Sometimes Excel is enough!
Data should be presented accurately and effectively. As a data scientist, my goal is to package results in a useful manner (and occasionally, it turns out to be interesting, bonus!). Sometimes data is sent to me via spreadsheet or maybe I'm under a time crunch (you've been there, right?) and I need to get something together to present shortly. This is to say, not all data visualizations are or need to be flashy. I know it would be so cool to be a d3 or bokeh whiz, but as I am personally not an expert in these specific tools, it's still possible to create something good enough that gets the job done. In fact, I keep on hearing this quote in the voice of a manager on an adjacent team, "Perfect is the enemy of the good", which has been a guiding mantra as I worked in roles at the company.
I recently came across a blog post about ways to improve line charts in Excel and I was curious about what kinds of improvements were meant by the author. Many kudos to Storytelling with Data for walking through the example by providing context, highlighting menu options within the application, and explaining the rationale behind the adjustments. The graph at the beginning is accurate and just....it's fine:
But I zeroed in on some items I would have quickly changed according to my training in academia and experience as a science communicator.
What's up with repeated information in the legend and title?
Maybe not orange.
The many zeroes are impressive but perhaps a bit distracting.
And maybe the most important question, what is the point of this graph?
Should I be impressed by the organization's growth number of meals served? Am I meant to focus on the peak number of meals served? Should I be concerned that the number of meals served is declining?
Do not leave your audience guessing and do not assume they can read your mind! It's one thing to give a presentation live and walk your audience through the findings, but you leave yourself open to distracted members trying to guess at the purpose of the graph rather than paying attention to what you are saying. If you were sending a deck or this graph as a standalone piece of information to a colleague, do they have the background information required to understand the point of the graph? Are you making an assumption?
Deep breath....setting aside the anxiety-provoking thoughts....and this is just a hypothetical example! Let's see how the author modifies the chart.
In the modified version of the graph, we see that the title of the graph summarizes the trend of the data and are made aware of which part of the graph the author is calling your attention to. I learned how to emphasize data points (in this case, the data for 2016), which seems to me to be a pretty handy tool! The author also removes the excess zeroes, the legend (there's only one set of data to consider), the grid lines, changes the line color, and even includes a call to action.
So regardless of the tool, let's make sure the data is presented accurately and is able to convey the message that is intended to get across. I appreciate seeing how others present things and what they keep in mind. I'll keep these tips in mind the next time I need to crank out a line chart in Excel and I'll definitely continue to check out this resource for visualization inspiration!