The HiPPO and his menagerie of managers looked at the data displayed on the screen and nodded in agreement. The swirls, lines, and colors of the data visualization painted a very clear picture about the company’s next move, amplifying the product manager’s recommendation. Decision made, the assembled herd moved on to the next agenda item.
Data has outsized power to sway and convince when it is presented in a way that is accessible to those looking at it. While spreadsheets and databases may be helpful to corral the data and do some analysis, it is through effective data visualization (charts, graphs, etc.) and storytelling that you can really sway hearts and minds and facilitate decision-making.
As a product manager, data visualization should be your sidekick, your Tonto, your Robin, your [insert favorite sidekick here]. Amassing data is relatively easy; analyzing it, drawing conclusions from it, and using it to support your arguments is the challenge. Taking that data and creating a compelling data visualization that is then used to assist in decision making is the real trick. And in those cases where you do not have an abundance of data, when you are faced with a data drought, there are still ways for you to eke out strong assessments supported by visuals..
That all sounds great, you might be mumbling to yourself, but how do I get started?In This Article:
- Intro to Data Visualization
- Data Science and the Sexy “Data Scientist”
- Data Visualization Tools
- The Setup: Using Charts and Graphs
- Storytelling With Data Visualizations
- Telling a Data Driven Story When You Don’t Have the Data
- The Future of Data Visualizations
- The TL;DR
Intro to Data Visualization
“I hear and I forget. I see and I remember. I do and I understand.” — Confucius
What is it?
Data visualization is broadly defined as “the techniques used to communicate data or information by encoding it as visual objects (e.g., points, lines or bars) contained in graphics. The goal is to communicate information clearly and efficiently to users.” In effect, it boils down to using graphics to reveal data, expose trends, uncover causality, and generally make the data and its analysis more accessible.
Data visualization is as much art as it is science and there are countless books, websites, and blogs focused on revealing the usefulness and beauty of well constructed graphs and charts. Professor Edward Tufte, the godfather of data visualization, was both prolific and exacting in producing useful, visually stunning pictures based on novel data sets. (Full disclosure: I attended one of his seminars many years ago, and have 4 well-worn books he wrote. But his aversion to powerpoint still stumps me.)
Why should you invest time in creating charts or other visual representations of data? The best ones help the reader draw conclusions they might otherwise have missed. Visualizations may be static or dynamic, two-dimensional or interactive. Giving users access to visual representations of your data, especially interactive ones, allows them to draw their own conclusions and gain a deeper understanding of what the numbers are saying.
Another major reason to use a data visualization: sometimes, you simply have too much data to work with and representing that information in a chart, graph, interactive graphic, infographic or the like is the only way to make sense of it:
“Visualizations convey information in a universal manner and make it simple to share ideas with others. It lets people ask others, ‘Do you see what I see?’ And it can even answer questions like ‘What would happen if we made an adjustment to that area?’” [source]
What isn’t it?
Data Visualization is not the same thing as Data Analytics. Visualization = making the data more user-friendly; Analytics = plumbing the data for insights. “[D]ata visualization is often a component of data analytics, but it should not be seen as the entire puzzle,” writes Colin Sebastian.
Data Science and the Sexy “Data Scientist”
All those websites, sensors, mobile phones, IoT appliances…they generate a colossal roving herd of data, an explosion of data. Data Science is a relatively new profession that is uniquely suited to wrangle and make sense of that data. And who are the people that put the data in science? That’s right, Data Scientists, named “Sexist Job of the 21st Century” by Harvard Business Review. (Yes, that HBR.) Sorry, George.
As a Product Manager, you may be lucky enough to have a data scientist (or scientists) on your team or within your organization. Aside from their apparent sex appeal, their mad analysis skills make them invaluable. Just make sure you know which strain of “-stein” you’re dealing with: Ein or Franken.
- Einsteins assess data searching for truth, or at least strive for objectivity, listening to what the data says.
- Frankensteins bend data to fit their desired conclusions.
Choose your -stein carefully.
Data Visualization Tools
What are the tools of the trade for creating great data visualizations?
Spreadsheets are the most basic option, they allow you to investigate data sets, create simple graphs and charts, and explore the data. But, scanning through rows of numbers will rarely reveal trends or patterns; it’s more helpful to have the data in an aggregated, visual format. Using a spreadsheet’s built in graphs and charts is a step up from raw numbers, facts, and data as these simple graphics make it easier to spot patterns.
Nick Brown has outlined a progression of tools which you can use depending on your technical prowess. His suggestions range from no-coding-required options like spreadsheets (already discussed), and Tableau, to some-coding approaches like R and SQL (focused on analysis, not presentation), to hardcore options like D3 and Python. Know thyself and choose thy weapon. Or, uh, don’t, and find someone who’s really proficient in one or more of these tools to help you out.
Want more? There are fit-to-purpose tools, and tools for mapping, media, timelines, charts, and infographics to name a few. The goal of using tools like this with great data sets? To create some incredible visuals.
The Setup: Using Charts and Graphs
Before you go too far, you must make sure you have great data to work with. Let’s just assume that you’ve got that covered.
“Everything should be made as simple as possible, but not simpler.” — Albert Einstein
Tips and Tricks for Creating Effective Visuals
When you are crafting visuals, here are a few best practices to keep in mind:
- Create data visualizations that empower your audience with the goal of inspiring action. For example, if creating a dashboard, ensure it is designed in such a way that it can be used to answer the most important strategic questions for the business, and indicates the threshold at which managers need to be concerned and (possibly) take action. Again, the goal is to have actionable and meaningful content.
- Context matters. Don’t leave users guessing what a number might mean in the bigger picture. Present metrics in comparison with set goals:“Color is an excellent data visualization technique to help demonstrate performance vs. goals.”
- Simple is better than complex. The point of creating a data visualization is to make the data more accessible. Focus on the most important messages first, keep them snackable.
Best Practices for Effectively Using Graphs and Charts
First off, match the type of visual to your purpose. Consider the following guidelines courtesy of Emily Rugaber in a post on GoodData:
- “Line Charts track changes or trends over time and show the relationship between two or more variables.
- Bar Charts are used to compare quantities of different categories.
- Scatter Plots show joint variation of two data items.
- Bubble Charts show joint variation of three data items.
- Pie Charts are used to compare parts of a whole and should be used carefully. Never compare two pie charts without clearly noting that the size of the pie may have changed as well.”
Reduce the Noise. The other consideration is whether the visualization is an efficient way to represent the information. “The most important consideration when designing for efficiency is that every bit of visual content will make it take longer to find any particular element in the visualization. The less data and visual noise there is on the page, the easier it will be for readers to find what they’re looking for.,” says Noah Iliinsky in a post on User Interface Engineering. Another way to reduce noise is to rely on legends and axes to establish context and consistency.
Emphasize the most relevant parts. Is there something in the visualization which you want to emphasize? If so, consider making it “bigger, bolder, brighter, or more detailed, or called out with circles, arrows, or labels. Alternately, the less-relevant content can be de-emphasized with less intense colors, lighter line weight, or lack of detail,” Iliinsky advises.
Make it pretty. You’ve spent the time to make sure the goal of the visualization is clear, so apply the right amount of visual polish to ensure people spend time looking at and using it. More advice from Iliinsky: “A familiar look and feel can make it easier or more comfortable for readers to accept the information being presented… At times, designers may want to make choices that could interfere with the usability of some or all of the visualization. This might be to emphasize one particular message at the cost of others, to make an artistic statement, to make the visualization fit into a limited space, or simply to make the visualization more pleasing or interesting to look at. These are all legitimate choices, as long as they are done with intention and understanding of their impact on the overall utility.” (P.S. you should really check out his article on effective visualizations for some more tips)
Storytelling With Data Visualizations
“It’s not what you look at that matters, it’s what you see.” –Henry David Thoreau
Hans Rosling is a master at using compelling visuals – data-rich, dynamic, interactive – to support vivid storytelling and to make a larger point: “Data and information are not inherently boring. The key is to select the appropriate (and accurate) data to support your message. Yet, it also matters how you bring the data alive, giving it context and meaning.”
As Garr Reynolds writes, Dr. Rosling is effective because he takes the time to set up the visualization: “Dr. Rosling consistently does something that few presenters ever do. That is, he takes the time to set-up for his audience the display of his data before revealing the actual visualization to them. Most people show everything all at once—without ever pointing out what we should look for or explaining the variables—and rush to their conclusion about what they say the data shows without us ever getting a chance to really see it with our own eyes.”
Dr. Rosling also offers some very practical hints for presenting dynamic data visualizations:
- “Use full screen to maximize the view
- Explain *first* the vertical and horizontal axes as well as the meaning of the size and color or bubbles
- Mouse over a few of the bubbles that you want people to pay special attention to
- Set the optimum speed and tell the audience when you’re going to start the animation
- Explain the meaning of the movement as it is happening”
Telling a Data Driven Story When You Don’t Have the Data
Sometimes, you just do not have good data, enough data or statistically significant data. How do you create a visual to support a data-limited story? This is a very real problem that many startups face, and one route they take is…“when there are no metrics, gather opinions.” (Just be sure not to treat those opinions as facts.)
The Future of Data Visualizations
A brief note on the future of data visualization: according to John Whittaker, executive director of marketing for Dell Software’s information management group, we’re on the cusp of a new phase where predictive capabilities are becoming more common, and their accuracy is improving. “The first wave was exploration. The next was, ‘this is interesting but what other data can I bring in?’ — big data oriented analysis. The next phase is looking forward to what may happen, rather than what just happened,” he says.
“Today we need micro analysis on a macro scale,” says Shay Har-Noy, senior director of geospatial big data at DigitalGlobe, “There’s a ‘show-me-where’ class of problems we call geospatial big data, where it’s ‘show me where there’s new construction in the country’ (or where they should be new construction in the country), ‘where supply chain bottlenecks are across an international supply chain,’ or ‘where there’s increased traffic around ports indicating suspicious activity or increased economic activity.'”
Finally, one of the most transformative changes has been the willingness to share the data, and givie the general public access to what were once proprietary data sets. With these datasets in hand, individuals and organizations are then able to apply their own visualizations, thus enhancing the usefulness of the data.
You have data, lots of data. How can you make sense of it all, and present it in such a way that it is understandable? Use a data visualization to reveal data, expose trends, uncover causality, and generally make the data and its analysis more accessible. Data visualizations are a great tool to assist in data driven decisionmaking by making data understandable, and they can also support a compelling story.