This past Wednesday, the Nina Hale, Inc. analytics department went to FARCon, a conference hosted by MinneAnalytics. It was a day jam-packed with seminars about everything analytics, specifically focusing on the financial and retail industries. Though there were lessons aplenty from this highly informative conference, here is a list that the department thought were the most useful.
1. Data is an amazing resource to help inform business decisions, but it doesn’t always explain the whole story.
In the past decade, businesses have started to realize the amazing potential of data, and how much it can help guide important decisions. Despite how powerful it is, it will never be able to explain everything that is happening. One speaker gave us an example of an electronics retailer in Detroit that was consistently performing drastically below what the data was saying it would.
The people working at the corporate campus (not in Detroit) couldn’t understand what was happening. Why was this particular branch performing so much worse than similar branches, including other branches in Detroit? They finally sent someone out to see if they could find an explanation, and they found out that the major highway near the branch had undergone radical changes, and the customer would have needed to drive 15 miles out of the way to reach the branch. This is information that obviously has a massive impact on a retail location, but is impossible to see in the data. So remember: data can provide amazing insights and predictions, but it will never be able to explain some crucial elements of business.
2. Have a strong, transparent relationship between analysts and the business decision makers.
A reoccurring theme at FARCon was that both analysts and the decision makers of a business need to work on their communication with one other. No one thought that the relationship between the two groups was stereotypically negative, just not as strong as it could be. One talk I went to, titled “Aristotle on Analytics” given by Scott Friesen (Principal, Quadratic), urged the audience to be prepared to explain as much of the math behind the analytics process that the decision makers find necessary to make the appropriate move. He then addressed the business decision makers, saying they shouldn’t just accept the data that is given to them, and assume it is perfect. Though most of the time it will be, it is never a bad idea to have another set of eyes on the process, to check it and maybe think of a different approach that could be taken to look at or use the data in a different way.
We were also given tips on how analysts should present the data. Though it’s important for the business decision makers to understand the process of the data, they don’t necessarily need to know how you calculated every single metric — or even need to hear the results of every metric. When presenting your findings, we were advised to always try and do three things. First, always give the results of the most important metrics up front, typically conversions or leads. Though it’s important to talk about the factors of what drove conversions and what non-conversion metrics helped with the bottom line, you need to present that the results the decision makers care about first so they remain interested in the supporting metrics. Second, control the story. Always relay back to the business’ goals, and don’t go off on a tangent. This goes for both the analysts and decision makers. While it may be interesting to explore a specific metric or group of metrics more in depth, you need to take a moment to reflect and ask yourself if it would help the bottom line. It’s surprisingly easy to start digging deeper into something that starts out beneficial, but before you know it you’ll be spiraling down a rabbit hole. Finally, similar to the previous tip, focus on the value. Be sure to let the decision makers know your value, and how you can improve. Even if you’re presenting a report with low performance, the decision makers will see you as a valuable asset if you explain what you did, and how you can improve your business and yourself.
3. Learn from every stage of the customer’s journey.
Though it is important to look at bottom-line metrics, like conversions, you need to be sure that you understand every step of the consumer’s journey and where the process may be improved. Often times you can explain why your leads had a drastic dip or lift by just looking at the individual steps that the user takes, and assigning metrics to each interaction to measure the performance. (Though not always, see tip #1). When you are measuring every step, you will be able to more quickly see exactly where you need to improve.
We were also advised that if for some reason you can’t find the data you need, don’t give up. Instead, go out and get the data. Not all the information you need will be able to be found by web analytics tools like Google Analytics or Tag Manager. Some data is more opinion based and need to be collected by the customer responding, not just watching their interactions. We were given an example of a company that was performing poorly, and they were unable to track the data they needed to answer why. Rather than guessing how to improve where they were lacking, they went out with surveys and collected all the data they needed. By hearing exactly how the users were feeling, they knew exactly what they needed to change, and almost immediately were back on track for the projected performance.
The information in this list is truly just the tip of the iceberg. If you’d like to learn more about how to enhance your analytics process and practices, I urge you to check out MinneAnalytics’ next conference.