The Effective Marketer

If before marketers had the challenging job of gathering data to analyze campaign effectiveness, now the opposite might be true. Email marketing, marketing automation, web analytics, CRM, and the myriad of software now permeating marketing organizations gives us more data than we can process in a timely fashion.

A recent survey by CMO.com shows that fewer than 20% of marketing respondents have full confidence in what should be fundamental abilities, including measuring overall campaign effectiveness, how to allocate budget with ROI in mind, and communicating performance up to c-level executives.

“The lack of confidence results from a perception that there is simply too much data and too many channels out there, making it difficult to capture and measure all relevant data.” – CMO.com

Understanding Marketing Analytics

A recent post on the Marketing Automation Software Guide Blog titled “Marketing Analytics vs. Website Analytics”  does a good job at separating two commonly mistaken data sets. On one hand you have page views, click paths, bounce rates, and all the nice stats Google Analytics gives you for free. But, as the article argues, “in marketing analytics systems, data is integrated in a way that enables you to see a direct relationship between channels“. And this is key to understanding how to measure marketing effectiveness.

Unless you can step back from the data deluge, it will be difficult to assess exactly what to do.

“There are literally hundreds of marketing metrics to choose from, and almost all of them measure something of value. The problem is that most of them relate very little to the metrics that concern a CFO, CEO and board member.”

Another consideration involves who you are reporting to. When analyzing results from your marketing efforts you have different stakeholders asking different kinds of questions. The quote above is from a Marketo whitepaper, The Definitive Guide to Marketing Metrics and Analytics, which I have reviewed before. And it is spot on. Different questions require different data sets.

So your first question shouldn’t be “what do I measure?” but, instead, “what question am I answering?“. Do you agree?