How We Lost User Engagement After a Redesign
icons8.com
icons8.com
Under that model of user expectations, I envision two groups of users: 1. the users who want to casually suggest an icon (and don't need it for anything critical or want to pay for it themselves, but who would happily use it if it became freely available) 2. the users who have a special request and would pay to have it done, regardless of how many other users want it.
Group number 1 won't bother clicking on "Request an icon", because they figure that the button is for members of group number 2. Group number 1 won't see an option for them, which changes site perceptions, because group number 1 has an impression that the site is for paid requests.
Group number 2 will click on "Request an icon", hoping to place an order, but will be disappointed and/or confused by what they see. Members of group number 2 might note that "Oh hey, it'd be nice if I could get this for free". But given that their need was sufficiently critical for them to want to pay commission, the option to submit an idea that may or may not be voted into existence at some unknown time won't seem promising enough for them to invest any more time in the matter.
That one word choice would have made a great deal of difference to me as a user understanding the feature. Yes there are glaring design issues with where they went as well. But that one choice of word is huge.
Back in the 2007-2010 era of websites, somebody mentioned that "ugly" websites work. Not that fancy or flat design is bad. But the sheer simplicity of ugly websites forces people to focus on the content, context and functionality. The ugly websites pointed out were MySpace and Craigslist mainly, but you could probably add a few others like Drudge Report and Plenty of Fish. Even Hacker News is pretty "ugly" but it obviously works great.
Reddit got popular when Digg was redesigned.
There are actually a few etymologies for the term in architecture. It does often imply a deliberate crudeness and a harshness, supposedly an honest response to the character of the materials. To the untrained eye this is close enough to "ugly" that most people don't distinguish. There's also a very strong argument that brutalism represents a turn away from the values of western humanism (typically expressed, historically, in classical architecture) toward a vulgar, bleak conception of modernity. Another aspect is that it tends to be "sublime" rather than "beautiful", which, again, translates to negative vibes for many people.
It's not a term that I'm happy to see being applied now to some essentially cosmetic aspects of web design.
I share a similar peeve as you with the use of "women" as an adjective. All through grade school and high school, the use of "women" as a noun was pummeled into my head. Now, I routinely see presumably college-educated journalists use it as an adjective ("women scientists", "women engineers", etc.)
If you would now go ahead and use it to describe a concept of a different scope where it does not apply at all, say web designs, you could do that, but you would expose that you just tried to name something in a sciency way to sound sophisticated without having any idea what the term actually means.
Indeed, this is by far the usual case. People collectively can, and do, reject almost all proposed novel usages, which are quickly forgotten. We just remember the few cases where things change due to survivorship bias.
"People" aren't really a useful measure here because there is slang that achieves popularity out of the subculture that produces it, subcultures composed of people, just the same as it may not propagate, in which case it naturally remains useful to the in-group. To reject slang as valid language until it achieves mainstream acceptance is a) not measurable as there is no line that is crossed, and b) antisocial.
Anyway, I came up with a better definition: accepting a word/usage is believing that it's a good way to express whatever it was.
I myself will never use "literally" to mean "very," which differentially lowers the chance of this new meaning becoming standardized. Over time, either most people will do the same and the proposed novel usage of "literally" to mean "very" will die out, or else most people will collectively adopt the new meaning.
Empirically, almost all novel usages die out very quickly and are forgotten. Read some slang from the 1920s -- it's already almost incomprehensible to a modern reader.
As for the term itself, there's always a risk of imprecision when laypeople discuss the concepts used in certified professions.
There's a 2004 paper along the lines you mention by Simon Yuill: http://www.lipparosa.org/essays/code_brut.pdf
Edit: Somebody should write a manifesto about it.
What am I trying to communicate? Can someone easily get that information from my site? If there is interaction - is it easy for the user to figure out what to do?
The questions that actually get asked when designing a website:
Does it feel modern? Am I using the latest technologies? Would I feel comfortable telling my buddies how I created the site?
I have nothing against modernizing a site, other than the fact that it's not that important, and it's really easy to screw it up.
http://www.yale.edu/fin-bus/resources/docs/SurvHandbookChap3...
Yes, and they were wrong for any meaningful sample of websites. You can point to ugly yet hugely successful websites but that ignores the millions of ugly, failed websites that have long since stopped serving visitors. And equally, there are examples of beautiful sites that work and beautiful sites that didn't.
How good a website looks has a short term impact on how it's perceived, but great content (where "content" has a lot of different meanings) is always more important. That's true regardless of how much effort you spend on design.
"aesthetics is orthogonal to usability". not "all ugly websites are usable".
Make something as simple as it needs to be, but no simpler. Most designers I see these days seem to forget that last bit.
In this specific example the whole purpose of the request system is to vote and secondarily get other people to vote with you. The new design eliminated almost everything that made the old one useful. It isn't simpler, it's simplistic.
People are copying the pointless "elegant simplicity" signalling of the new Apple Macbook. I think this is probably one of the best filters you can find, if you are perceptive enough to be able to use it: See whether or not someone prioritizes signalling over actual functionality/semantic content/logic. Observe how often this happens. Compare the effort that goes into signalling versus more substantive observation and analysis.
If someone isn't self aware of their signalling, and hasn't differentiated that concept from actual value/competence, quietly disengage.
My reaction was Nope, roll back to the previous version.
The new interface feels very unconventional, in fact. It may become convenient once you've learned what everything means, but until then you'd feel puzzled.
~~ Of course, I could not try to work with the old site, but the old screenshots looked more informative than the new ones. Not surprised at all by the results.
It's hard to imagine something simpler. HN even maintains a good information density.
One obvious example being Amazon.
As a consumer I will generally go with whoever gives me the most information about whatever I'm doing/learning/buying with the least clicking around. Indulging in the modern "design" trend of giving me 5 million white pixels is not how you sell stuff to me. I'm not looking to buy white pixels, I'm all stocked up on those.
That surprises me, since when I visit Amazon's UK homepage my fan kicks-in and there's a noticable delay before I can start typing-in the product for which I want to search; all so that they can load an enormous Prime promotion carousel and rows upon rows of 'product suggestions' in which I have never shown any interest. Today it includes pearl ear-rings.
OT: I really miss Yegge's writing. I wish he publish more.
"To be honest, the graphic below is pretty much useless. I just think it’s pretty, so it’s staying in."
"I was FORCED to conduct usability studies"
That medium gray text is harder on the eyes.
Am I required to 'share' my icon in order for it to be made, like on of those "Like on Facebook for coupon!" sites? Where can I share it? Can anyone vote for my icon, or only people who see it from my 'share'?
I would suggest that simply "Free" would be a much clearer, with the social/share component available but not required. Then, if they want more people to vote on their icon, they can choose to share it or not.
"Free for share" makes it sound transactional, like I'm paying for the icon with a 'share', which is a concept I absolutely hate.
Okay, back to finish the article...
Also, that entire page was impossible to parse. I had to go to their main webpage to even try to understand the concept of what they're doing ,and try to put the page snippets in context. Page snippets on a white background, placed over another white background, are confusing as hell.
I understand in cases where it's a big switchover and basically a redesign riding the coattails of a huge middleware/backend rewrite. I don't understand in a case like this when it wouldn't break major back office processes to just put the old site back.
The new page is just confusing. What am I supposed to do on it? Nothing looks like button or a hyperlink.
They specifically mention that they did not lose any traffic/visitor numbers: "Important: our overall traffic didn’t change. The same amount of people came to this feature’s page."
It appears as though the "lost users" were really lost interactions with the vote button: "our voting numbers decreased by ~50%."
It's fascinating to me how far people are willing to go when skewing the information to pique interest. I'll bet a lot less people would read it without "lost users" in the title.
If you had lost 50% of your users, you would have half of the overall site traffic that you had last time.
Let's take HN for example (I don't know if HN considers this their metric or what, just using it as an example.). Let's say they made a change that now requires everyone to login to the site each time they visit. So if you closed your browser window or were inactive for a period of time you would have to log back in. While you could still read posts and comments, you would be unable to submit new stories or comment until you logged in. While you are not logged in you would be considered a visitor. But when you log in, you now become a user.
While at first the traffic, or visitors, may remain the same, without participating users on the site the number of visitors will eventually drop as not as many people are submitting or commenting.
A site that does not require interaction, may on the other hand, simply use the two terms (visitor & user) interchangeably.
Specifically, did solving the issues identified restore traffic back to normal levels?
It's just me, who sees some drawbacks of this drop in the future. For others in my company, it may be just 20 votes reduction.
Even if I already fully understand their system and are highly motivated to vote, I wouldn't know how to actually do that in the new design.
In the new design, you have to hover over one of the numbers (without there being any indication that hovering does anything) - after which a grey up-arrow without any text appears. If you're familiar with reddit, etc and know that particular language of symbols, and also know this is a voting system, and you're not distracted by the gray color which could indicate "action not available", you can make an educated guess that this is likely the vote button.
In the old design, it was a button with the word "vote" on it.
Was there any sort of A/B testing done beforehand to see if the changes would bring actual improvement or was it more just trying to mimic current design trends?
One thing that would have been cool to see your writeup would be how you used direct customer feedback. Maybe it was none at all, but I suspect talking to customers (or even internal employees) helped you reach some of the conclusions you made. It's really difficult to look at a graph and go, yes this is the thing that caused the change.
- user generated requestst - crowdsourced voting - how easy submitting a request is, and what it entails - the volume of requests you have
I see this in the larger context of icons8 and I get the sense that is akin to voting for a feature. This is something I could do, if I wanted.
The new design shows me:
- Weird titles that may or may not be some sort of request
And that's about it. Only five of them. Take that in the context of icons8 and I really can't parse what it means so I would probably just completely ignore it. On the request page, I don't see how approachable the request feature is, so I ignore it.
Uservoice does it better.
UserVoice has no doubt put a tremendous amount of effort into ensuring that users interact with the support system as much as possible. Not very smart to assume that you can re-do all of that work yourself.
Full disclosure: Happy user of the UserVoice system myself, grandfathered into the free plan but otherwise no relationship.
Went from being Simple left to right, top to bottom style to some weird layout that it very hard to navigate.
I would give that sucker a grade of “F”.
Great reminder to have users test products before shipping something.
As well as use common sense in your designs.
[1] http://kare.com/
On the one hand designs tend to be overcomplicate - any design that is noticeable is overcomplicate - and on the other hand every 2 or 3 years people feel a urge to fix things that work.
Yes Google might not have chosen poor words in this case, but it does create friction when you switch around the whole UI.
The example I always cite is when Microsoft switched the whole UI in Office from the 2003 version to 2007 version. People who had mastered pivot tables were suddenly up the river without a paddle