A recent exchange that I had on LinkedIn underscored a humbling but crucial truth: the same chart design can seem clear and obvious to some but genuinely confusing to others, even among smart, well-intentioned people.
Full article: www.practicalreporting.com/blog/2025/9/...
Someone else proposed this use case but, to my eye anyway, bars would be considerably quicker and easier to read:
I avoid using radar/spiderweb graphs.
Find out why and what chart types I use instead in my latest post: www.practicalreporting.com/blog/2025/8/...
Really excellent. Nice work, @shrikhalpada.dev .
For a while now, I've been trying to think of ways to visually and viscerally communicate just how ludicrous the wealth that the top 1% has accumulated is. This one (not mine) is pretty good...
Just heard Musk say that xAI's latest benchmark testing proves that "...Grok 4 is better than PhD-level at everything...".
I guess PhDs don't know that showing bars on truncated scales is a bad idea.
While I (thank god) haven't had to do a lot of hands-on work w Power BI, I've seen enough to know that it would be soul-destroying for anyone who cares about making effective charts.
I'm sometimes asked to adapt my courses to Power BI's limitations; my answer: www.linkedin.com/pulse/why-wo...
So, all dashboards should be ignored?
Imagine receiving a letter that threatens your country's entire economy, and that also contains so many grammatical errors and poorly constructed sentences that it would get an F if it were submitted in a high school English writing class. Just... surreal.
"How to rescue your failing dashboard" webinar recording now available! My 47-minute interview with Ian Stuart of Rupe Ltd covers the top reasons why dashboards die, and what you can do about them.
youtu.be/xL7qHCy-CEE
𝗛𝗼𝘄 𝘁𝗼 𝗿𝗲𝘀𝗰𝘂𝗲 𝘆𝗼𝘂𝗿 𝗳𝗮𝗶𝗹𝗶𝗻𝗴 𝗱𝗮𝘀𝗵𝗯𝗼𝗮𝗿𝗱 😬
Unfortunately, many dashboards end up under-used or (worse) abandoned entirely. Why? I'll be covering the top reasons in a FREE webinar at noon BST Jun20. Hope to see you there!
Pre-register for the webinar here: events.teams.microsoft.com/event/4908a1...
Nice. I have yet to find a good use case for radar graphs, apart from some highly specialized applications in which the audience thinks of the data as "circular," e.g., wind roses.
𝗛𝗼𝘄 𝘀𝗵𝗼𝘂𝗹𝗱 𝘆𝗼𝘂 𝘀𝗵𝗼𝘄 𝗢𝗨𝗧𝗟𝗜𝗘𝗥𝗦 𝗶𝗻 𝗰𝗵𝗮𝗿𝘁𝘀?
Check out three common-but-not-ideal methods and one good one (IMHO, anyway) in my latest post: www.linkedin.com/feed/update/...
Inset charts are my favorite #dataviz trick🔥 Many chart creators aren't aware of them but I use them all the time. What are they? Find out in my latest video (5 mins): youtu.be/n1RCV-NMct0
BTW, registration is now open for my Jun2 live online workshop! www.practicalreporting.com/june-2025-on...
Agreed. I cover this situation in my "Practical Charts" course and book:
😱 Don't make these mistakes when showing missing values in your charts! 😱
Are there missing values in your data? Many charts don't show these properly, which can confuse and even mislead readers. How can you avoid this in your charts? Check out my latest post: www.linkedin.com/feed/update/...
The dumbbell version would address the "steep slopes" perception issue, but it adds visual busy-ness and I'm not sure it makes the deviations between both series any more salient. The line chart also communicates that one series was always lower than the other at a glance...
The discussion and engagement was so good on this over the last few weeks that I went ahead and added 2 whole new sections to this blog post (worth checking out if you already read this!):
I reflect on the discussions and fun that we had as well as provide more resources and a call to action.
"𝗘𝗰𝗼𝗻𝗼𝗺𝗶𝗰 𝗖𝗵𝗮𝗼𝘀" 𝗦𝗮𝗹𝗲 𝗼𝗻 𝗡𝗼𝘄! Until May 20, get 𝟮𝟱% 𝗼𝗳𝗳 on...
👉 Upcoming live online workshop (kicks off June 2)
👉 Practical Charts On Demand (6.5 hrs of video in 45 lessons)
👉 Practical Charts + More Practical Charts books (PDF+eBook editions)
www.practicalreporting.com/blog/2025/5/...
Love this.
(...though the minimalism-fascism connection feels like a bit of stretch, IMHO.)
@duvekot.bsky.social I see your point but, again, pros and cons. If you only show deviation vals, the scale of the "lag" is lost. In a chart like that, it could be that, e.g., in the grand scheme of things, Boeing is only lagging Airbus by a tiny amount relative to total sales, or it could be huge.
▶️ New video (4 mins) ▶️ Should you put key insights in chart titles, on in annotations directly in the chart?
I've recently been leaning toward the latter, but would love to know your opinion in the comments.
www.youtube.com/watch?v=OU3s...
@duvekot.bsky.social Also a viable option, yes. No color associations needed, but the trade-off is that the chart is visually busier with each category name listed four times (and yet busier with all 20 quantitative values labeled instead of a quant scale with 4 or 5 value labels). Pros and cons...
▶️New video!▶️Why I don't use colored category labels in my charts (4 mins)
You've probably seen charts where the colors in the chart are identified by coloring category name labels instead of using "traditional" color swatches. I don't use this technique in my charts, though. youtu.be/RLdsgluOqnM
@mjskay.com @frank.computer
🤣 Here's my version:
I wrote about the data-to-ink ratio and a new game I've invented called "ink golf" for data visualization.
I hope you enjoy:
www.frank.computer/blog/2025/04...
⏯️New video!⏯️ Watch me redesign a chart so that it "does its job" (10 mins) youtu.be/QnqpuGeKIxw
BTW, live online workshop kicks off on June 2 with early-bird prices ending May 9! Info/registration: www.practicalreporting.com/june-2025-on...
So, basically, screw small businesses.
😱Unpopular opinion warning: 😱 GenAI has made shockingly little progress in the past 24 months when it comes to many/most *real world* tasks: www.linkedin.com/feed/update/...
Congrats!