Gartner Blog Network


The Art and Science of Data-Driven Marketing

by Bill Gassman  |  April 18, 2013  |  3 Comments

Little doubt should remain that data-driven marketing has broad potential.  Leading brands, like NY Times, Disney, Wal-Mart, Intuit and Geico prove its worth every day.  There is also a strongly collaborative community of digital measurement specialists (see Twitter #measure), committed to helping each other get more value from data and celebrate better conversion rates.  The weak link though is buy-in from the creative side of marketing, to be driven by data.

Gartner’s recently published survey on data-driven marketing shows most organizations understand the potential of analytics.  They allocate an average of 21% of their marketing budget towards analytics.  A survey summary is freely available at this link.

What the survey doesn’t show is how effectively organizations use the results of analysis and how efficiently they spend their analytics budget.  How much of that analytics budget is wasted?  Getting value from data is more difficult than sitting in the CMO captain’s chair and commanding; “make it so”.

The selection of tools available to the data-driven marketer is already nothing short of amazing.  You can tell which campaigns are attracting your audience, how people behave, create micro-segments and target an individual based on context.  But, what good is all of this if you don’t know what will persuade your users to take action, or are unwilling to expose creative ideas to the transparency of an A/B test?

Analytics tell you what is working and are pretty good at predicting what will work, but someone has to think up the possibilities in the first place.  New ideas takes creativity and analytics can help people that are creative do things better.  The best work gets done within the intersection of art and science.  Organizations that figure this out first have a marketing advantage over those that come around more slowly.

Where should you focus your data-driven efforts?  On the data, of course.

  • Hire talent that knows the science of analytics, but has passion for the artistic side of marketing.
  • Measure everything you can.  Build telemetry into the development process of online applications, content and campaigns.  Mash with externally purchased and internal data.
  • Embed analytics into every marketing decision where it makes sense.  Train the creative folks, so it becomes essential to their work.
  • Create an independent analytics omnibus to measure results from the CMO’s point of view.  Here, all individually optimized tasks come together into (ideally) harmony with business goals.

Bottom Line: The science behind data-driven marketing works, but works best when there is trust, not suspicion among the creative staff.

Points for discussion – what have you seen work, or not work, to enlarge the intersection of the art and science of digital marketing?  Do you agree it is important?

Category: data-driven-marketing  digital-analytics  digital-marketing  digital-marketing-programs  

Bill Gassman
Research Director
15 years at Gartner
35 years IT industry

Bill Gassman specializes in helping marketing leaders manage, measure and optimize their Web, mobile, social and search marketing programs. For more than 15 years, he has advised marketing teams on how to use technology ... Read Full Bio


Thoughts on The Art and Science of Data-Driven Marketing


  1. Ellie says:

    I agree that data is becoming more important in marketing because of the new channels now available to us, such as social media, email automation tools, etc. However, as a seasoned marketer and as someone who began her career as a Marketing Analyst when only data from databases was available, I still believe that to a certain extent data can be manipulated to tell the story you want it to tell. You can cut and analyze data multiple ways.
    I find and appreciate the amount of data available to us and realize to mine through this data is still a challenge. I am certain it will become easier to gather and analyze data in the next five years. But what is still a common belief and guidance in the marketing community is to look at the data but go with your gut feeling. I am also certain that this view point will change with the change of generations in the marketing roles as the newer generations are growing professionally with an enormous amount of data available to them.

  2. Bill Gassman says:

    Thanks for the comment Ellie. Hopefully, gut feel will never be discouraged, but neither should it be hidden from exposure to analysis. The $50K flashy home page splash picture may feel good to the graphic artist’s gut or the manager that hired out the job, but if it moves the needles strongly in the wrong direction, it should be called what it is; money wasted.

    I went to an agency’s web site yesterday, which had a pop-over on the home page, asking me to sign up for their newsletter. My gut says this is a terrible practice, but it turns out they get quite a few to sign up. What they may not know is; how many real prospects just bounce away. That’s what A/B testing is all about.

    To your point about data manipulation, I agree! Data can be manipulated to tell the story you want. This is why an omnibus approach is useful. It analyzes results independently from the point solution tools used within marketing. This is especially important as agencies provide analytics as part of their service. Data quality and manipulation problems will always be with us, but the newer techniques – such as tag management – are improving trust. That should help those that decide on gut feel make better decisions and feel good about them.

  3. Bill and Ellie –

    You are both on target. Data can easily be manipulated by a solid analyst who wants to tell a specific story – but the answer is not to ignore data. What challenges many marketers is that they aren’t asking the right question. For example, using analytics, we’ve seen that bounce rates are much higher for visitors interested in a “gift” – where gift or something like it was part of their expression of intent. Based on that analysis, you might decide not to bid or optimize for gifts. BUT – that same traffic had higher conversion rates and order sizes – because the intent was so high. MEaning that if the gift you wanted was there, you bought. If not, you bounced.

    The other challenge for ecommerce marketers is that the number of variables and the number of optimization changes you need to make every day is simply beyond the capabilities of the marketing or IT organizations. We (Bloomreach) believe this is where a generation of intent-aware applications, big data marketing applications, are going to empower marketers to create great content, leverage web-scale insight and analytics and adapt to act on those analytics in web speed.



Comments are closed

Comments or opinions expressed on this blog are those of the individual contributors only, and do not necessarily represent the views of Gartner, Inc. or its management. Readers may copy and redistribute blog postings on other blogs, or otherwise for private, non-commercial or journalistic purposes, with attribution to Gartner. This content may not be used for any other purposes in any other formats or media. The content on this blog is provided on an "as-is" basis. Gartner shall not be liable for any damages whatsoever arising out of the content or use of this blog.