New Relic’s marketing data analyst Maria Stopak is a fan of data visualization guru Edward Tufte. “Tufte teaches that there are two key goals whenever you are presenting data: convey the story you want to tell and make sure you’ve established credibility,” she explains. As Maria has learned firsthand, it’s not always an easy feat, but it’s certainly an exciting challenge.
A member of New Relic’s Marketing Operations team (lovingly referred to here as the MOPsters), Maria is responsible for all marketing-related reporting and analysis, helping us understand everything from the effectiveness of our campaigns to broader questions like how many customers we have paying for a particular product. “I segment and analyze the leads that marketing sends to our sales team to uncover greater business insights,” said Maria, who does so using a unique blend of SQL skills and software.
No longer confined to the rows and columns of Microsoft Excel, marketing analysts like Maria are enjoying greater flexibility in terms of how they can query data and what types of insights they can uncover. We recently sat down with Maria to talk about trends in the modern MOPs landscape, the beauty of data visualizations, and the potential traps data analysts should try to avoid.
What’s in a modern MOPster’s toolkit?
A former product analyst with a B.S. and M.S. in mathematics, Maria has used a variety of reporting and analysis tools throughout her career. At New Relic, however, this “mathe-magician” relies on:
- Heroku Connect. A tool that replicates Salesforce data models inside the SQL database-as-a-service Heroku Postgres, which she likes for its breadth of reporting capabilities.
- R. A free, open source statistical computing tool that lets you manipulate, massage, and model data in endless ways.
- Shiny. A Web application framework for R that lets you turn your data analyses into interactive Web applications.
“R and Shiny are very widely used by analysts in all types of job functions, but Heroku Connect is a newer service that represents a dramatic shift in the analytics mind-set among marketing and sales analysts,” said Maria. “Seeing that the new fad seems to be automation and writing software programs that can be used for marketing analysis, I anticipate a lot of new users coming in who’ll find it very useful.”
(Interested in learning more about Heroku Connect and how the New Relic Marketing team uses it? Maria will be participating in the Heroku Connect panel discussion at Salesforce.com’s Dreamforce conference on Wednesday, September 16 at 9:30 a.m., so be sure to stop by!)
Data visualizations to inform and delight
One of the most common questions that comes up both here in Marketing as well as in other parts of the business is, How many people are paying for all or some of our products?
To answer that question, Maria created the below Venn diagram (an animated gif in its original form) that illustrates the number of paying customers we have per product and how those numbers changes over time. “Data visualizations can be very powerful, since you can tell a deeper story more efficiently when you have the right graphic,” explained Maria. “This diagram in particular is one of the many things I’m known for in the company. It gets looked at by Marketing, the Exec staff, the Product Management team, and more.”
Another big part of Maria’s job is analyzing customer activity throughout each stage of the purchasing cycle. “I’m looking at lead volume, conversion rate, and the velocity at which leads are moving through the stages,” said Maria. “Ultimately, what we’re trying to do is create the smoothest, most streamlined process we can.”
To that end, Maria created the following Sankey diagram that shows the flow of user activity from one step to the next. The purpose of this visualization is to show all the various paths that people take and which products get picked up the most along the way.
A different way to think about campaign analysis
Although Salesforce.com has its own built-in reporting capability, the New Relic Marketing Ops team relies on its own custom campaign analysis to avoid any misleading interpretations. “I’d say it’s a much more direct approach in that we don’t include what Salesforce calls ‘influenced opportunities,’” said Maria. “I prefer to focus on a specific person versus a group of people at the company because it keeps our reporting accurate. This way, we can pull apart the story of what really drove the customer win.”
What story is your data telling you?
What it all comes down to is sticking to the truth. As Maria said, “Tufte stresses that ‘cherry-picked’ data is the biggest threat to your presentation’s credibility.”
Maria’s own work philosophy echoes this sentiment: Your story shouldn’t shape the data; the data should shape your story. “Some people try to use data to prove a point or tell a specific story they already have in mind, but that’s not the right way to think about it. It’s important to let go of assumptions and listen to what the data has to say. In the end, we’re all working toward the same goal, which is to see the team succeed, and I believe in trusting the stories that the data tells us.”
Think you have what it takes to be a modern MOPster? We’re hiring! Get the details here.
To learn more about New Relic MOPs, read:
- Tech’s Newest Marketing Skill: Engineering—A Future Talk with Isaac Wyatt and Baxter Denney
- What Marketing Ops Can Learn From DevOps
- Using Insights with Marketo Webhooks to Build Real-time Dashboards and Analytics