« The trouble with two lines | Main | Spinning the climate »


Jon Peltier

I guess they missed the memo that area is a poor way to encode values. Color's not great either, but at least they used a discontinuous scale. Very discontinuous, as you've discovered.


The Times has a graphics team that at least sometimes shows extraordinary care and cleverness in presentation of data. WTF happened this time?

Michael Pierce

Wow, what a mess. The change in area is so hard to perceive and they aren't even consistent in the direction of the growth/reduction. Notice Kohl's, which grew, actually shrunk on the front edge, yet grew in the other direction. Cinemark, on the other hand, grew the opposite direction. Crazy.

My other question: why these companies? Are these the 27 largest companies (don't think so)? Or did they "randomly" pick most of the worst performing companies in order to make their point?

Funny thing, many of these companies don't even operate in a "mall" setting, which makes the metaphor all that more forced.

In the end, a bar chart would have painted the picture much more effectively. Heck, even the table of data enhanced with red/green indicators would have been a better choice.

Joe Mako

I would like to know why some cars disappeared, why mostly ones in the center, and does the color of the car mean anything.

If they were going for a mall metaphor, why not group the restaurants into a food court?

Alejandro Sacasa

I was also surprised by this incredible waste of space, and on the back page of the Sunday Business section! The chart also didn't help to explain any differences between businesses in the same industry. Jack In The Box lost sales while Burger King went up for example.


Harlan: to give credit where it's due, this chart is by "LEFTLOFT, a design firm"

David Gorsline

The callouts to indicate the companies, using the visual metaphor of flags on pins, float high above the plane of the chart, confusingly so. And the meaningless "ball" of each pin is too strong relative to the flag and the shaft, causing the reader to believe that it's the pointer to the company's colored area. Look at the callout for "Build-a-Bear Workshop," for instance.

The only thing this graphic gets right is the change in the number of "cars" in the parking lot at the upper right of each panel.

Gerald Higgins

Wow, the more you look at this, the more awful it becomes. To expand on some points already made by other commenters . . .
1) Change in area is indeed hard to perceive. They could have made it easier to perceive by using simple shapes - squares or rectangles only, for example. Still not good, but better. What makes it even worse is that some stores actually CHANGE SHAPE between the two views. E.g. McDonalds changes from "L" to rectangle, Build a Bear Workshop rotates through 90 degrees, Big 5 Sporting Goods and Cinemark both seem to expand their frontage across their neighbours - incidentally suggesting that they are both growing - Cinemark is but B5SG is actually SHRINKING, just not as fast as Gap.
2) The Pushpins are indeed awful. What makes them worse is that some are poorly positioned so that they seem to be planted on the edge of their stores, thus raising doubt as to whether they are actually meant to be in that store or not, for example McDonalds and Wet Seal. Also, to make it even more difficult to identify which flag goes with which store, there are several instances of the flag of one store crossing the pin of another store, e.g. Starbucks and McDonalds, Papa John's and Cheesecake Factory. Surely a few minutes' extra work could have eliminated all of these overlaps.
3) Does the relative size of the stores actually mean anything ? For example, Walmart's sales, according to the table, look as if they are bigger than ALL the other listed stores COMBINED. But on the chart, Walmart looks as if it has a similar area to both JC Penney and Kohl's, both of which were less than 5% of Walmart's sales in 2008. What looks like the 4th largest store in the top version, Saks, had less than 1% of Walmart's sales. It seems to me that if you wanted to use area meaningfully, it should either be related to total sales, OR, area in the top version should be equal for all stores, to get a better view of the change in size.

The comments to this entry are closed.


Link to Principal Analytics Prep

See our curriculum, instructors. Apply.
Kaiser Fung. Business analytics and data visualization expert. Author and Speaker.
Visit my website. Follow my Twitter. See my articles at Daily Beast, 538, HBR.

See my Youtube and Flickr.

Book Blog

Link to junkcharts

Graphics design by Amanda Lee

The Read

Keep in Touch

follow me on Twitter