Clocks and pies
Oct 28, 2007
Keith A submitted this graphical idea from the folks at Ikea (via Boing Boing).
Based on the comments, it seems like some people really like this presentation!
Consider these for amusement:
- Does the "9" on Sunday mean 9 am or 9 pm? (This chart mixes A.M. and P.M. hours in a totally nonchalant way.)
- If the above is too easy, try the "9" for Saturday!
- Why was "9" displayed on Sunday anyway? Meanwhile, why wasn't "7" displayed for Saturday? (How were the hour labels chosen?)
- Why was "Closed" written on the chart while "High", "Mid", and "Low" were put into the legend?
- Since pie charts show proportions, is it possible to describe what proportions were plotted?
Reminds me of this pie chart.
Usually your analyses strike me as very good, but I've got to comment on this one because I just don't get it.
This chart actually strikes me as quite a good representation of a complex subject in a compact and fairly clear format. But perhaps I've just been brainwashed by IKEA since I was there yesterday : )
In answer to your questions:
- 9am/pm means both am and pm.
- Same answer : )
- 9 is displayed on Sunday because it is central to understanding the chart. It is the division between am and pm on the first two charts, so it is important that this navigational feature not disappear from the third chart.
- "Closed" is probably the most important chunk on the chart, so it is highlighted and clearly segmented by hour markers. It I show up at Ikea at 5:30 on Sunday (30 minutes into the red segment) that's not a big deal. If I show up on Sunday at 9:30 (am), or 8:30 (pm) that is a pretty big deal from a customer experience standpoint. Unless I want to get breakfast at the restaurant . . .
- Yes, the proportions plotted are time Closed vs. Light Traffic vs. Medium Traffic vs. Heavy Traffic during the 12-hour period between 9am and 9pm every day of the week.
Thanks for the site!
Posted by: Ethan Jewett | Oct 28, 2007 at 10:20 AM
I agree with Ethan. That chart does a pretty outstanding job on conveying both store hours and their expected volume.
The target for this chart isn't a statistician, but there customer. I can picture what it would look like if they had the full 24 hours on there and it's pretty useless for their purpose.
Posted by: Rick B | Oct 28, 2007 at 10:36 AM
I like this one a lot better.
Posted by: Filip Salomonsson | Oct 28, 2007 at 10:44 AM
The target for this chart isn't a statistician, but there customer.
There's just no need to make cracks like that. The point of competent information graphics is to make things simpler for the customer.
What's wrong with a Gantt chart?
I have assumed the store opens at 10:00AM and closes at 9:00PM on Saturday because, despite Ethan's answers, the opening and closing times are still an unknown quantity for Saturday. Note I have also invented more complicated volume profiles for Monday to Friday, and for early weekends, simply to show that a better graph can, without losing clarity.
Posted by: derek | Oct 28, 2007 at 12:18 PM
True that different Ikea stores have different hours. I'd imagine that this one was open 9am - 9pm on Saturday. Some stores are open 9am-10pm on Saturday, in which case this pie/clock approach wouldn't work very well.
It would be interesting to know where this photo was taken.
Posted by: Ethan Jewett | Oct 28, 2007 at 04:06 PM
This isn't a pie chart, but merely a circular graphic to mimic the passage of time on a standard analog timepiece. There are no proportions being plotted, other than the proportion of the half day between 9 am and 9 pm.
I would keep the pie chart styling, since it does nicely resemble a watchface (the Gantt and tabular approaches require a bit longer to figure out). I would use better labels, like 10 am, Noon, 3 pm, etc. On Sundary I would label 10 am and 8 pm, but not 9 (am or pm). I would put a heavy line at the border of closed and open, so if the store opened at 9 am and closed at 9 pm, a dark line at 9 would indicate this. I would also put High, Mid, and Low labels right into the chart.
Posted by: Jon Peltier | Oct 28, 2007 at 06:38 PM
On the one hand, I don't think it's as bad as you suggest - I think the fact that it runs from 9 to 9 is pretty easy to work out. On the other, I don't think it's very intuitive either: I don't associate the right-hand half of a clock dial with the afternoon, for example. Clock faces with hands are easily understood; I'm not sure they are when they're solid blocks. Perhaps simply labelling the start and end of the sectors with their hours might make it clearer.
Posted by: Tom | Oct 28, 2007 at 06:46 PM
At first look, I didn't think the pie was so bad, but after seeing the Gantt chart I have changed my opinion. I find the Gantt chart much easier to understand (and easy to compare different days). I like the idea of using a clock/pie chart, but the times are confusing (ie take longer to understand the the other charts).
Posted by: Andrew | Oct 29, 2007 at 10:03 AM
The Gantt solution is really clear. This makes me think that there is more information that would be useful, like what kind of shoppers are shopping when. Rick B's link suggests that it is businessmen only, but when are the single women there? What about families with kids? Gay couples?
Posted by: Chris P | Oct 29, 2007 at 11:27 AM
The one Filip linked to from a different Ikea store is a nicer and simpler variant of the Gantt chart. I like that one the best.
Posted by: Kaiser | Oct 29, 2007 at 11:01 PM
I think it's useful to take the context of this sign into consideration. Here's a charitable interpretation: This sign is posted somewhere visible, near an entrance or exit. Store hours are inevitably posted nearby, so there's little need for that on the chart. The sign will answer two questions for customers. First, have I arrived at a busy time? It's reasonable to have a presentation directly tied to the most common representation of time that people carry around with them: a wrist watch with an analog face. Second, if I were to come back again, when would be a good time? Assuming that people are creatures of habit, they may think, "Oh, if I got here an hour earlier or an hour later, that would be better."
Or maybe I'm totally off-base. Still, a graphic analyzed in the abstract may seem terrible, but it might make reasonable sense given the viewer, the physical environment, and the activity.
Posted by: RSA | Oct 30, 2007 at 08:25 PM