• by Andrew Latzman , December 2, 2014

The fundamental role of analytics is to use data to gain insight and make decisions. But, like comparing fast food to a gourmet meal, there’s a really broad menu of options for success. Analytics, when done well, is a wonderful combination of many different things: mastering the obvious, a few moments of clarity, great listening skills, connecting the dots, and some really smart strokes of insight and organization.

Whether you’re an analytics pro or a newcomer, it’s easy to get overwhelmed by metrics and key performance indicators (KPIs). To make things less complex, let’s explore a few rules that I like to follow when assessing any project. From determining campaign impact to brand tracking, it’s critical to decide on an approach and follow it.  

Less is More

It’s vital to decide on a couple of very limited KPIs that will be the “stars” of the show. While there will be many other KPIs to track, having a small set that are the main metrics is very important. From this set of limited KPIs, you can then build relationships/correlations/models between these key metrics and other metrics that you’re tracking. For instance, in the health care industry, use a non-digital data point such as “procedures” or “appointments” and test its relationships with digital stats like site traffic and online appointments. That allows linkages between critical parts of a business and interactions with customers. 

From your set of key KPIs, you can then begin to pull in metrics from other disciplines like brand tracking, media impressions and call tracking to see how data relationships can be used to help predict future success. You can begin to understand whether other metrics can be leading (or lagging) indicators of your key metrics. However, you really cannot accomplish this without first knowing what your most important KPIs are.

Line up your data and organize it

Try to create a foundation of the data collection that is the same across all of your data points. Having messy data is like having a messy room. You might know where everything is, but it’s going to take you some time to get organized. If some of your data is weekly, monthly, by user, and by device, it’s incredibly difficult to piece it all together and draw common conclusions. 

So create a master set of data that has all of your KPIs on a similar analytic framework. By establishing a set of as many metrics as possible on, for example, a monthly basis, a common framework for tracking and relationships is built. This will allow you to create relationships between data points that you would have previously thought were not comparable. While you should still analyze the data in a more disaggregated nature, using the conclusions generated from things like a longitudinal “roll-up” allows your analysis to cascade down to the more detailed data.

Have fun and build a story

Like all good detective work, there is a sequence of events in  marketing, and mapping them out in a timeline or a story really helps you understand your KPIs and why they move the way they do. Creating a timeline and storyline is key to understanding data. Looking at it in this framework makes it fun and helps you make sense of your brand.