1

I have make a graphic where I need to show user trends.
I will call them A, B, C, D -- brands of cell phones.

My users are not very good at reading charts.
Are there ways to combine features of chart 1 and chart 2 to make a more intiuitive graph?

Version 1

This is your standard line chart, the height representing the 4 different variables. Here I emphasize the four variables are separate and independent.

enter image description here

Version 2

The variables are represented by the space between the curves. Here I am emphasizing that A, B, C ,D contribute to a whole.

enter image description here

  • 3
    Why would you not want to use version 2? What do you want to achieve with a combination? If your "users are not very good at reading charts", is a chart the right thing to use? – peterchen Mar 25 '14 at 16:50
  • @peterchen I am worried version 2 is misleading... features of A spill over into the other three lines. – john mangual Mar 25 '14 at 18:33
1

For user trends, the stacked graph seems to be appropriate, because your different data share the same unit, and the sum of the values is a meaningful value (total number of users). It makes it slightly easier to express both changes in percentage and total number, although at the cost of direct comparison.

However, some quesitons remain - see my comment.

1

Do you absolutely need to emphasise that the variable is continuous?

I'd suggest displaying version two as a stacked column graph if you can get away with it. Humans are much better at comparing area changes in one dimension rather than two at the same time.

You might be able to overlay a line graph over the stacked column graph to get the effect you want, but that is likely to be very difficult to decipher for an audience that isn't used to viewing graphs.


Realistically, if it were me, I'd use two graphs.

mockup

download bmml source – Wireframes created with Balsamiq Mockups

One line graph like your first one to show your trends over time, and a second column graph, with only one column (covering the entire date range), showing the split between the different brands.

If you're presenting to marketers, I might even shudder consider replacing the second graph with a pie chart. Since that's the format they're used to seeing when they're comparing a part-to-whole relationship.

In either case, I'd lose the vertical gridlines. They're a large part of what is making your graph hard to decipher.

For learning to present this kind of information effectively, I cannot recommend this book enough.

  • I think that was an awesome reply, @Racheet ! I don't visually like too may graphs ; it offends me! – Jean Mar 18 '15 at 11:40

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.