UI redesigns are mostly a waste of time (2019)
debugandrelease.blogspot.com
debugandrelease.blogspot.com
They are very effective at getting the executives who sponsor them promoted, because they look great in a PowerPoint. The actual customer of an executive is the finance department, because they are the ones who actually control the headcount & budgets that an executive needs, and they are usually not the user. How the product works is often invisible to someone who is not a daily user, and financial analysts often don't have time to be power users in all the products they are evaluating. How it looks is very visible. Therefore, when deciding which projects to fund, they are often biased toward shiny.
Notice that founder-run companies, and ones that are too small to have normal corporate functional separation, usually don't invest in visual redesigns unless the existing UI is really bad. Big corporations do it continuously.
No UI is really intuitive enough that there isn't a formal or informal learning and even training cost to it. Then the redesign invariably finds a new way to hide the controls (hamburger menus, ... menus, gosh knows what else).
>And with Calc, that’s exactly what happened: Massive technical improvements. No visual improvement. And nobody noticed. In fact, the complaints just keep coming. “Look at Calc, same as it always was.”
> Anyway, my point is that—whether you like it or not—if you don’t change the UI, nobody notices. That’s why so much effort is spent on new UI.
I think this is wrong on two counts. First, if the change you made to the business logic really improved the things that users value, they'll absolutely notice even if the UI doesn't change.
Second, that comment implies that one way or another, a user-noticeable change must happen. It could very well be that the software is ideal for its user base as-is and no change at all is preferable to change for change's sake.
To quote the users: "but why should I pay for bug fixes?"
User-noticeable change is essential.
I also run a site and when I ask people what I should improve next, nearly no one says design, even when the design is clearly not very good.
An example of a site that never changes is Craigslist. I can't even recall the last time it looked different. It's probably overdue for a redesign even.
Of course their last redesign was so bad that they are already working on a new one: https://sh.reddit.com/. It seems better so far, but lets hope it ships before they turn off old.reddit.com
That's the system working as intended. Why would they invest in making the site better on smaller displays (i.e. mobile) when their priority is to get you to use the app? From the normal not long term enough business perspective, doing so could only cannabalize those conversions to the app that would have otherwise happened. And conversion rate from web to app is almost certainly a explicitly tracked metric.
All UI changes bring some amount of pain to existing users. People will accept that pain if the benefit of the change is greater than the cost, no matter how large or small the change is.
The thing that's become common is for UI changes to happen that bring little, if any, benefit to the user.
The ideal engineering solution, even for UIs, should be something that is abstracted properly, written only once, and then accumulates new features without any friction whatsoever. The new features rest on top of the platform seamlessly, giving users more options with no downsides.
A good UI, imho, is one that users can configure, it only adds new things and never take out existing features, and is robust, reliable and responsive. This is a serious technical challenge and most organizations are not setup in a way to solve this problem properly without the solution deteriorating and needing a rewrite... and yes in part it's because on the business side there is no perceptin that this is actually needed and so it is deemed easier to just rewrite and build something new every time.
It is better if UI works good out of the box. Of course, it is impossible to adapt to everyone's preferences, but configuration should be not too complicated, otherwise most users will not configure anything
Citation needed.
This isn't specific to UI redesigns, or even to software in general. People have been doing things "just because" since before the discovery of fire.
The author asks about smooth redesign and gradual deployment. I guess like google docs do.
I am curious, what car model and from what year this gentleman uses? I hope everything around him is only made by following his advice and principle.
There is a problem in UX/UI design. Twofold problem.
Developers hate to work on frontend implementation, and designers have mediocre knowledge about their craft and refuse to think objectively.
Both of the problems can be solved with actual effort in education.
Instead, big corps are wasting money and energy to replace all of them. AI will solve all of this by removing capitalism, human expertise, and design. Right?
Sidenote. Today, I have tried seven times to update my Mac Mini. After hours of fight with Apple logic, I give up. What's going on there?
I know a lot of developers who strongly prefer front-end work. I know a lot that hate it. A bit like how some developers prefer greenfield development and others prefer maintenance work.
I don't think it's supportable to make a blanket declaration that all developers have the same tastes or interests.
Let's make a poll. Why, when CSS is advanced as never before, devs are using Tailwind?
There's very little productive new ground to grow on if you're working on YouTube.
There's existing glaring site problems, but they're on purpose and ideologically driven (shorts, placebo dislike, subscriptions mean nothing) so no one is going to touch them. It's politically safer and more advantageous to mess with the UI and irritate all the users every couple years.
YouTube know that the choices they've made are qualitatively bad because users overwhelmingly complained about them (dislike button being the most obvious example of that)
...but quantitatively numbers didn't go down enough to force any kind of retraction or even recognition of discontent, so now we're stuck with the results.
So yeah sometimes the path of least resistance is building something new.
In comparison, I'm still using foobar2000 and Craigslist 20 years later.
Other things like non-monetized blogs can stand being retro.
I'm sure PMs and Data Analytics teams have actual data to back me up on this, and if the data shows to the contrary please correct me. I'm working with a small subset of friends and family who are, for the most part, your "average" user, so it's all anecdata for me.
On the flip side, stop fucking with messaging app UI.
I don't care about a splash of CSS to make it look fresh - just leave the rest of the UI alone, please.