3 ms·

I'm a product manager that oversees several hundred apps over roughly 25 stores on 8 platforms (Android, iOS, BlackBerry/QNX Windows, Symbian, Bada, Brew, J2ME)

I'm a product manager that oversees several hundred apps over roughly 25 stores on 8 platforms (Android, iOS, BlackBerry/QNX Windows, Symbian, Bada, Brew, J2ME). We are do a steady 2.5 million in sales per year, aiming for 4m next year. The problem is that startups in the Valley are extremely narrow minded and aesthetically snobbish in their mobile approach. 1) There are 1 billion feature phone users around the world who have perfectly usable app stores. 2) You need to launch in more languages than just english, and on more app stores than just the US market -- Chinese, Hindi, Spanish and Portuguese at a minimum. 3) Apps that are not SAAS are throw away, impulse buys. They should be developed with that in mind. 4) Apps have a short lifespan and sell based on timeliness. 5) If you want guaranteed profit in mobile, build mobile extensions for established brands, don't try to build the next Angry Birds or Apple App clone. 6) Design matters, but not that much. Touch interface has made UI far more visceral, but the sale happens long before the user judges the design 7) IAP -- its icky but it works 8) Try lots of stuff rapidly and fail fast. 2-3 month cycles. You can get up to about 10 apps a month if you have a larger team and a good process. Finally, this is nothing like 1999. Companies were failing with no product, spending 1 year + and not even launching. There were late round fundings for companies that literally had products that were figments of their imagination. Now two guys in their garage can build a hit app with two macs and no funding what-so-ever. Don't confuse a highly active, easy to penetrate market with a bubble.

anywherenotes 14y ago
What is IAP? Thank you for the post btw.
- thegoleffect 14y ago
  In App Purchases.
frankieim_111 14y ago
The artist in me finds your shovelware business model sad, although my inner accountant nods approvingly. I guess profitability trumps a lot of other concerns, since you can't put out too much software if you're bankrupt. I don't understand your point (6) - intuition and a small amount of experience tells me that users judge software overwhelmingly by its design - can you elaborate on this?
- tsunamifury 14y ago
  My inner artist actually struggles a lot with it as well, however it has allowed us the runway to invest in larger, higher quality projects. Regarding (6): Users judge by design, if design is all your app has. If you are providing an enormous amount of content thats of value to the user, design becomes secondary. For example, 2 of our highest performing apps (about 70-100k per year) have poor interfaces, but killer content that specific users really want access to.
  - nreece 14y ago
    Regarding (6): Spot on. "The lower your product is on the Maslow’s hierarchy of needs, the crappier it can look." http://engineering.gomiso.com/2012/02/01/youre-overthinking-it/ http://engineering.gomiso.com/2012/02/01/youre-overthinking-...
  - joering2 14y ago
    since you so kindly share with us -- can you please name some of those apps? I would love to download (pay if needed), play with it, see the value with my own eyes. Thanks!
- prostoalex 14y ago
  I think what he meant is that design guidelines, libraries, and UI frameworks provided by OS resonate perfectly well with majority of the userbase. Yet every app that comes out of Valley seems to need to redefine basic navigation patterns and color themes.
- jasonhitchcock 14y ago
  Users make snap judgments in the app stores based off of star ratings, some comments, and possibly looking at a screen shot or two. The interface isn't fully appreciated until after the money is forked over, hence the devaluation of its importance in terms of making a quick sale on a throwaway app.
- utopkara 14y ago
  Your disapproval of "shovelware" business model and the related shallowness of design does make sense; however, failing fast and often does not imply the malicious intent as you imply. Think of this as a search for an anchor point, a good business, so that the whole enterprise can take a breath. If you hit the spot in the first try, all power to you! Chances are even when you have a good hunch, some unlikely/unforeseen thing will block the way to success. Why feel sorry for having all the eggs in one basket, when there are many things in place that would make it possible for you to do otherwise, e.g. several easy ways to reach customers, in several countries, and relatively untapped markets etc. This may not be valid for big companies with lots of spare money to cover the losses until the app has the power to take off, or developers who are in love with an idea to the point that they will sacrifice themselves until giving it a real push.
hospadam 14y ago
Willing to share how big of a team you need to have to turn out 10 apps per month? I'm honestly curious!
- tsunamifury 14y ago
  Don't want to give away all the secret sauce, but you can do it with about 10 US workers and about triple that in India.
bazookaBen 14y ago
what're your thoughts on HTML5 games? Since you're so invested in so many apps across multiple platforms, there has to be some thought on a cross-platform strategy. Any way I can reach you directly?
- tsunamifury 14y ago
  The reality of HTML5 is that it is extremely processor intensive, radically different from mobile browser to mobile browser, and simply not supported at all on 4 of the 8 platforms I create for. Building a XML-based object architecture and then interpreting it into ObjC and Java, then standardizing the data structure it accesses ends up being far better than hacking together a series of poorly performing HTML-based apps. In short -- native is still best, by far. Put of the contents of an app itself in a standard database that all the native apps can access, to minimize your native code dev time. Or just use Unity.
  - drh 14y ago
    > Or just use Unity. I'm curious how you deploy to the non-smartphone devices with Unity, as it only has iOS and Android target support built in. Do you use the output of some other target's build with your own glue libs for each supported device? Or are there Unity plugins for this kind of thing? Or is it some other kind of wizardry?

anywherenotes 14y ago

What is IAP? Thank you for the post btw.

thegoleffect 14y ago
In App Purchases.

frankieim_111 14y ago

The artist in me finds your shovelware business model sad, although my inner accountant nods approvingly. I guess profitability trumps a lot of other concerns, since you can't put out too much software if you're bankrupt. I don't understand your point (6) - intuition and a small amount of experience tells me that users judge software overwhelmingly by its design - can you elaborate on this?

tsunamifury 14y ago
My inner artist actually struggles a lot with it as well, however it has allowed us the runway to invest in larger, higher quality projects. Regarding (6): Users judge by design, if design is all your app has. If you are providing an enormous amount of content thats of value to the user, design becomes secondary. For example, 2 of our highest performing apps (about 70-100k per year) have poor interfaces, but killer content that specific users really want access to.
- nreece 14y ago
  Regarding (6): Spot on. "The lower your product is on the Maslow’s hierarchy of needs, the crappier it can look." http://engineering.gomiso.com/2012/02/01/youre-overthinking-it/ http://engineering.gomiso.com/2012/02/01/youre-overthinking-...
- joering2 14y ago
  since you so kindly share with us -- can you please name some of those apps? I would love to download (pay if needed), play with it, see the value with my own eyes. Thanks!
prostoalex 14y ago
I think what he meant is that design guidelines, libraries, and UI frameworks provided by OS resonate perfectly well with majority of the userbase. Yet every app that comes out of Valley seems to need to redefine basic navigation patterns and color themes.
jasonhitchcock 14y ago
Users make snap judgments in the app stores based off of star ratings, some comments, and possibly looking at a screen shot or two. The interface isn't fully appreciated until after the money is forked over, hence the devaluation of its importance in terms of making a quick sale on a throwaway app.
utopkara 14y ago
Your disapproval of "shovelware" business model and the related shallowness of design does make sense; however, failing fast and often does not imply the malicious intent as you imply. Think of this as a search for an anchor point, a good business, so that the whole enterprise can take a breath. If you hit the spot in the first try, all power to you! Chances are even when you have a good hunch, some unlikely/unforeseen thing will block the way to success. Why feel sorry for having all the eggs in one basket, when there are many things in place that would make it possible for you to do otherwise, e.g. several easy ways to reach customers, in several countries, and relatively untapped markets etc. This may not be valid for big companies with lots of spare money to cover the losses until the app has the power to take off, or developers who are in love with an idea to the point that they will sacrifice themselves until giving it a real push.

hospadam 14y ago

Willing to share how big of a team you need to have to turn out 10 apps per month? I'm honestly curious!

tsunamifury 14y ago
Don't want to give away all the secret sauce, but you can do it with about 10 US workers and about triple that in India.

bazookaBen 14y ago

what're your thoughts on HTML5 games? Since you're so invested in so many apps across multiple platforms, there has to be some thought on a cross-platform strategy. Any way I can reach you directly?

tsunamifury 14y ago
The reality of HTML5 is that it is extremely processor intensive, radically different from mobile browser to mobile browser, and simply not supported at all on 4 of the 8 platforms I create for. Building a XML-based object architecture and then interpreting it into ObjC and Java, then standardizing the data structure it accesses ends up being far better than hacking together a series of poorly performing HTML-based apps. In short -- native is still best, by far. Put of the contents of an app itself in a standard database that all the native apps can access, to minimize your native code dev time. Or just use Unity.
- drh 14y ago
  > Or just use Unity. I'm curious how you deploy to the non-smartphone devices with Unity, as it only has iOS and Android target support built in. Do you use the output of some other target's build with your own glue libs for each supported device? Or are there Unity plugins for this kind of thing? Or is it some other kind of wizardry?