How Marketers Can Use Data Without Getting Lost in the Forest

Data is here, there, and everywhere. Big data, data sets, data analysis, data analysts, data scientists, the list goes on and on. As a marketer, it’s easy to become lost sifting through gigabytes of information. What is important? What do we want to track? How do we know we are asking the right questions?

The key to successfully using data is to realize it’s just another tool, and that it should support the company’s overall goals and mission. Data is not the end goal, but rather a means to an end.

So, what does this mean? If you’re analyzing data to find a needle in the haystack, take a step back. This is not a productive way to approach information. The best marketers take a systematic, top down approach to leveraging data that directly ties into the core drivers and mission of the company.

Using this approach, you might structure your initial data analysis with the following questions:

• What are the key metrics that drive business success? For some companies, this might be unit sales, or subscriptions, or registrations, or retention.
• Put your key metrics up on a whiteboard. Then, begin to drill down on the sub-metrics that sum to these key metrics. Many call this the sales/marketing funnel. For unit sales in an ecommerce company, it might be conversion rate, defined as the number of visitors to a site versus the number who complete a sale.
• Take a look at these sub-metrics to understand which have the most impact on the key metrics. Why? Begin to look at key differentiating factors: traffic source, gender, income, or other information that is relevant to your business.
• Develop some hypotheses. As you are drilling down, you start to develop hypotheses. Design an analysis with data to test that hypothesis.

As you begin to test and analyze, you might find some interesting deviations or differences. These can become the fodder for additional analysis. Over time, this analysis can become quite complex, involving regression, huge data sets, and the kind of rigor associated with a powerful data shop. But it all starts with asking some key questions that align with the goals of the business.

Data Analysis Example

Let’s use an example to illustrate this approach.

You are a company that sells online subscriptions. You have a massive data set distributed across multiple databases that include basic and demographic information on your customers, data on where they have come from, data on what they do on the site, third party psychographic information, information on users who registered but didn’t buy, and payment information.

You have access to Excel, Tableau, R, and any other data tool you want.

So, how do you logically start extracting data from the information that you have?

First, identify the key metrics that fuel the business. In this case, you have identified two key metrics: 1) initial subscribers; 2) retention.

You decide to drill down on initial subscribers and identify the sub-metric for this. After some discussion with your team, you identify two sub-metrics:

• Visits
• Conversion rate

After examining the data, you notice that visitors who spend twice the time on the site as others have 3x the conversion rate. You dig some more and notice that these visitors are reading a certain set of articles about a particular topic. This kicks off an entire analysis around the type of topics, whether these users are coming specifically for this topic or find them upon reaching the site, the profile of users who are interested in this topic, and more. The goal of the analysis is straightforward, can you improve or expand these articles to increase visits and conversion rate, ultimately leading to more initial subscribers? The tools and methods you use will depend on the analysis that is necessary to answer your questions. In this way, you can spawn many lines of inquiry in which the answers have a direct impact on the success of the business.

Don’t become overwhelmed by the all of the tools, acronyms, and sheer amount of information at your disposal. By taking a top-down approach, you can put a real strategy around your data efforts that will provide clarity and true value added.