The Key to Effective Analytics: Know the Question You Are Trying to Answer

Dr. Wendy Tate
Director, Research Operations, Forte
December 27th, 2017

Many research institutions understand the value of analytics and measuring their clinical trial performance, but it can be difficult to find the right starting point. It’s exciting to think of the possibilities – every data point and metric – and want to track them all. However, people can become obsessed with numbers and dashboards without understanding what they’re really looking for. You won’t miraculously find insights by simply looking hard enough at the data.

Finding an answer and then trying to figure out the question you were supposed to ask to get it is a backwards way of thinking. That’s why it’s important to take a step back and think critically about what it is you really need to report on, rather than reporting for the sake of reporting.

To avoid having an unfocused analytics strategy that doesn’t create meaningful value, begin with identifying the questions you want the data to answer. Take the time to write down all the questions you have and the answers you want to find. The quality of the question will determine the quality of the insight you receive (e.g., if you ask generic questions, you will get generic answers). Get feedback from others in your department as well as leadership to make sure you have a complete set.

Next, do you know why you want each answer for the questions on your list? Do you know the subsequent actions you will take? Think about the following:

  • What are you going to do with this information?
  • Will it help bring forth meaningful improvements?
  • What are you going to change based on the answer?

If you’re not able to answer these questions for each insight you’re looking to receive, that insight probably isn’t worth tracking. Or, the question you’re looking to answer may need to be asked in a different way. Without clearly defined and specific questions, you won’t get specific actionable answers.

For example, when reviewing protocol performance, the question “How is our enrollment doing?” might come up. This is extremely generic. You may be able to gather some data after asking this question, such as the number of subjects enrolled on a trial, but it’s unlikely this will result in enough information.

A better actionable question would be, “Which protocols have underperforming enrollment and need intervention to meet their accrual goal?” or “Which protocols are so far from their accrual goals that they should be closed?” These questions allow you to take immediate action (e.g., strategize on plans if the trial will stay open, close out the trial, etc.).

Asking good questions is as important as knowing the right answer, and the better the questions, the more valuable the answers. The data itself isn’t all that helpful if you haven’t figured out what problems you’re trying to solve. Once you do this, you can begin to answer the questions that will have the biggest impact to your organization’s success and decision making.

Learn more about using operational performance metrics to find answers to your critical questions. Download this free eBook.

To provide institutions with the information needed to influence current actions and impact future outcomes of their clinical trial performance, we created Forte Insights, a cloud-based analytics tool. With ready-to-go dashboards populated with data from OnCore Enterprise Research system, Forte Insights helps inform strategic decision making by providing quick answers to complex clinical research questions.

Learn more about Forte Insights



2 thoughts on “The Key to Effective Analytics: Know the Question You Are Trying to Answer

  1. Very nice article. I absolutely agree that before you ask a question, you should know why you are asking it and if you really need to know the answer. I find this to be true in analytics, protocol testing requirements, and in case report forms. Asking for unnecessary answers or data simply for the sake of “covering all bases” can waste a lot of valuable time and resources in the long term goals of a trial. Spending a little more time in the beginning phases of a study, funneling down to just the questions needed to be answered and the basis for asking them, can help greatly in the long term to get the needed answers without costly complications with study process, data quality, and time spent monitoring metrics that provide little or no appreciable oversight value. Thank you for a great article.

    1. Thanks for your response, Rina. Great points on the additional activities this exercise can be applied to as well. I couldn’t agree more on the need to think through more upfront to prevent wasted time on the trial downstream.

      Thanks for sharing!

Comments are closed.