Many of you are hunkering down during Hurricane Beryl as I type this, and I hope you all did/are doing alright!
Last week, I wrote about being more conscientious when choosing color to map data displays. This week, I am expanding beyond maps to charts and graphs. Also, there’s a difference between a chart and a graph apparently...and I’ll act like I knew that before I did my research for this week’s newsletter 😉
A chart is a table, diagram, or picture that organizes large data clearly and concisely and is often used to interpret current data and make predictions. A graph, however, focuses on raw data and trends over time.
Glad we got that straightened out. Now, let’s dig in!
Sarah Compton
Editor, Enspired
Innovative Data Displays Pt 2
Variety is the spice of… data displays. As a geoscientist, I’m used to seeing, and making, three types of graphs:
Scatter plots. Sometimes the data are color coded or size-coded to convey a third and fourth dimension. Ooooooh.
Line graphs. I refer to these very technically as “squiggle charts.” Our log plots fall into this category and are the most common line graphs we encounter. Production graphs might be in there too, but usually we deal with log plots on the daily.
Christmas pie charts or doughnuts. Ok, probably literally no one calls them that, but these are our production pies and are often red and green, sometimes with blue thrown in. These might be overlain on a map to show geospatial relationships.
A data display smorgasbord. I invite all of us to expand our horizons. Why have only pies and doughnuts when you can sample from the entire data display dessert tray?
Consider mixing in these charts:
Area graphs—especially with a cutoff to show above/below values, such as your break-even production numbers.
Tree maps. If your pie chart has so many sections you can’t see what they are (whether that’s due to a lack of space on your slide or because there are so many categories), consider a tree map.
Sankey diagrams. My brain would not let me read this as anything except “snakey” diagrams. You will not unsee that because it fits! These show volume and direction of flows through the stages of a process or system.
Show then tell. Ideally, your audience won’t need you to explain your beautifully built display, but they will need you to explain why they should give a darn. That’s where the true art comes in.
A message from MicroSeismic
Welcome to a lively conversation between thought leader and industry visionary Peter Duncan, Microseismic’s founder and CEO and AAPG’s Dr. Susan Nash. Peter guides us through the history of how Microseismic’s innovative approach to monitoring wells made it possible to hear the “snap, crackle, pop” of the rocks fracturing, and thus map the induced fractures, something that turned out to be extremely valuable for operations.
Visual aids are all well and good, but without context, perhaps provided by a narrative, they are little more than pretty pictures.
Let the data guide you. There’s a tendency to think that storytelling requires cherry-picking talking points or skipping over portions of the data, but you can take steps to avoid biases.
Proof your data and visuals. There’s nothing more frustrating than realizing your calculation was off or you grabbed the wrong data. Proof. Then proof again.
Save and step back. I often find the skills required to build great visuals can hinder my interpretation of them, because building requires being in the weeds and trees, while interpreting requires seeing the forest. I like to get my visuals built and organized, then step back for at least 15–20 minutes to clear my head.
Start in isolation. Know and understand the implications of each piece separately, then think big picture: zoom out and see how or why those implications might fit together.
Iterate. Once I start interpreting, I find other questions come up, so I’ll have to make more visuals or add more data to what I’ve already made. This is where skills like programming and BI tools can help. I don’t want to go back into Excel and recreate my complicated visual with 15 separate data series.
Outliers can be key. I like to say we can’t uninvite outliers to our birthday party simply because we don’t like them. This isn’t second grade, and there could be great insight in those outliers. Assuming there isn’t a justified reason (e.g. the tool broke) to toss them out, keep the outliers in and dig into what’s going on there.
How you say it matters. As geoscientists, we’re trained on the information that makes up the story of the Earth, but how we tell it matters. Think of the difference between these two narratives.
250–65 million years ago, creatures known as the dinosaurs were the predominant land animals.
Some of the largest terrestrial animals in the history of the planet ruled Earth between 250 and 65 million years ago.
Those two sentences are both factually accurate, but one is more likely to stick than the other. The volume and order of words and ideas matter, so keep the important stuff in mind and structure your message appropriately. When in doubt, simpler is better.
While we might not want, and likely don’t need, to provide “entertainment,” the fact is that people remember things better when they’re actively engaged, so crafting your narrative in an engaging way is key to explaining the data to stakeholders.
Some more info on how to craft your narrative with data can be found here.
👍 If you enjoyed this edition of Enspired, consider supporting AAPG's brand of newsletters by forwarding to a friend or colleague and signing up for our other newsletters here.
➡️ Was this email forwarded to you? Sign up for Enspired here.
AAPG thanks our advertisers for their support. Sponsorship has no influence on editorial content. If you're interested in supporting AAPG digital products, reach out to Melissa Roberts.
.png?upscale=true&width=220&upscale=true&name=Sarah-Compton-Headshot-Signature%20(1).png)
