- Last post
- 16 Responses
Is there such a thing as overusing too many icons?
I am trying to build a case to convince these guys that they have too many icons on the screen (22 at a time). And I'm not talking about the general visual language representing a "close", "delete", or "submit" action. They want to abstract stuff like "Awaiting check generation", "Acknowledge trip", and other stuff impossible to guess unless you rollover to read the tooltip.
The initial layouts were simple and clean, with a handful of icons but with no more than maybe a dozen at a time. But they are just bending it to their will and it looks like the cockpit of an airplane.
Sorry, but I can't show any screenshots. I am going bald from pulling my own hair.
What would you do? If you can think of an article that explains why less is more, that would be great.
A few things I would ask myself:
Do the icons mean anything to anyone?
Do the icons help users better understand the label/category?
I've heard of studies that say icon combined with copy is the best. Even if it's not a super easy icon to understand once a person relates the two it can help find an item in a list quicker.
I'd say icons that have no reference without interaction are more of a hinderance. Who's going to hover or click on every item to find waht they need?
Unless icons are ubiquitous (e.g. the refresh icon or "x" for close), they need text labels.
You simply say, "Let's test out"
You can’t answer this without knowing who the user is, their level of expertise, and their level of familiarity with the subject matter and your design verncular. Then you can temper that with the limits of cognitive overload for those users.
icons are harder to figure out than text labels on first time use, but are better recognisable for returning users
The problem is having too many icons together on the same screen. Labeling them is not the solution. I am just trying to convince them that if they keep adding more to the bunch it will just confuse the user even more.
It sounds to me like a the client wants a clusterfuck of nonsense. I've always felt icons are only for secondary or tertiary navigation/utilities (close, log out, refesh, et cetera).
My recommendation to you would be to try and find a still-live Geocities website from 1999 or so, one that's loaded with spinny chrome balls and skull animated GIFs, rainbow headlines, blinky text, the <marquee> tag every third line, a dark paisley repeating tile background, and self-playing Midi files ... and then simply tell Dear Client that he's asking for the 2018 version of exactly this ancient, unusable monstrosity.*
*Bonus points if it has bevelled horizontal rules, and content set in tables without the borders turned off.
This sounds like a functioning admin tool, rather than a basic website that happens to feature icons, right?
if so, what's the user profile, and will they be using it frequently? If so, I think Icons and a small learning curve is tolerable.
After all, the programs we use each and every day rely on them, don't they? I know that each new piece of software I've come across in recent years (thinking Rhino and Fusion360 in particular) has been a confusing mess when I first set eyes on them, then familiar and functional after practice.
Show the client a different clusterfuck of icons and ask them to move the plumbus into the rectotherb so the samoflange hits the greebus limit.
A little bit of context: I work for a company that serves as the middleman between insurance companies and dealers that buy totaled vehicles.
I'm in charge of re-designing their enterprise software. The applications their employees use to deal with the entire process (towing, dismantling, auctions, etc...)
As long as the business team is happy with their numbers, they really don't care how the apps look. Their goal is not winning awards for usability or innovating user interaction. They just want to see their sales increase.
- So you don't really need to do UX, you need to do MeX by just adding more icons and say, "Looks good to me!"robotron3k
- I'm sure productivity and avoiding human error are important to them since both affect the bottom line.mort_
- Oh. Well, then by all means, add a dancing banana or two. Or a dancing Spider-Man.Continuity
- ^ yesMaaku
- ^ That yes was for mort_, lolMaaku
You mentioned 22 icons on a screen at a time.
How the elements are grouped would be the thing to watch, rather than the overall number of icons.
e.g. if all 22 are in a single long navigation list, that list is too long regardless of whether there are icons are not.
But if say the navigation has 8 - 10 links, and some icons are on buttons or headings and some are status indicators and some are part of a right-hand-side column, 22 could be perfectly fine.
They say average person can hold between 4-7 distinct items in their short-term memory at once. (High IQ skews higher here, IIRC). Which means, if your set of choices exceeds that number, it will be more of a cognitive load to parse the list - it won't feel easy.
Grouping items can help. Familiarity can help. But in general, if you have a system that people work on infrequently, or that is used by a broad swath of people, then you'll benefit from fewer choices at a time.
For software where a specialist is using it constantly, and that person becomes familiar with it, a larger number of items can be visible at once in order to reduce the number of actions a user must take to initiate a task (increasing efficiency).
Keyboard shortcuts are a further step beyond that, and are usually the mark of an expert user. A novice won't use them.
Thank you to everyone the took the time to reply.