Holiday Toys for Children's Hospital A.I. DuPont: The Spirit of Giving – Learn More ›
Popular Today in Tech: All Popular Articles

Predictive Analytics: Dare To Try It At Home?

Tech & Gadgets

Since the New York Times’ February 2012 article about “the intersection of data and human behavior”, How Companies Learn Your Secrets, I have had many conversations about what Predictive Analytics means for marketers.

One key factor is the change in how we consume and share information about ourselves. The accuracy of our message to our audience is infinitely more important – quality trumps quantity.

I have always been interested in pop culture and trends. I almost chose an anthropology major in college. In hindsight, that might not have been such a bad idea. I joined the ranks of IT marketers in the mid-90s.

During the .com boom, I learned a lot about new markets – both start-ups as well as the e-commerce counterparts to brick-and-mortar stores. Over the years, IT firms have brought more discipline to marketing. We have progressed in our measurement of campaign results, integration of on-line tactics, and use of increasingly sophisticated B2B marketing benchmarks. I believe Predictive Analysis, the intersection of data and human behavior, is an opportunity to take our practice to the next level.

Predictive Analytics: Dare To Try It At Home? image 274637 l srgb s gl 1024x697

Related Resources from B2C
» Free Webcast: How to Create Killer Email Conversion Copy

Predictive Analysis delivers greater business value to the business. It also gives us the chance to utilize new skills. Whether you like the right brain or left brain sides of marketing, targeting and delivering the optimal message requires both.

The tools used by retailers and the 2012 Moneyball Presidential campaign are enviable. I began to ask myself how to put this into practice. I have the feeling I am ignoring the disclaimer before Evil Knievel jumps and similar stunts, “don’t try this at home.” How do we try this at home? We gathered the tools available and took these 4 steps.

Step 1 – Historical Performance

We started with this statistical test that our database marketing team ran on campaign responders. It is no secret that the core audience for many IT campaigns is IT buyers. It is also no surprise that the quantity of messages this audience receives is enough to make anyone an avid “opt-outer.” Because the IT line of business in our target market is the only viable audience for our teams solutions, we needed to get smarter about how we were marketing to them. A .05% response rate to an email would not make a dent in our company’s sales goals.

Predictive Analytics: Dare To Try It At Home? image Graphic StacyRies

Step 2 – What does the data tell us?

Based on lead conversion analysis, an ordering of IT job titles can be associated with solutions offered.

  • Down each column, the ‘best’ titles can be seen associated with each solution
  • Across each row, each title can be associated with a rank ordering of solutions

The team’s analysis showed us two key points:

  • ‘Strategic’ titles such as CIO and VP are associated with more strategic decisions (P1) – Solution 2
  • ‘Tactical’ titles such as Director and DBA are associated more with production solutions (P1) – Solutions 1, 3, 4

Step 3 – Use the Trends to Inform the Campaign Strategy

From the statistical analysis, it was clear that a strategic message would produce better results with the CIO/VP audience and a tactical message would resonate more with Director/Mid-level IT Managers. I know this seems like a “no-duh” observation but the practical matter of getting campaign messaging that does not suffer from “feature bloat”, as software product managers like to call it, is no small feat. We were now armed with the evidence we needed to hone our message to the key point.

We filtered the email for the CIO/VP to focus entirely on the TCO value of the solution, with one whitepaper offered. We tailored the Director/Mid-level IT Manager message to focus entirely on the reliability of the solution, with one brief offered. The results were 12x greater – 6% vs. .05% of contacts responded to our email offer.

Step 4 – What’s Next?

We are happy with the results and we plan to make this persona strategy a standard campaign process. But this feels like we’re at the crawl stage of using Predictive Analysis in marketing and sales. We’ll continue to share what we learn, both successes and failures, as we move from crawl to hopefully walk and run stages. I invite you to follow this blog, share your thoughts and experiences, and do try this at home!

Comments on this Article: 0

Add a Comment

Add a Comment:


Thank you for adding to the conversation!

Our comments are moderated. Your comment may not appear immediately.