Data Driven: Making Sense of It All
Data are everywhere. Anyone with a smartphone is his or her own (not so private) data point. Technology—and an ever-growing need to track, measure and optimize ROIs—has fostered a corresponding need to analyze and understand data.
In today's world, marketers are exposed to an abundance of data and analytics. The challenge is making sense of it all and using it to guide actions and decisions, not to paralyze or impair them.
Understanding the intended action that will come from information and its analysis is a good place to start. In rather simple terms, data and analytics are used for three broad purposes:
1. Describe: Data can tell a story about what has happened. It is often a snapshot in time that provides information on—or a description of—something that has occurred.
2. Predict: Data are often used to forecast a future occurrence or outcome. There are a host of predictive techniques that use sets of data points to make educated assumptions about the future.
3. Decide: Data are almost always used to help inform and guide some business decision and action. Data are at the heart of most organizations' performance measurement systems and are used to make key decisions.
Data and analytics can be used for any one of these purposes, or for all three. A framework for interpretation and use of data and analytics leans heavily on the basic scientific method we were taught in grade school. The framework relies on a four-step process, with each step having its own set of golden rules.
Step 1: Know the
Question and the
Answer in Advance
If you don't start with a clear understanding of what you are trying to find out or what problem you are attempting to solve, making sense of the data can be difficult. In addition, go into the process with a reasonable definition of success, failure or any gray area in between. A clear or even broad idea of how to define success at the beginning of the process can result in a timelier, productive analysis at the end.
Step 2: Properly Absorb and Understand the Data
Interpreting data can be like reading tea leaves. Common tips for a proper understanding include:
• Scan all of the data in advance of making any judgment or analysis. Allow the cognitive portion of your brain to absorb what you are looking at.
• Know your data source and its pros and cons. This includes where it comes from, what systems create it and what time period it covers. In written presentations of data, always read the footnotes and fine print.
• Nomenclature is critical. Fully understand how data sources or collectors define key terms. Because of a lack of commonly used industry standards, terms such as "gross sales" or even "Web visits" can mean very different things to different people, organizations and systems.
• Realize that not all data are created equal. Some data are more important to your business or your efforts than others. Balance the quantity of data points with the quality of the data and its ability to help tell a story, support a point of view and truly track performance.
Step 3: Derive Meaning and Form a Hypothesis
This is where things get tricky and ultimately where the value of data and analysis come into play. Logic is the overriding mantra for good, solid data analysis. Logic can guide an ability to see patterns and trends in the data, or even identify relationships between various data points. Cause/effect relationships from seemingly independent data points can be a valuable observation to help form a hypothesis.
Whenever possible, validate key data points against other data points. This triangulation can often minimize any inherent risk in your analysis or uncertainty of a hypothesis. And keep in mind; the use of any single data point to derive intelligence or meaning can be risky.
External references, industry standards and benchmarks can help provide a comparison and context for your data. However, be aware of making an apples to tangerines comparison when apples to apples are in order.
Keep in mind, data derived from small universes or a relatively short period of time have inherent value issues. Also, beware of averages. While an average plays a role, it can also greatly misinform. If you are looking at an average, make sure you fully understand the underlying range of numbers that it came from.
Step 4: Effectively Communicate Your Findings
Data, and the analysis of data, has little use if it is not properly translated and communicated. And as a marketer, your role is to know in advance what you want to accomplish and ultimately communicate. Data should be used to help tell a story, inform an audience and support findings or recommendations.
Never present data without some explanation. Clearly highlight or identify some of the most relevant and interesting data points and key findings. Cite sources, formulas and other critical information inherent with the presentation of data through charts, graphs or tables.
Visuals can be very effective in presenting relationships, patterns or trends between data points. In order to avoid misleading data communication, draw visuals to scale. Also avoid one simple data presentation misstep (or trick, depending on your perspective) by starting or scaling your chart from the value of 0 (zero) when showing an x-axis and/or y-axis relationship.
And finally, be aware of data and analysis bias. Bias or prejudice (intended or unintended) is possible in most data and analytics. It can come from a bias in a system or a process. It can come from a bias in human or constituent agendas (overt or hidden), or simply from a bias in the numbers themselves. The key is in understanding and acknowledging bias and, where possible, minimize its influence on your ability to achieve or support your ultimate purpose: to describe, predict and/or decide.
Chad Giddings is executive vice president of marketing and planning at the Mission, Kan., direct marketing agency J. Schmid & Associates. He can be reached via e-mail at email@example.com.