Big data is big business.
Major industries—from health care, to energy, to retail—shell out billions of dollars for data collection and analysis.
If information is the “new oil,” then reserves are booming. Amazingly, 80 percent of the world’s data was generated in the last two years alone.
Distilling such a massive volume of numbers into meaningful and actionable chunks is a formidable challenge, one that this humble blogger will leave to statisticians and data scientists.
But big data presents a unique challenge to communications professionals, too. For collecting and analyzing big data is one thing. Communicating what it means is an entirely different undertaking.
Data is the centerpiece of much workplace communication, including annual reports, market analyses, executive summaries, and other correspondence. All these forms share something in common: they will be read by a mostly non-technical audience. Yet this same audience may include stakeholders who have to make important decisions and take actions based on the data.
When reporting data, your goal is twofold: (1) Make the data “digestible” and (2) explain its significance. Your communications should focus less on the method of analysis and more on the big-picture results. Accuracy and clarity are paramount.
Organize for Clarity
Your data-based writing can achieve clarity by focusing on one main idea. Following a three-part organizational pattern similar to SEA will help you develop this idea. This method states the main idea first, follows with details to support it, and ends by calling the reader to action.
When writing about data, use a similar three-part structure:
- Opening: Introduce the main idea. The main idea is the “big idea,” or insight, that comes from the data. The idea could be an emerging trend, a data-based prediction, a meaningful comparison, a consequential outlier, or something else of value. You should be able to state the idea in a single sentence.
- Middle: Support the main idea with data. You can do so by pointing out high and low points, changes over time, and other pertinent numbers. If you include tables or graphics, make sure to explain what they mean.
- Closing: Restate the main idea to show how it impacts your audience. If appropriate, make recommendations based on the data.
For more complex data analysis, secondary results may need reporting. Highlight these in the middle part with separate supporting paragraphs for each new insight. These paragraphs should follow a modified version of the three-part structure: 1) Begin by identifying the insight; 2) support the insight with data; and 3) close with a conclusion based on the insight.
The middle part is also where you should report and respond to any data that differs from or contradicts your main idea.
Simplify Word Choice
Another way to clarify your data-based writing to simplify your words. Do so by:
- using plain language;
- removing unnecessary jargon and complicated language; and
- defining acronyms and technical terms.
Notice the difference in word choice in these two examples:
A) By deconstructing numerical research of the wood flooring industry, one can conclude revenue does not necessarily flourish in connection with MBF. If you direct your attention to the MBF and revenue patterns in West Virginia in Table 1, you will infer that the region is a bullish market for sales ($94 million) and a bearish market for production (45 MBF). Meanwhile, Georgia produces 215 MBF annually, while generating less than $27 million in sales. The proposition of a regional production-to-sales correlation is a falsity.
B) Our analysis of the wood flooring industry fails to show a geographic correlation between sales and production. Table 1 shows that West Virginia is the second leading market for wood flooring ($94 million annual sales) yet is home to only four manufacturers producing 45 million board feet (MBF) annually. Conversely, Georgia floor manufacturers produce upwards of 215 MBF but sold just $26.93 million within the state in 2013. This data suggests production does not drive revenue within geographic regions.
Can you see why the second example is clearer to a general reader? It uses plain language, defines acronyms, and cuts unnecessary jargon.
Visualizations are powerful communication tools. They often reveal the “big idea” of a data set more clearly than words, helping the audience digest the information. Not surprisingly, visualizations are also more engaging than numbers and words. Some are downright beautiful.
Thankfully, it doesn’t take a graphic design artist to create an engaging visualization. Many easy-to-use visualization tools are available for free online. Applications like Google Charts, Many Eyes, Tableau, and Visual.ly are a great place to start.
If you decide to build a data visualization for publication, keep in mind these helpful tips from the Data Journalism Handbook:
- Focus on one big idea. Your visualization should reflect the “big idea” of the data set. Think about the one impression you want to leave with the reader. Enhance that idea by removing unessential data or information.
- Design for two types of readers. The visualization should be easy enough to understand at a glance, but also offer something of value that will invite the viewer to study it more closely.
As with any communication, accuracy is vital. The final step to writing about data, or creating a visualization for publication, is having one or more trusted individuals check and revise your work.
Good communication is good business. That doesn’t change in our increasingly numbers-driven world. Wherever people are busy crunching numbers, someone needs to communicate what it all means.