The Modeling Agency logo

Analytics Transformed™
  • This field is for validation purposes and should be left unchanged.

Trick Question: Which Data Points Should Your Organization Be Tracking?

BoromirDataMiningThis is a question which came up during the January 16th session of TMA’s free webinar: Data Mining, Failure to Launch. One of the participants asked data mining gurus Tony and Scott to provide him with a list of data points that his organization should be tracking.

Of course–that couldn’t be done, not in the way the participant was probably hoping for. After all, one simply cannot generalize that kind of list.

Such a list could only be generated after taking a hard look at the project environment, aligning the objectives of the organization’s entire project teams, and exploring the data that is already available.

Nobody can look at your organization and say, “Well, you should definitely be tracking transaction data,” or “You should definitely be zeroing in on all voided transactions.” There is no one-size-fits-all solution because it all relates back to what the data needs to accomplish.

Data mining actually flip-flops the traditional scientific method that every school child learns. And this is often a source of confusion for organizations who are struggling to do a good job handling their data.

In the traditional scientific method you form a hypothesis. Then, you gather data.

But most organizations have already been gathering more data than they know what to do with for a very long time. This is one reason why nobody can just provide organizations with a list called, “The Top 50 Data Points Your Organization Should Be Sure to Track Right Now.”

Chances are, you already have all the data you need, and then some. The challenge is to figure out which of those data points are actually relevant to your objective.

Data which is relevant to tracking fraud, for example, may be 100% irrelevant to data which helps your direct mail campaign become more effective.

Don’t be mesmerized by the data itself. Be laser-focused on problems and priorities.

Then, you can develop your own list–a different list. The list of the Top 10 problems that you’d like to solve or objectives that you’d like to meet.

Now there is a list which would actually be useful.

Would you like to get access to gurus Tony and Scott for your own short free consultation session? Register for the next session of Data Mining: Failure to Launch and bring your most pressing questions.


Data Mining Webinar

Learn How to Get Predictive Modeling
Off the Ground and Into Orbit
1 Hour Live Interactive Event

Why Train With TMA?

Determine whether TMA training is right for you, and learn why TMA is truly the best option for live classroom analytics training.