Making your graphs better
"Everything starts out looking like a toy" (No. 17)
This week’s toy: when you can’t remember a song, just hum the song to Google and it will find a potential match. I’ve been wondering when Shazam would have this feature since … well, a long time. Edition No. 17 of this newsletter is here - it’s October 24, 2020.
The Big Idea
Photo by Markus Winkler on Unsplash
We expect metrics to be done and valid when they are presented. For example, when you look at a chart of month over month comparisons, you expect all of the parts of the graph to be comparable. But what happens when you’re in the middle of a month and trying to show the most up-to-date data?
Frustration like this might occur:
Sam Parr ⚪️ @theSamParrDownloads for our podcast. Took a while to get going...but good content and audience creation...it compounds!!! It just takes a while. So, if you're in the content game, and you think you're any good...keep going. If you stink, change! But if you have some talent...keep going! https://t.co/NTzoYcbBBH
Josh’s point is well taken. When you are looking at data visualization, you very infrequently have a view into the underlying data that provides that information. Are you looking at derived information (partial month, scaled to a full month), real-time information (we haven’t finished the month, so we’re not ready to compare to previous months yet) or the most recent completed period (all data through last month?
Making your Charts More Understandable
Here’s a modest suggestion for making data visualizations more readable (and to set the context). Adjacent to the graph (in a note or otherwise), state the form of the underlying data and its periodicity so that it’s possible to know when you are getting the exact measurement (weekly sales data, for example) and when you are getting scaled information that’s based on a larger or smaller time span.
It’s not possible to account for all data misunderstanding, but much easier to get the idea of a visualization when you know what it was originally intended to convey. And there’s also the question of how to create a visual reminder that you are looking at partial data. This one’s a bit harder.
Show Partial Data Consistently
Here are a few ways to indicate partial data:
Truncate the time frame to the most recent completed period
Create a new data series showing a completion estimate for the current period scaled on the current data and percentage completed
Estimate completion for the current period taking into account seasonality and prior year periods
Just label your graphs clearly ;)
What’s the takeaway? There’s no one way to make your data clear for everyone. However, if you’re creating a visualization and showing partial data, it’s a lot easier to understand if you either call out the partially completed data with a visual (different color or treatment) or omit it from the data visualization and explain in the notes.
We’d like to know …
With the upcoming Thanksgiving holiday, many of us would be traveling in a typical year, either in town or across the country. Are you traveling this year, or staying at home? Guests or no guests?
Click the tweet to tell us how you’re voting.
Links for Reading and Sharing
These are links that caught my eye.
1/ Advertising is coming to the freezer aisle - Cooler Screens is creating an ad experience anchored on the cold case at supermarkets. This - combined with A/R on phones - could either be really annoying or allow you find that flavor of ice cream you’re looking for … faster.
2/ MECHA CGI - This trailer (it’s in Portuguese, but you get the idea anyway) shows the possibility of fan created scifi. I think the future looks great, and hoping that Netflix picks up this show and gives this creator a platform.
3/ The COVID-19 Recession is not equal - It’s been long enough that we can see clear impacts from the COVID-19 Pandemic. One of the things that’s happening, diagrammed by the Washington Post, is that the effects are not equal. Some people are surviving just fine, while for others their livelihood has disappeared and may not return.
On the Reading/Watching List
On Teaching Yourself How to Think: if you’re interested in learning how to think in first principles - that is, assumptions that themselves do not need to be proven - this guide is a great place to start.
You might not know that Mission Impossible: Ghost Protocol (2011) was directed by Brad Bird of Incredibles fame … it’s a great 🍿 movie for when you just want to turn off your brain and just enjoy the almost-impossible. stream on Amazon Prime
What to do next
Hit reply if you’ve got links to share, data stories, or want to say hello.
I’m grateful you read this far. Thank you. If you found this useful, consider sharing with a friend.
The next big thing always starts out being dismissed as a “toy.” - Chris Dixon