Digital analytics is booming, but it’s still so new and everchanging that there aren’t any textbooks yet. Sure, there are online certifications like Google Analytics, but those are really specific to a channel. And those keep changing too! After teaching a course on digital analytics and interacting with students every day, here’s what I have learned:
You don’t need to have technical skills to be an excellent analyst. You need to be curious. Ask questions. Be unafraid to shake the boat. Oftentimes, the big insight isn’t in the raw data – it’s in the patterns.
I’ll walk you through an example.
After running a digital campaign with interactive banner ads (the user could draw on them), we had a standard summary report which included impressions, placements, clicks, actions, interactivity with the units, time of day, and all the normal stats that come with a summary report. But then we were curious – did the interactive units really move the needle? Would it have been better to have just run static ads with more impressions? Here’s the data:
Rich Media additional stats:
- Display time – 322,395,407 seconds (613.39 years)
- Interaction time – 970,907 seconds (1.75 years)
- Average interaction time – 5.0 seconds
- Interaction rate – 3.6%
- Total interactions – 184,702
If we would have just looked at CTR, then the static images were far better (0.23% compared to .03%). If we were to just look at the submission rate, then the rich media units barely beat out the static ones (.236% compared to .231%). But then we started to weigh interactions. On a static banner, one can just see it and click. On a rich media unit, one can interact with it. The rich media units had a 3.6% interaction rate, with each user spending five seconds in the unit, spending a cumulative time of 920,907 seconds (1.75 years) interacting with the units.
So to take it the next step further, does 1.75 years of interactivity outweigh a lower click through rate? Does it justify a higher ad serving cost? We did the math and found the cost-per-click to be 9x higher with the rich media units. However, if you looked at cost-per-action (interaction or click) the cost was actually 2x lower with the rich media units. The end goal was awareness and submissions, and the rich media units played a vital part in the success of the campaign.
Lastly, if you’re asking yourself why the conversion rate was so low – first please apply for a job at Penna Powers, but then rest assured that we dove into this as well. We looked at the conversion funnel and identified an area where most users were dropping off and were able to correct it.
Obviously every campaign is different. If one learned to always look at CTR, then one would miss the impact of the rich media units.
Curiosity helps in the approach. None of these calculations required more math than division or multiplication.
When teaching my students how to approach problems analytically, we start with creativity exercises, such as writing down 20 uses for a thumbtack. We then do a mountain of case studies. Even though every problem is different in the real world, the case studies help students become familiar looking at patterns. Lastly, we just do a lot of analysis on real companies. Again, every analytical problem is a little different, but for example it helps to see a few ecommerce problems along the way.
Teaching students entering the workforce has so many parallels to our self-learning at work. I’m amazed at how few professionals develop the analytical skills necessary to excel in their jobs. Whenever someone tells me that they’re “not a numbers person” I just think to myself that few people are. We’re humans and all like pretty pictures over an Excel spreadsheet. But we all are curious. So take an hour out of your day and go hunting for some patterns. You might be surprised what insights you find.