Digital Analytics Strategy
Two Key Takeaways from eMetrics & GAUGE San Francisco 2012
First, let me say that the eMetrics and GAUGE conferences in San Francisco were incredible. There were great keynotes, informative sessions, actionable round-table discussions and intriguing conversations with peers in the industry. I would highly recommend anyone who analyzes data to attend eMetrics. And if your job allows you to touch Google Analytics at any point, GAUGE will certainly teach you something new that you can bring back and improve performance and increase insights.
That being said, I think there were two key takeaways from this past week:
1. Voice Of Customer (or VOC) is HUGE.
2. Methods in which you show data can make or break any report.
Getting Actual Voice Of Customer Feedback
Everyone from the heavy hitters online (Expedia, Intuit, etc) to the little guys are gleaning important data from Voice of Customer (VOC) applications. This allows the user to actually tell you about their experience rather than inferring it from the data. They point out specific problems, such as inaccurate or changing pricing to difficult-to-find information.
Tools (and vendor sponsors) like ForeSee or 4Q are easy to set up with a single tag and VOC data immediately comes pouring in. 4Q has the additional ability to integrate data with Google Analytics. You can then dig deep into any secondary dimension that you might usually use for great insight.
Active Feedback from Non-Customers
Outside of on-site tools, there are vendors such as UserTesting that allow a user to attempt whatever you ask them to while their screen and voice narrating the experience are recorded. You literally hear the voice of the customer attempt to use your site. This method is fairly inexpensive and easy to start. You won’t get a huge sample size (unless you have deep pockets), but you can get great insight from actual users within minutes.
Overall, looking past the quantitative and into the qualitative can provide priceless insight.
Telling A Story With Data
Another subject that came up several times throughout the conference was not just presenting data, but how it is presented. It should tell a story and keep the audience in mind. One keynote was from Steve Petitpas from Microsoft, who has experience presenting reports to Steve Ballmer directly. He discussed what data to show, explained actions that can be taken, and how you need to convey a lot of information in a short time frame. Knowing what to present is key.
Steve broke it down into the “Four That Score”:
- Revenue and Market Share
- Efficiency and Cost
- Customer Loyalty
- Talent and Capability
Revving Up Your Dashboards
Telling this story is essential for reporting, but the situation doesn’t always call for that. More often than not, you’ll need a quick look at a dashboard. For that situation, here are tips from dashboard master, Stephan Few:
- Provide a high level overview, but allow the user to drill into specifics easily.
- Data is easily and quickly absorbed with images. Text is absorbed linearly, but visuals are absorbed simultaneously. Also ensure the visual is congruent with the data, such as comparing data with a bar chart instead of a pie chart. The literal representation of a dashboard dial – like RPM or speedometer – are meaningless, and better expressed through other visuals.
- Color is a great way to get information at a glance, but the standard red/yellow/green is not ideal. Gradients of colors is best for conveying intensity and works best for colorblind users.
- Keep it all within one screen. A user shouldn’t have to scroll a dashboard.
- Provide context to the data – benchmarks, previous date range comparisons, projections, predicted impact, etc.
- Dashboards are meant for action, not simply “FYI” information.
The GA experts at GAUGE showed great ways to create dashboards for specific departments and segment them by Acquisition, Engagement and Outcomes.
Putting It All Together
From over 40 hours of sessions and 100+ pages of notes, these two key areas were touched on the most with real, actionable outcomes. Analysts from all experience levels can take advantage of these tips to better understand their data, role and responsibilities. Remember to ask your customers to tell you about their experiences directly, and watch them improve immediately. And then craft a story around that data for easy-to-digest dashboards and a happy c-suite.