KWFinder - find long tail keywords with low SEO difficulty

Modern Marketing Blog Influencer Series: Improving Marketing Performance with Analytics …

By Christopher Penn

“The Modern Marketing Influencer Blog Series asked top influencers from across the marketing spectrum what’s on their minds and what topics and pressing issues in their fields they feel are begging for more insight. Here they share their thoughts on data, the ever-changing field of marketing, and how it all comes together.

Analytics has become a common component of marketing strategy and planning. On any given day, thousands of marketing managers are asking questions like:

  • How is the social campaign performing?
  • What’s the analysis of the first-touch open rates?
  • What websites are our key demographic likely to use in the next month?

A growing number of stand-alone software solutions—as well as new capabilities within marketing automation platforms (MAPs) and customer-relationship management systems (CRMs)—are making it easier for marketing professionals to use analytics.

This is good news because marketers are overwhelmed by the amount of data they have, and the pace of new data is only going to accelerate. Proper analytics strategy and execution can turn that data into new insights about buyer behavior, marketing program performance, sales activity, and business impact.

Because of these new tools and general excitement about what’s possible with analytics, it might seem as if marketers are quickly maturing in this area. But the truth is, marketing remains in its analytics infancy.

Analytics Possibilities and Roadblocks

I am excited about what analytics could do for marketing. Anyone who reads my blog, newsletter, or data experiments like the Cheese of the Week forecast or has seen me speak at industry events knows this.

One example of what’s possible with analytics is intelligent time-series forecasting. With this, marketing teams would know when content about specific products and services is most likely to be found by buyers via search. The Cheese of the Week forecast is a silly but functional example of this: It uses five years of search data and proven algorithms to forecast when cheese enthusiasts will be searching for a particular type of cheese over the next 52 weeks.

If we were building a 52-week editorial calendar for cheese-related content, this insight certainly would improve the chances of our content leading to a cheese purchase. Content development, placement, timing, amplification—everything we do—would all be more effective with this insight. Obviously, you’d use this strategy and technology with your own data; this is merely an example of what’s possible and in market now.

Unfortunately, intelligent forecast is not common in marketing. Why not? Primarily because it requires true machine learning, and marketing professionals generally are not machine learning experts. If there is not a shared resource to tap, they don’t know where to start and can’t clearly articulate the ROI of investing in machine learning to internal stakeholders. Machine learning has enormous potential to help marketers be more successful, but what executive is going to take a risk with something that is outside of the traditional marketing-skills wheelhouse?

This is one of the major roadblocks in analytics maturation in marketing: lack of education and training about groundbreaking advances in technology, including analytics. On average, 3.9% of a marketing team’s budget is devoted to training and education, according to the August 2018 CMO Survey, a survey of 2,895 marketing executives that is conducted twice a year. In the same survey taken in February of 2014, the average was 3.4%—so a virtually flat 4-year trend yet consider how data and technology has changed marketing during those 4 years.

The other major roadblock is a lack of integration of separated data sets and digital workflows. While it can be beneficial to analyze data at specific points within the buyer’s journey or a marketing stack or program, the highest value of analytics comes from more sophisticated models that can be predictive and prescriptive. That is, they tell you not just when buyers will be searching for what, but also alert you to change your marketing mix in response to changes in sales. Conducting this level of analysis requires real-time integration of marketing and e-commerce data.

The impact of these two roadblocks is evident in another statistic from the CMO survey: Marketing leaders use marketing analytics for decision making only 35.8% of the time—that’s only about 1 in 3 decisions. (It’s worth noting that this average has actually fallen from a high of 42.1% in February of 2018.)

Going Forward and Getting Better

Clearly, marketing is on the right path with analytics, but as my observations and the CMO Survey data demonstrate, the steps have been small, and so the relative impact is small. That doesn’t mean doing more isn’t possible. It is, but it takes commitment to invest in education and training, as well as technology planning that elevates analytics capabilities.

Fortunately, marketing technology is evolving quickly, and that should accelerate advancement. For example, developer applications, such as Oracle Advanced Analytics 12c and Oracle R Enterprise, now make it easier to build and deploy machine learning in enterprise applications, including MAPs and CRMs. This means marketing teams with access to developers should have an easier time building predictive analytics applications.

Vendors also are starting to build in integrations, such as the integration of DemandBase’s context-aware content recommendations in Oracle Marketing Cloud. This is a trend that is likely to continue, and as it does, it will eliminate the need to manually integrate data and build custom applications, which is the main choice available to many marketers right now.

Another consideration is all of the untapped territory for “analytics solutions,” i.e., long-standing marketing challenges begging for a fix. Think about marketing and sales attribution. As more data sets are merged for analysis—operational, sales, audience, etc.—more insight on the impact of individual actions will be gleaned, and it will be easier to understand where to assign attribution. When this happens, it will be an enormous benefit for marketing leaders because pressure to demonstrate the top-line value of marketing is relentless. Right now, due to the complexity of the computations, true attribution analysis is the purview of only a select few data scientists, typically at larger companies.

The big takeaway here is that although marketing is still comparatively young in analytics maturity, it’s certain that analytics will be a bigger part of marketing in the future. To be ahead of that trend and gain the most possible benefit, marketing leaders will need to commit to learning more about analytics and choosing technology that can easily evolve to match the current analytics best practices.

See data in action as it makes a difference in how a company helps its customers with “eharmony Uses AI to Help Their Users Find Love.”

Read the blog