 

Practice What You Preach!

I’m not going to get religious here, but if you really want a data driven organization you can’t be selective about when you listen to the data, and when you don’t. I have worked at organizations that were data driven, and those that claimed they were, but really weren’t. The biggest difference between the two is that a data driven organization welcomes challenges and experimentation. A data driven organization is all about failing fast, and is not risk adverse. Data is a security blanket, and shines a light on possibilities – it is not feared.

 

If you want to build a data driven organization, these mantras are a great place to start.  But it’s not all about culture. Start by taking a look at the technology you are using.  Do you have a single source of the truth when it comes to marketing data?

Here at Impact Radius we absolutely love data. In fact, we have 1.5 petabytes of storage in our Hadoop cluster. According to ComputerWeekly.com, one petabyte is enough to store the DNA of the entire population of the US – and then clone them – twice! Like I said, we love data!

 

1 thought on “5 Mantras for a Measurement-Driven Organization”

Leave a Reply to ErikHDavis Cancel reply

Your email address will not be published. Required fields are marked *