How To Tell a Story With Data…in Ten Steps!
Organizations are increasingly using research, data, and analytics to inform decisions and ensure they are correct the first time around. This growing focus on data-driven decision making spans corporate, non-profit, and education sectors as a way to ensure that organizations, regardless of industry, location, or size, are making the best decisions in the most efficient and cost-effective manner. Although conducting research, gathering data, and crunching numbers are integral steps in the data-driven decision making process, another critical component that is often overlooked focuses on delivering information in a manner that drives action. If a research study or analytic initiative doesn’t drive action, there is little to no benefit to spending time, money, or resources on the data process – regardless of how sophisticated the research design or statistical analyses are. It is important to translate information into a cohesive and compelling story to motivate and inform decision makers to act on the data-driven insights.
It is important to translate information into a cohesive and compelling story to motivate and inform decision makers to act on the data-driven insights.
Storytelling refers to the translation of research and numbers into words and images to present and communicate information in a cohesive and compelling manner. Although the term “story” implies that information will be transferred verbally, it is also likely that you will need to develop some form of deliverable (i.e., a report, presentation, or one-pager) for the decision makers’ future use. This deliverable not only needs to tell a cohesive and compelling story, but also needs to stand on its own and be clear in your absence. This post provides a step-by-step guide to help tell the story of your data through a deliverable such as a report, presentation, or one-pager.
Identify your audience. The first step to telling a good story is identifying and understanding your audience. You can do this by answering a few questions such as…who are they, why are they interested in this information, what will they do with it, and how much time will they have to consume it. Be sure to identify potential secondary and tertiary audiences as well. For example, you may present information to a small group of employees, who will then pass it along to their board of directors, who will then request that the information be posted on a company website for customers or community members to read. You should identify and understand all audience targets before moving to the next step.
Select your delivery mechanism. Once you understand your audience, you can determine the best way to deliver the information to them – that is, your delivery mechanism or deliverable. Your delivery mechanism could be a detailed report, high-level slide deck, or simple one-pager. To complete this step, think about the answers to the questions in Step 1, especially why do they want the information and how much time will they have to consume it. You should also take into account their interest and level of knowledge regarding the context, research study, data, and analyses. You may need to create more than one deliverable depending on the size and complexity of your audience. For example, a detailed report would fulfill the needs of the small group of employees, but the board of directors may benefit from a high-level slide deck, and a one-page infographic may be perfect for the organization's website.
Create an analysis plan. To avoid having a bunch of information (i.e., metrics, data, output) that you don’t know what to do with while still missing vital information, it is important to create a solid analysis plan. This analysis plan should, at a minimum, outline the questions you are trying to answer, the data you will be acquiring, and the analyses you will run. It is beneficial to organize this in a way that links each analysis back to specific data elements and research questions. When you are creating your plan, you should constantly be thinking about the questions that need to be answered. This will ensure you create an exhaustive list of analyses, but do not conduct analyses that are not relevant to the study. Basically, this helps you avoid analysis paralysis!
Analyze and document. Once you have acquired your data, you can move into the analysis phase. Your analysis plan will help you stay organized and focused on your research questions. While analyzing data, it is helpful to document all of your output. Excel is a great place to do this because you can create a tab for each set of analyses (e.g., descriptive statistics, research question #1, research question #2). There’s no need to spend time making the file pretty or adding any graphs at this stage. You are just getting all of your output into one place so you can determine what the data is telling you.
Conduct the “think, feel, do” exercise. After you have conducted all of your analyses, you can review your output (probably a few times) and identify what the data is telling you – hence, the answers to your research questions. Ask yourself the following question: What do I want my audience to think, feel, and do based on this information. You should have one statement for think (e.g., I want them to believe the program is effective), feel (e.g., I want them to be happy that their investment is producing a return), and do (e.g., I want them to expand the program to other groups). If you have multiple audiences, you should do this exercise for each one.
Storyboard your findings. To ensure your story flows properly, you should storyboard before you open any type of publishing software. A storyboard is a visual outline that can be created on paper, sticky notes, or my favorite – a big whiteboard with one box or sticky note representing each page, slide, or section of your deliverable. You can keep it pretty high-level here. Think about the results of your think, feel, do exercise and ask yourself how do I get them there. Most likely, you will start with some background information and research questions, then some information on your study (participants, design, data elements), then go into the results, and finally end with key insights and recommended actions. One great thing about sticky notes is that you can move them around in order to find the best flow. Remember that you are telling a story so you need to include all parts so it is complete and cohesive, but also concise and compelling. A good way to check your completeness is by asking yourself: Did I cover the question, answer, and recommended action – or Q&A2. I recommend doing this exercise with a partner, preferably someone who is not as involved in the study as you. Keep in mind that you may need to do this multiple times if you are creating multiple deliverables. In this situation, I recommend finishing the steps with one deliverable, then using it to create the others rather than going through the steps simultaneously.
Visualize. Once you have your storyboard created, determine the optimal data visualizations for each of your findings. Not all types of visualization are good for all type of data. Therefore, it is imperative that you choose the right type of visualization for the data you have. This is where all of your documentation can help. You should go back and review the analyses you ran, the output you obtained, and the think, feel, do exercise. If you need help determining which data visualization to use, there are a lot of resources that can help. For example, a quick Pinterest or Google image search for “significant difference infographic” or “correlation data visualization” can help generate ideas. There are also a lot of books and blogs that provide great examples and support. I recommend following the 3-second rule for any data visualization – if it can’t be interpreted in 3 seconds or less by any audience, try a new visualization method. Keep in mind that you are just ideating and sketching here. You will create the actual visualizations in Step 8.
Transfer everything to your deliverable. Now that you’ve done all the hard work, open your publishing tool (i.e., Word, PowerPoint, Publisher, etc.) and start transferring your storyboard to your delivery mechanism. Make sure all of your headers and sub-headers interpret the slide or section for your audience – this maximizes the extent to which they will think, feel, and do what you want. You should also consider adding a table of contents or executive summary. This will be easy now that you have a storyboard completed, and also ensures that the story will stand on its own. Canva, Piktochart, Unsplash, and The Noun Project are all great sites to help with graphics creation in addition to the typical Excel, Tableau, etc. for creating graphs and tables.
Complete the storytelling checklist. Once you have a first draft of your deliverable created, I recommend reviewing our Storytelling Checklist to ensure you’ve covered every step completely. Reach out if you would like a copy of the checklist!
Conduct a test run. The final step in our storytelling guide is to deliver your story to people who are similar to your audience, especially as it relates to their level of knowledge regarding the research study, data, and analytics. Test run participants may be fellow coworkers or even friends and family willing to donate some of their time to you. If you are part of a team that does this type of work on a regular basis, you may benefit from adding a peer—review step to your process where all members of the team participate and provide feedback during the test run. After you have delivered the information, ask them about their thoughts, feelings, and intentions and make sure their answers match what you said you wanted the audience to think, feel, and do. If they don’t, you should make some adjustments. Be sure to also get their feedback on the story, flow, data visualizations, and headers.
We hope this step-by-step guide (and our checklist) will help you turn data into action! If you find yourself stuck at any particular step or need a bit more support in general, feel free to reach out!
