Hacker News new | past | comments | ask | show | jobs | submit login
From 0 to 450k, an App Store Success Story - With Charts and Numbers (struct.ca)
222 points by sp4rki on Oct 19, 2010 | hide | past | web | favorite | 63 comments

The most interesting thing to me is that the app sat in relative obscurity for 4 months before being featured by Apple, at which point sales exploded.

And judging by the App Store, being 'featured' only meant that the app icon was added to the New & Noteworthy section....not the giant banner you get for being App of the Week (currently Nike GPS) or Game of the Week (currently Cut the Rope)...unless I missed it earlier in the month?

I'd consider that "featured" because you touch the little "featured" spotlight icon in the bottom of the App Store to get to these apps.

I probably buy a game from these sections every week, or 2 weeks, but I almost never scour the store for hidden gems myself.

I think this section of the App Store is a HUGE draw for developers because it allows them to use their imagination when creating a new game and they still have an inkling of hope that they'll get noticed or featured.

In the Android Market however - if you create something awesome but original, you're pretty much doomed to obscurity unless you find some other marketing avenue to drive traffic to your game.

Or maybe by some miracle Google will feature you, but they have pretty limited space available for featured apps, and nobody seems to know how those are chosen.

That's a good point, actually. The way the App Store looks and feels on the iPhone itself is a lot different than the version on the iPad or in iTunes.

You get a lot more visibility being featured on the iPhone compared to the other two.

Note that the 450k download number is of the free version of the game, Trainyard Express.

There is also an impressive sales graph for the paid version at the end of the post, suggesting that the game has been rather lucrative since the beginning of October: http://struct.ca/wp-content/uploads/2010/10/TheWholeShebang....

Note also that 450k download number for the free version is now > 1 million!

From its App Store description: Downloaded over a million times, woah!

[Edit: make that 2 million now http://news.ycombinator.com/item?id=1810010]

The little bump at oct 2 is $240, so it looks like each major horiz line is 2k/day. It seems the day he posted this, it made just north of 9k/day.

Looks more like $1k per horizon line to me, but $4.5k/day is impressive nonetheless!

Hey this is Matt, without actually saying the total value, I'll just tell you that the horizon lines are $1.2k ;)

So that's a peak of about $4.7k.

No, i think peak value is 1.2k * 4.5 ~= $5.4k

Whoops, you're right. I'm guessing $5.6.

This is certainly an inspiration for developers that want to take the time to make a quality game, instead of trying to "game" the app-store with submissions like iFart, etc.

I was just reading about this article in relation to how the submission system works here at HN...


It's not just the title that was wrong: when I load the article, there is nothing compelling in the first 3 pages of text! It takes scrolling a lot before some useful data shows up here and there. No wonder it didn't get any upvotes the first time.

Yeah when I first read it I thought the same, but you do have to admit that the title made all the difference in regards to exposure. I even think i could have made a better job of the title had I actually though about it more, but since this was only a little experiment to prove the importance of titles in HN I guess it proved a point ;)

This is awesome. The story, I mean; unfortunately, I do not have an iOS device, which means I cannot play the game (I would if I could!). It gives me hope knowing that there are great success stories out there like this still. I know the chances are slim, but it's still really cool to hear these sorts of stories.

Congrats on the success of your game!

"always release on a Thursday" - never heard that one before

Anyone care to speculate or enlighten the rest of us on where this pearl comes from?

Enlightenment not implied, but here's my retrospective theory:

  Sunday: few people looking to start something new, languish for a few days.
  Monday: it's *Monday*.
  Tuesday: Still beginning of the week, things are still being piled on
  Wednesday: nothing wrong with it
  Thursday: early adopters try, advantage of full Friday + an extra day of users before weekend
  Friday: people finish / wind down / have more free time to start things
  Saturday: In the swing of X, or looking to start X.
With an assumption that you will always have early adopters any day of the week, at roughly level values (or increasing on Friday evening through Sunday).

My final take from that: Thursday gives you part of a day with early birds to find game-breakers ASAP, and Friday and Saturday are useful days to have early to boost your numbers at the start. You're also sure to get all of Friday by starting on the day before. Wednesday gives little advantage over Thursday, but costs you another day before bug-fixing weekend / next week. And early week releases are, well, too close to Monday. Nobody likes that.

All of which is wildly subjective, and is merely justification and not reason. But it makes some sense in this light (to me, at least).

There are two parts to the Thursday logic: The first is that Apple updates the "featured apps" list every Thursday, so in theory if your game is coming out that week, you could get featured on your launch day.

The second part is the fact that Saturdays are the biggest App Store sales day of the week, so by launching on a Thursday, you have a couple days to build up steam and be high in the charts by the time Saturday rolls around.

i would imagine that applies more to the entertainment content. A game released on Thursday, will get played and told to the friends about Friday thru Sunday

App store sales spike on weekends, so better to ride that trend.

If you release on Thursday and there are problems, you can always fix them on Friday. If you try and release on Friday, any problem will make for an unhappy releaser.

fixes have to go through app store approval, so you can't fix it on friday. it takes about a week for it to get released

I noticed the paid app was sold at varying times for $2.99, $1.99, and $0.99. Anyone have more information about the physiology/performance/trade-offs of those different prices?

Do $0.99 apps sell way more than other priced apps?

Trainyard was at $1.99 at launch and for a while after that. I figured it was a low introductory price, but not so low that I'd get careless users who wouldn't put any effort into the game and would just give it 1 star reviews, which was a bigger problem back a few months ago before Apple took out "rate-on-delete".

These days, I feel like there are less tangible disadvantages to the 99c price point, other than the fact that you make less per sale. There's still the issue that users are willing to put more effort into a product they pay more for, and will even go to further lengths to defend it, because they want to justify their purchase and their taste.

On the App Store, it all comes down to perceived risk. If a user is unsure about an app, they probably won't spend much more than 99c because they don't want to overspend on a dud. Once a user sees an app in Apple's "featured list" or has used the lite version of that app, they'll be much more willing to splurge on a higher price, because they know what they're getting.

For Trainyard, I felt that there was no longer a big difference between the $1.99 and $2.99 price points, because by the time a user arrived at the purchasing page, they already knew they wanted it.

All that being said, there is a large contingent of users that seem to only buy apps at $0.99, but most of the time, your apps won't be visible to those users. The moment I reached the top 50, I knew I would have tons of that sort of user looking at Trainyard, so I did the price drop.

I went the opposite route in the Android Market. When the lite version of my game hit the top 20 in its category, I made a few nice updates, and I felt the full version earned its higher price (~$4). After making the change, the number of sales barely went down, but revenue went up.

But now you have me thinking. Are MORE people in the the $0.99-only crowd seeing my app, and are there enough of those people to make DROPPING the price more financially beneficial???

I suppose I could run a "sale", but once you start getting comfortable with a nice daily income, taking risks becomes harder and harder.

Yep, that was exactly my dilemma. I was making a huge amount of income, so it was a pretty big risk to drop the price, but I knew I would only have the opportunity during the week I was featured. It very well could have dropped and failed.

I'm not sure how the Android charts work, but in the App Store, it's ranked by units sold, not by profit. This is the KEY thing that is crucial to how the market works. Even if I make LESS money than I did the day before, I will be higher up in the charts, which will get me more exposure, and eventually more sales. Going high in the App Store is all about upward momentum. The volume of sales doesn't change linearly with rank either, it's quite exponential.

   In the App Store, it's ranked by units sold ...
   I will be higher up in the charts, which will get me 
   more exposure ... volume of sales doesn't change 
   linearly with rank either, it's quite exponential.
This is a great piece of insight, much appreciated.

Yep, I seriously believe this is the absolute key to using the way the App Store works to your advantage. At a certain point, the rank of your app will help to increase the rank of your app. Strange but true.

If you were going to game the system and you spent $10000 or even $20000 to get your app into the top 5, you'd make your money back within a day or two, and your app would be remain "sticky" at the top (provided it wasn't awful).

It seems your original article is currently down for me, so excuse me if this is a repeated question.

Do you think it was a mistake to start at $1.99 based on your theory? Should the entire aim to be to go as cheap as possible to get into the top 10 in the first place? (Or is that unlikely unless there is some other form of catalyst, e.g. Apple feature and therefore you may as well have the price at $1.99 / $2.99 until that happens?).

I think tha maybe I should have started at $2.99, but definitely not 99c. Every user that found Trainyar at that time would already know about it and wouldn't have just stumbled upon it. I think the 99c price point is only worth doing if you're reaching users who haven't heard of it, so they're willing to take a risk.

Interesting, so start a reasonable price until you get a trigger point (e.g. Featured by apple) and then lower price to remain in the top apps.

> At a certain point, the rank of your app will help to increase the rank of your app.

Do you believe this also applies for users searching for a specific keyword or combination of keywords? The 'above-the-fold' phenomenon. For example, 'baby feeding'? How did you do on keyword searches before you were selected by Apple (though Trainyard is a pretty unique name)?

Congratulations on your inspiring success! Are you going to release the original Tic Tac Toe game now?

I have to be honest, I really don't think generic keywords matter at all, the vast majority of users only use the charts unless they already know what they're looking for.

I may do the tic-tac-toe game at some point as a multiplayer game.

So, if you get high up on the charts, do you plan to put the price back at $2.99 or stay/go back to $0.99 since everyday will be like you're featured?

Looking at Gamasutra (http://www.gamasutra.com/view/news/31045/Top_iPhone_Game_App...) it would appear that top-10 == $0.99. I'd guess this is because, as has been speculated here, there is an enormous number of people who are simply going to spend a maximum of $.99 on an iPhone game and that's that. Either you sell yours for $.99 or they'll find something else for that price. The evidence suggests to me that if you don't realize this, you'll simply be replaced in the top 10 by someone who does.

Of course, next week when a $9.99 game is in the top 10 I'll have to revise this opinion. :-)

Part of the reason is just because these are GREAT games that are all worth much more than 99c, but because of the fact that they are 99c, they're also fantastic value. Why would a user pay $2.99 for a fantastic game when they can get one for 99c?

I think his best plan would be to put the price back at 1.99 after he gets the publicity he needs and leave it there while he works on the next iteration or game. Then make a sale or 0.99 a few weeks before the new launch to create some media exposure and leave the game at 1.99 again close to the release of the new product and leave it there until it dies or new signups become so low that 0.99 becomes the new price for the old game (while focusing on the new iterations off course).

Well in my case, is kind of already happened, and I kept it at 99c. Tough decision, but I think it worked out pretty well.

But there's also a Top Grossing chart - do you think the Top Paid is that much more important? My initial impression is that getting to the Top 10 of either one will give you a lot more visibility (cause you're visible on the App Store homepage).

Top Grossing - from what I can tell - is significantly less important than the Top Paid list. It is rather hard to tell for sure, because usually you're in both or you're in neither, but I really think most users just look at either Top Paid or Top Free.

I believe you need to be aware of your markets behavior first. There will come a point where your revenue will start stalling and even decreasing, and that's a great moment to release a major update or start a sale. You want to increase your exposure to level decreasing revenue and it has the benefit of marketing your app to a completely different market with different rules on pricing. When you get this people on board and you take the price up again, you'll be taking some of this customers with you, for any next iterations updates, or even your other products if you decide to make another.

You sell your app for 1.99 and get a bunch of customers, who speak to other people and in time sales in this particular market start to slow down considerably. At this point you increase the price to make up for lost revenue on new sales. Then after sales start going under a threshold of popularity, you come out with a free version to appease markets that have been untouched by your project yet, but it also happens to have the massive effect of a new type of consumer that's more concerned with the expected vale vs real money value is now interested in your app. Some of them will buy the 2.99 version, but most will probably not. Then you come out with a sale, and boom it's now in reach for all those people that though 3 dollars was too much. Now you have massive sales and publicity, and when it starts to slow down you're back to increasing the price.

Then comes a time where there are no new markets to exploit, so you build a a new app or a continuation of the old one, a sequel. At this time you also lower the price to 0.99 just to make sure you have the most exposure possible. Only now you have the advantage of past and present publicity and if rightly timed, momentum... And then off course you repeat the formula depending on your market and the viability of your app.

Yep that's a great summary.

"Then comes a time where there are no new markets to exploit"

The crazy thing with iOS development is that there are 250k new users every day who are always looking for a few apps to put on their new phone, which is how a game like Angry Birds can stay high in the charts for months.

Oh yeah totally right, but there comes a time where no matter what you do, you're going to keep making less and less money out of the game because even though there are 250k new users ever day, the people that where interested in buying your app have already done so, and now only the passerby's and window-shoppers of the iOS world are looking at your app.

The cool thing is that now that you've been the developer of Trainyard, when you do reach a point where you should be releasing a new game, you have gotten so much credibility it isn't even funny. And off course that translate to more immediate sales at the beginning and a big chunk of extra steady clash flow at the end.

Yep, and I now have 2 million (!) downloads of the free version, Trainyard Express. That's a huge base of users I can reach when I eventually release a new game or a new feature in Trainyard. I go into a little more detail about the stuff that happened this week here: http://struct.ca/2010/the-week-that-was

My startup tracks pricing on iOS Apps (among other products) and its fascinating seeing how much pricing volatility occurs.

For instance, we've found that that Apps generally fit into two camps: those that change their pricing consistently; such as:


Versus those Apps which try to keep pricing stable and only have one-off promotions; such as bejewelled:


The general notion and hope behind such price changes is to get an influx of people in to buy rapidly (to grab the limited time discount price). This in turn hopefully pushes the App up the sales chart where it will receive more visibility and in turn more sales from users who would not otherwise be aware of the product.

I got Trainyard as soon as I got my iPod Touch recently just because of this post on HN. Actually, it was the first (and only, so far) game I've downloaded, just because of hearing about it here. Awesome game, and definitely something I think most hackers would like as it makes you think through the solutions. Great way to kill a few minutes between projects =D

BTW, just like PopCap does with their games, I think this one could live on a Ton of platforms ... including PCs/Macs. That'd get in the people without an iOS device...

Well deserved success. It's an excellent game and I love seeing solo developers at the top of the charts!

Your story is very inspirational and makes me want to get back to work on my simple game!

Are you looking to bring this to the Android market, and if you do, would you consider the 'free but with ads' style that is being offered by Angry Birds producer Rovio?

Yeah, absolutely. I've heard way too many stories about return policies and all that on Android, so I think I'd try to do the same thing they did. It really depends on the market at that point in time though.

Trainyard express is showing #1 top free and Trainyard (full) is showing #2 for me. Trainyard is also featured #1.

I'm not much into games but curiosity got the better of me and Trainyard Express is downloading.

I wonder if Matt is one of Toronto's "App Kings": http://www.torontolife.com/magazine/2010/11/

Nope. :)

too bad, sounds like you deserve it

I just bought Trainyard to thank you for having written such an interesting blog post.

Great story.

BTW, your dreamhost account was able to handle 100K submissions? not bad!

400k now. Yeah pretty crazy, I've almost used 40 gb of disk space, and if the traffic keeps up, it'll be a TB of submissions. That being said, it's also a bit unreliable and users often have connections fail when submitting, but it's really pretty solid. I've now switched to Dreamhost PS, so it's a little more expensive, just to make sure it's a bit more reliable.

Any examples of iPhone game developers who made a decent game but didn't make six figure downloads?

It depends on your definition of "decent", but I wouldn't be surprised if there are many. This very article would be an example, if the author had not decided to release a free version and the Italian press hadn't picked it up. Prior to that, the exact same game languished in obscurity.

Exactly, but all it takes is the right mention from the right influential person, and it can snowball. None of this would have worked if I hadn't made a good game though.

great breakdown, from start to finish

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact