While having data is a good start, you need to put that data into context, meaning you have to figure out the “who, what, when, where, why and how” before you know what the numbers mean. Discovering this meta-data can completely change your perspective on a set of data so you can interpret them appropriately. Once you understand what is interesting or abnormal about a data set, you can communicate it well to your boss, clients and other involved parties.
Why “Who” Matters: If you’re conducting a study to determine what improvements can be made to your customer service, the opinions of your long time customers might differ from customers who bought your software in the last month. So you’ll want to know who took the study and who administered the study as you perform data analysis.
Why “What” Matters: Finding out what the data contains through statistical methods is important and it’s critical to compare what is in the data to other related studies to ensure that your conclusions make sense.
Why “When” Matters: Timing changes data results. For example, if you are collecting data after a PR disaster takes place, you might see answers that are not in support of your company’s practices due to the bad press. Further, if you are utilizing old market research data from your company to perform data analysis on, it’s important to recognize that the implications of the study may not be valid due to the gap in time.
Why “Where” Matters: If your participants are from different countries or states, their opinions and data may differ. Therefore, making global assumptions based on US data is not a good idea. Also knowing where data was collected can help you make inferences! For example, you may find data collected in person may be less accurate than data collected via the web as people on the web can be more anonymous and truthful in their responses.
Why “Why” Matters: Often times, confirmation bias can appear in data where researchers are looking for data to confirm their hypothesis. If you know a particular study is taking place to identify problems with customer service, you may want to watch out for strong data that shows there are no problems with customer service as this is the information a researcher would want to bring to their boss.
Why “How” Matters: Data collection techniques can often influence the data. If your sample size is incorrect or the methodology chosen is not meant to capture the data you need, you’ll find inconsistencies within your data and your conclusions may not be the right actions to take.
If you are discovering unusual patterns within your data analysis or your statistical significance is not strong enough, you might not have enough data to make valid conclusions. Without doing data analysis, you don’t get the opportunity to evaluate your data before making actionable plans.
In the end, data is meaningless without context and without context, you cannot turn data into information. And information is useless without being able to apply it to something, creating wisdom. For example, through data analysis you may uncover that 95% of customers switch to your competitor based on renewal price increases. This is turning data into information. Then you can use that information to make actionable plans to lower renewal prices so that you’ll keep more customers.
Using the correct methodologies and data collection techniques can help you make accurate conclusions and actionable plans from your data analysis. But before any of this can take place, you have to get your superiors and clients to buy into the process. In order to show your bosses, clients and co-workers the value of market research, download our free eBook “How to Communicate the Value of Market Research to Your Clients” today. You’ll learn about the clear connection between market research and increased loyalty, sales and profits while learning how to turn market research insights into focused action and measurable ROI: