IPhone Apps Design Mistakes: Over-Blown Visuals (2009) | Hacker News Reader

IPhone Apps Design Mistakes: Over-Blown Visuals (2009)

smashingmagazine.com

98 pointsacqq28 comments

To this day, Motion-X absolutely kills it with their "overblown" designs. This 2009 article is harping on an app that was dominant for the two years following its publication, and a design strategy that is still working for the company.

It's true that you can make an excellent UI using the standard components, and this also leads to other advantages like being able to adapt to new screen sizes easily. But to say Motion-X's UI somehow fails is a disproven hypothesis.

They went on to use the same kind of design approach for Motion-X GPS Drive, which let me check... yep, makes more money than Angry Birds in the US App Store, ringing in at 11th grossing app overall (not in the category, on the entire App Store).

Spot on, Smashing Magazine... you literally could not have picked a worse example. Maybe I should be quiet though, since we compete with them on some apps :)

How does the fact that the app sells well disprove the fact that the app is badly designed? I have both Motion-X apps, and I love their features, but I find myself using simpler apps because of the Motion-X UI clutter. Motion-X offers an impressive feature set for the price, but why on this earth do I have to deal with un-familiar graphical switches to change a setting? Where in the real, tangible world does having more switches and buttons make something better? I'd actually use the apps more than once if I didn't have to study the UI in order to make a task-related decision.

For example, take a look at a screenshot from Motion-X's Drive app (http://i.imgur.com/uWXdk.jpg). Compare that to the less-flashy Tomtom app UI (http://i.imgur.com/6mPKe.jpg)

Which UI is more functional?

And for run tracking, which UI would you find yourself more inclined to use, Nike GPS' (http://i.imgur.com/80pLl.jpg) or Motion-X GPS' (http://i.imgur.com/tTdMN.jpg)?

I never can figure out how to use the Motion-X apps. Tease me all you want, but the UI is a chore to me. I prefer the dumbed-down Tomtom app for navigation and the Nike GPS app for my runs. While they may lose, feature-wise, they win on functionality. Why should a user have to deal with artificial UI abstractions that add flair at the cost of increased cognitive load?

How does the fact that the app sells well disprove the fact that the app is badly designed?

Being well selling doesn't mean that it's not badly designed, it means there's a financial interest in not changing it.

Exhibit A: the fucking ribbon.

You can't make this claim without know how it would have done with the alternative design. All you can say is that it's design didn't kill it.

My latest iphone/iPad app would probably be considered as "overdesigned" by the article's standards.

http://itunes.apple.com/us/app/fantasy-football-draft-monste...

Do you know why? Because overdesigning SELLS. My iPad app is currently ranked top 75 overall and my iPhone app has been in the Top 100 overall. With 400K apps in the app store, having a unique looking UI will attract more people to try it.

I totally agree with having good UX and being consistent with the HIG, but the user can't appreciate that until after they've purchased the app. So if you want to actually sell a few copies, then suck it up and pimp out your UI. You'll be glad you did.

That would actually be quite sad. I personally can’t stand customized apps. I think they are a possibility in theory but they have to be extremely – and I mean extremely – well done.

The likelihood that somewhere something is screwed up or not as good as the default is just so high. I have found that my personal standards are nearly always higher than the developers’.

I can’t believe, though, that I’m typical in that respect so ugly apps that stick out may well sell better.

Based on my experience, your tastes are definitely not typical (at least of app store customers). Customized UIs are very popular, and in my opinion they are definitely not ugly. But opinions matter very little in the marketplace...

It’s not the customized apps themselves that are (to my eyes) ugly – though that may sometimes be the case – it’s the inconsistencies, the lack of polish and the lack of thought that usually comes with having a customized UI.

Sure, but my point is that buyers don't know whether the UI is consistent or not until they actually buy it. All they have to go off of are screenshots, and if they see screenshots of standard looking SDK components they'll gloss right over it. The app store is essentially window shopping, that's why overdesigning sells. So given a choice between pleasing UI snobs and actually making money, it's not a tough decision to make =)

EDIT: Re trials: most paid apps (mine included) have limited free versions that you can try.

Sure, I can understand that.

The app store really needs trials. I have more than once bought an app that turned out to have a horrible UI.

What matters most is distribution and solving a problem that customers have. It can be an advantage in a crowded segment to have an extremely well polished app for differentiation purposes. But without distribution any app is going to die.

In your case, your app seems to be distributing mainly because you have a customer base from previous apps, which boosts your inherent distribution (pools are created by groups of friends which will talk about your app) and your search results in the app store. Also, from what I can see you are the only one that supports both Yahoo and ESPN, which is pretty cool :)

I released an app recently. First, a quick prototype with a stock UI, then a "beautiful", custom UI. Changes in downloads = 0. Still, I'm not saying it does not matter, because it clearly can, but it's very hard to pull it off. In your case, we can't really tell, unless you released a version with stock UI? Hard to prove that your UI affected sales without a test.

Congrats for your apps.

> Do you know why? Because overdesigning SELLS.

Isn't that an assumption though? You haven't built two versions, one overdesigned and one not, and then A/B tested them right? It could just be that over factors besides design makes your app sell.

I don't think this is possible without jumping through hoops, releasing multiple versions of effectively the same app probably wouldn't make it past the review process.

I don't think your app would be considered over-designed - check their "police scanner codes" example.

Your app still uses normal buttons, and even the "tabs" are close to Apple's toggle-button-array design. Your search field is exactly like the OS standard. Compare that to the article's examples.

I would also point out that Apple is doing a lot more "overdesigning" itself these days. I think the address book in Lion or the Reminders iOS 5 app are evidence.

Agreed, I don't find it overdesigned at all, whereas I do think the example in the article is a bit overboard.

How do you know that it is the design that sells it? Did you A/B test with the native controls vs your controls?

No, it does not. It's what the application DOES that sells the app. I can almost guarantee you that you would sell more if you didn't have such an ugly (in my opinion) UI. I bought MotionX because it was only $0.99. A few weeks later I stopped using it because I could not stand to look at the UI any longer. Now I'm just back to using the Maps app that comes with the iPhone despite it not having turn-by-turn directions.

Sorry, but I just can't take one person's armchair quarterback opinion over hundreds of thousands of paying customers. If you have actual sales data to back up your theory I'm all ears but your opinion seems to be an untested hypothesis. I've A/B tested my UI many times and I have 16 months of sales data to back up my argument that app store customers love "overdesigned" UIs. This is not mutually exclusive with good UX btw.

EDIT: I realize there are less "overdesigned" fantasy apps that rank higher but one is sold by the NFL and the other by MSNBC (I think I actually outrank Rotoworld on iPad). Hardly a fair comparison from a marketing muscle point of view. I can only make judgments based on my own sales data. And all of it points to the fact that all else being equal, users are more willing to buy apps with interesting looking UIs.

The two fantasy football apps above yours in the iTunes store (Rotoworld and NFL Fantasy Cheat Sheet) are less "overdesigned" than your app. Doesn't this suggest that you're making a spurious correlation between the level of an app's sales and its level of "overdesigned-ness"?

The app store doesn't do demos so it is certainly not what your app does that sells it. How could it be, no one who buys your app has ever run it before. The only thing that sells your app is the advertising.

The author uses the fact that most apps are not used much after they are first downloaded to to justify his argument that non-standard controls are bad. But he fails to make a causal connection.

On the contrary, people will use apps if they are useful, and will learn any UI to achieve their goals. Does Angry Birds use any standard UI controls? Nope! But people play it anyway, because it's a great app. Standard UI controls would just make it ugly.

Decent points, but it's from 2009. Also, using 3D pie charts in an article about proper design... maybe not the most effective technique.

Glad the author used the Bloomberg stock finder input as an example.

I've been teaching my father in law how to use the Bloomberg app on his iPad and this was one of his biggest stumbling blocks - he just couldnt work out why the keyboard wouldn't appear.

With the adoption of these types of devices moving to include the "non-tech savvy" population, who aren't as quick to work out unclear UI, we start to see how small UI choices can have a dramatic effect on usability.

I agree with the principle of "over design" - we old guys call it "if it isn't broken don't fix it".

However, some of his examples aren't convincing for me. I actually like the ugly Motion-X toggle switches compared to the IOS toggles which are a bit confusing I find - I can immediately see what is active / selected on the motion-X screen but on his examples of the correct" design I couldn't immediately work out what is selected/active. Also, the Motion-X tabs are much clear to me - there I said it.

To be honest the author comes across as one of those who insist on people following the "correct" rules even when they get in the way or there is a better solution.

I'll bet there are a lot of people who also disagreed with some of his examples but who won't say anything for fear of appearing "ignorant" in the eyes of the "design police".

Another important point. He states

"On average, only 3% of people who have downloaded an app use it after 30 days. Why? Because the majority of iPhone apps don’t make any sense to users."

But where is the evidence for this? He shows App usage over time graphs which don't separate between "over-designed" and "correctly designed apps" and then goes on to make an unwarranted assertion that the fall off is due users not understanding how to use the apps! How about the utility value of the apps not being enough or users getting bored with their $0.99 fun purchase ?

Lastly, if you go to the author's web site. You'll see he's designed an app for a pizza company. I downloaded it and it's beautiful (in fact I fancy a pizza now)! However, It does not use ANY standard IOS controls and has gone for a very nice custom dedicated UI!

His article spends a lot of time basically saying that the standard UI should be used but as his own app demonstrates there are cases where a custom UI is appropriate. His article would be much better if he provided guidance on when using a custom UI is better and provided evidence for his assertions.

If you want a look at the long-term consequences of a culture of pervasive custom UIs, look at the current state of Windows apps. Every single app looks different in some minor way for no discernible reason. The sum of them on the desktop adds up to visual noise and clutter that is needless. I realize the differences in look are not intentional: GTK here, .NET there, each with a different layout engine.

But the sad fact is, nobody seems to care about it. There is no standard UI on Windows anymore, and it drove me nuts.

My favorite whipping boy in this argument is Trillian: http://pcworld.hu/apix/0911/trillian_astra.jpg

I think the Reeder App is over-designed. The developer basically blazes his own path and the app only follows conventions nominally. However, at the same time it's done with great taste. In some cases, following HIG to the letter might make it look like you slapped together an app with widgets.

Also, every application by Tapbots pretty much flaunts all conventions with highly custom UIs, yet they post great sales. I suppose the same could be said of all TapTapTap apps, they also sell great.

However, one thing they all get right is great functionality and performance stability. You really can't trump that.

Reply on news.ycombinator.com