Effective leaders often credit their team members for successes and achiev...Read More
Social Media and Investment Advisers
Drop in the Bucket
From Vision to Action
The Right Move
Needle in a Haystack
Corner the Market
By the Books
"Never Eat Alone," by Keith Ferrazzi
Are you a credible leader?
One for the Books
A Capital Idea
Travel & Tourism
Davis and Weber Counties
Management guru Peter Drucker once said, “What gets measured gets managed.” Or something like that. As with most notable quotes, Drucker’s exact words have likely been twisted to fit this or that context, but in any event, his meaning is clear: good information informs decisions and sharpens performance.
In today’s digital landscape, web analytics can provide the most concrete and granular marketing data in the history of business. It can also bury decision makers in mounds of sensory overload.
Experts in the rising realm of online data are developing a set of benchmarks to guide managers and executives in their marketing efforts and help maximize the potential of powerful analytics software packages.
Deciphering the Data
“Analytics give you the opportunity to measure how effective your marketing really is in a much more precise way than in the past,” says David Cree, technology manager at the Utah-based marketing firm Rare Method.
With traditional media, marketers have relied on projections of how many people read a newspaper or watch a TV station, with proof of performance limited to anecdotal evidence in most cases. Analytics, on the other hand, allow precision in gathering and interpreting data. Digital measures aren’t turnkey, though. Firms have to invest not only in software, but in people to make it work.“Most people just install some code and watch their website traffic,” Cree says. “That really tells you nothing.”
The first step, then, is to designate someone to nurture a firm understanding of the vast and varied capabilities of analytics, which cover everything from tracking page views to detailed segmentation of customers, and integrate the data into key company decisions.
Most software programs are intuitive enough for just about anyone to watch the tide of visitors ebb and flow, but the learning curve can be a little steeper when it comes to tracking data across web platforms and identifying the variables that actually affect the business, so the point person ought to be tech-savvy while possessing enough business sense to point the data at what really matters to the firm.
“Until you can actually act on the data, the insights you get don’t really mean much,” says Brent Dykes, evangelist for analytics products at Adobe and author of Web Analytics Action Hero.
Web data can be like competitive cycling or running: the challenges and payoffs get bigger the deeper you get. There are plenty of thoughts about how to categorize analytics, but one way is to think of four levels: basic, exploratory, multichannel and predictive.
Basic data consist of simply plugging tags into web code to collect information (or, in more aggressive instances, initiating efforts in data mining or web scraping). This is a technical component, usually reserved for whoever manages the website, and is the first step in collecting data. This level stops short of even basic analysis and is relegated to installing the infrastructure required to gather data.
Exploratory data dives a little deeper and seeks to tell a story with that basic data. Where are these visitors coming from, and what does their activity indicate about motives for visiting? Most companies’ use of analytics ends here, mainly because the data are easy to get at and interpret.
“You can learn some valuable things from exploratory data, but you are limited in a lot of ways,” Dykes says. “It is much more useful to dive deeper and get a 360-degree view of your customers.”
The real power of analytics is manifest in the deeper levels. With multichannel data, marketers can combine online and offline data to generate a more complete picture of customers. This allows managers to optimize customer relationship management (CRM) programs and identify patterns in what turns passive grazers into paying customers. This is no cakewalk—the price for such sophistication is dedicating a team, resources and attention to the cause.
While most firms fail to get beyond exploratory data, even fewer take the final step into predictive analytics. At this level, the analytics infrastructure includes pieces like “machine learning,” which is a series of algorithms that allow the software to recognize and adapt to patterns and make prognostications about how visitors might behave.
Like most tools, analytics are only as powerful as the skill and attention they are given by their users. Even the artificial intelligence of machine learning is pretty useless without a company culture of being data-driven.
“It sounds obvious, but it’s difficult to really institutionalize a data culture,” Dykes says. “It’s about being disciplined and using data on a consistent basis.”