There are not much threads about improving data visualization skills, so I decided I open one here.
Does this diagram explain to you what the Apdex is?
What is unusual or confusing here?
What would you change about it?
Interesting! I don’t think it explains what Apdex is, but it’s not hard to figure out from what it shows.
My question would be, why does it include half of Tolerating when it’s the 0.5s Apdex? From the bar chart, it looks like 0.5s is the cutoff for Satisfied, so I wouldn’t expect anything else to be included here.
The logic here is to see how many requests have passed with decent ping. So to get 1.00 or 100%, all requests should be in satisfied bin. I would write the formula as:
\frac{satisfied + 0.5*tolerating + 0*frustrated}{satisfied + tolerating + frustrated}
So half of tolerating is just to get 50% penalty while summing to get the final score.
What I think could make it more clear is to draw time spans on the vertical axis, but I am not sure that is a big improvement.
I mean, I’m in the industry (but I imagine most potential customers are too), but I would want to know if you are measuring the time until the response begins (TTFB) or something else. If TTFB, I’m not sure I’d be convinced that the user is satisfied - there could be considerable delay between first byte and first paint.
That graph is from updown.io – Website monitoring, simple and inexpensive which only measures the response time from the server, to detect if it is down or not, but time-to-render is a valid concert too to keep users satisfied. In fact Apdex seems to be a universal metrics for that Apdex - Wikipedia
I like the design that updown.io uses to explain the concept - lines and colors, but the logical part of the diagram is hard to read, so I wonder if it can be improved. Like, what professors in visual communication would do with it. 