(Featured illustration: Hanns-Peter Nagel)
Editor’s note: This post is the first of two posts that highlight careers in digital analytics. The second post in this series will highlight five characteristics of a great digital analyst and how to build skills that will help get you hired.
Who is coming to your website or social media channels? What are they doing? Are they converting into fans or customers? The answers to our most vital digital media questions lie in data, which is why we need to be not just conversant but fluent in analytics. Data not only help us to figure out what is happening but also why. Our best business decisions are driven by what we’ve learned.
As data get more complex, more organizations are adding dedicated digital analysts to their teams. As with many other digital media positions, this role comes in a wide variety of flavors. Some require deep statistical knowledge and expertise with SQL or SAS, while other positions are filled by experts in Google Analytics or similar tools. Job titles include Web Analyst, Marketing Analyst, Digital Media Analyst, Social Analyst, and other variations on the analyst theme.
Earlier this year, I spoke with Connie Rock, a student in the Communication Leadership (Comm Lead) graduate program at the University of Washington and Comm Lead alumna Navni Garg. Both Rock and Garg have solid analytics experience. Rock, currently a freelance writer and content curator at Artifact Technologies, previously worked in web analytics at Microsoft and social analytics at Deloitte Digital. Garg is a Senior Data Solutions Manager at Turn, which provides analytics solutions for advertisers.
I asked Rock and Garg about best practices that analysts—and anyone tasked with analytics—should adhere to, and here’s what they suggested.
Best Practice 1: Start with Your End Goals
Any strategy worth its salt starts with determining business objectives: What, exactly, do you want to accomplish? Only once goals are defined, key performance indicators (KPIs) can be determined that will be tracked, measured and analyzed. Because business goals are so important, analytics is not just an after-the fact reporting function, but a key strategic role.
Let’s say a client wants to increase brand awareness through a social media campaign. That’s a very different desired outcome than increasing direct sales, so the metrics that the analyst will need to implement and track will be different, too.
“If brand awareness is your goal, one of the things I would measure would be the volume of conversation about the brand in posts and tweets to see if your social efforts are increasing mentions,” says Rock. “In addition, I would measure engagement rates. How many people are sharing your posts or commenting? And what is the trend—are you getting more people to engage over time?”
Best Practice 2: Be Accurate
Simple metrics, such as page views or Facebook likes, can be a red flag when used by themselves without other supporting data. These metrics can be a sign that data have been cherry-picked to support a position, or, at the very least, that the data are too simplistic to be scientifically accurate. This is always problematic, but even more so when there’s a lot of money at stake, such as when advertisers are determining their ad buying mix.
“It is very important to be mindful of your impact and be accurate,” says Garg. She also warns about blindly applying general data that have not been tested to see if they are applicable to the client. “You may generally know that spending on advertising in a certain channel is high value, but you never know for sure until you test. Your overall data may be biased in a lot of ways.”
Best Practice 3: Tell a Simple but Compelling Story
Once the numbers are crunched, the analyst needs to metamorphose from data geek to storyteller. Analysts tell their stories in many different forms, including dashboards, data visualizations, infographics and client presentations.
Dashboards allow the analyst to group similar types of insights and can be interactive. Dashboards provide data points from many different angles, but don’t preclude anyone from picking one insight that resonates. Zeroing in on a data point is not necessarily the wrong thing to do, according to Garg. “But if you do that, it is important to test that the selected data are not biased or skewed.”
During presentations, avoid cluttering visuals with too much text and focus on what truly matters: The data that are connected to your business goals.
“Start with an executive summary, define the problem, and provide a simple answer to the most important questions before you provide more depth,” Garg recommends. “Break it down so that the people you are presenting to can quickly grasp the insights you are bringing to the table.”
Best Practice 4: Focus on the Next Step
Don’t fall in the trap of measuring a multitude of metrics and KPIs just because you can. “It’s all about getting from data to action,” Rock stresses. On the flip side, when the metrics you are tracking don’t lead to action, change them and find data that will rise above the noise level.
In many cases, clients take specific actions that will have a direct effect on their bottom lines. Garg, who helps advertisers figure out how to allocate budgets effectively, points out that her clients have a lot of options for spending their sizeable budgets—analytics isn’t just a nice-to-have. “For advertisers, return on actual dollars is very important.”
To figure out what works and what doesn’t, analysts must attribute conversions correctly, so that clients know to increase investment in proven channels and tactics.
Best Practice 5: Test, Test, Test
Metrics can also be used to predict performance through initial tests. “You need to check every decision you make. If you don’t test, you could go down the wrong path and not know it,” says Garg. “Allocating money towards research is really important. Constantly run small a/b tests and research projects.”