On 2.0 and rethinking an app

I’m a fan of MoneyWell. Version 2.0 has just come out, and I’ve been hanging around their support forums to try and learn a bit more about how to use it effectively, and to see which bugs are known about and being tracked. With version 2.0, a lot has changed: they rethought the entire interface and started again, with the benefit of all they learned from making the 1.x series, which resulted in a streamlined app that was pared back to its essential features. It also has a UI of the modern Apple style, with monochrome buttons and non-adjustable columns. Unsurprisingly, user reaction has been fairly polarised.

I started this post as a reply to this message by Romain Maurer:

I thought I would leave this message to tell you that I actually really like version 2.0… because we seem to only see unhappy customers leaving messages on the forum…

I really appreciate the spending plan compared to what it was in the first version and that’s definitely a major improvement!

(Romain then went on to offer a few suggestions for improvement, including suggesting they post a road map and ETA.)

Here’s what I started writing as a reply to that post:

I really like it too. I’ve offered more than my fair share of suggestions, but that’s because it’s an app that’s worth me spending time writing them up!

Above everything else, I love that a developer of an established product was brave enough to start again and rethink things. Any time a developer compromises and adds a setting to revert back to the old way of doing something, it’s a cop-out solution. It’s like when Apple built the first Mac, they didn’t put arrow keys on it so that people couldn’t just fall back on old habits and never learn what was so good about the new way of doing things.

Does that mean I think MoneyWell 2 is perfect? Nope. But show me a product that is perfect. Certainly the things I’ve shipped aren’t, and that’s because if you waited for perfection you’d never ship!

Regarding asking for an ETA, consider Hofstadter’s Law: it always takes longer than you think, even if you take into account Hofstadter’s law when making an estimate! I don’t blame No Thirst for not giving estimates, as people tend to treat them as unmissable deadlines then complain when there’s a delay. 

The main thing I’m taking away from reading the support forums: out of the complaints, they fall into a few categories:

1. People experiencing bugs (i.e. the app doesn’t work the way Kevin and the team intend it to). Obviously those should be fixed.
2. People who are finding the new way of working hard to understand. I’ve fallen slightly into this camp myself. Good documentation and videos will help here.
3. People who are missing a specific feature from 1.7.
4. People who resent that there is a new way of working at all.
5. People who don’t necessarily mind change, but don’t like the direction the app is going in.

These remind me a lot of the complaints we got when MacOS X was released, for those of you who remember that far back. Consider these:

1. It crashes sometimes when I’m trying to wake from sleep!
2. How do I see all the devices on my network? There’s no Chooser, and the Network icon does nothing.
3. I can’t burn a CD! My printer doesn’t work!
4. Why did Apple have to go and invent a new interface? Couldn’t they stick the NeXT kernel under the Platinum UI that has proven itself over many years to be slick and efficient? All the icons now are huge and waste loads of space: I’m not blind, I don’t need a font this big!
5. They killed the spatial finder! People find it really helpful to organise documents based on where they are in a window. That was one of the great strengths of the Mac!

And of course with Mac OS X, groups 1 and 2 are now completely satisfied, and group 3 are mostly satisfied (or at least appeased by other advantages). 

Groups 4 and 5 are more interesting: they’ll definitely have got used to the new changes now, and again if nothing else they’ll have realised that with the new OS came a lot of other things that wouldn’t have been possible before. Thing is, if a company has a certain vision for a product, and that vision is something new and exciting, it will by definition take users out of their comfort zone, because it’ll be something nobody else could anticipate. 

The issues arise when the developer is seeing things in terms of what the user is trying to accomplish (on a high level), and the user is seeing things in terms of what steps they want to be able to take to accomplish their goal (a lower level). No Thirst are making an app to allow people to use envelope budgeting in the digital age: I think that’s how they see their mission at the highest level. 

When a user says “I want to be able to fill this bucket by this much on the first half of every month” (i.e. a feature that was there in version 1 but doesn’t exist in the same form in version 2), No Thirst are translating the request into what the user seeks to accomplish. For that example, the user wants their budget to be recurring, so they don’t need to manage it separately every month. They want to have it under their control but informed by their past spending. They want it to be quick to set up. So No Thirst come up with a new system: the Spending Plan. If used as intended, it satisfies the first two of those points, and isn’t too much worse on the third point (speed of setup). Plus, it has other advantages which the old system couldn’t cope with. But the user isn’t seeing things like that. They’re seeing things on the level of the task they were doing, and now they can’t type a number into a little box per bucket. The user doesn’t want to have to think about how changing their workflow could let them meet their goals in a better way.

That reaction is instinctive. The users aren’t trying to be awkward, they’re just observing that a change to their software has caused them to have to spend time adapting what they do. The thing that the software company are relying on is that in the long run, a major version upgrade that is a rethink of the core values of an app will improve people’s workflow, and also will make it easier to improve the app in the future. The developers are banking on these improvements being worth the temporary discomfort around the upgrade, and that they’ll gain more new users over the lifetime of 2.x than they’ll lose old ones.

So, to summarise: I’m glad No Thirst rethought their app from the ground up. That’s how we get better apps. I have faith that bugs will be fixed, documentation will be released, and any rough edges in the new workflow will be smoothed. I think the user reaction to workflow changes is understandable, but that shouldn’t stop software developers making drastic redesigns. Also, everyone here uses a Mac: you know, the computer made by the company that is king of the start-again-from-scratch rethink? Recall the iPod nano (too expensive), iMovie 08 (too dumbed-down),  the Intel transition (I have to re-buy all my apps?), Xcode 4 (but I have multiple monitors!), the iMac (but people still use floppy disks!), the new Mac mini (but people still use CDs!), iCal (it's… orange…), the Macbook Air (but Firewire?), the iPod ($400 for 5GB and it’s Mac only?), the new AppleTV (no local storage!)… I could go on. All of them got terrible press at first, but went on to become awesome after a few revisions (ok, maybe not iCal), and now they’re leading their category. It’s companies that dare to rethink a successful product that end up with something amazing.