We attempted, and failed, to bring World of Goo to iPhone in the summer of 2009. Development was sporadic and slow as we worked with a series of four different contractors who, for various reasons, did not bring the project to completion. So we dropped it.
When the iPad came out, we thought we’d try again, this time with a fresh round of confidence that the bigger, more powerful device would be able to support a console quality port of World of Goo. We started working with Ron’s brother, Gil Carmel, in November 2010, who finally got this project done. There is nothing all that remarkable to tell about the development of World of Goo for iPad, but this being our first App Store title, the month leading up to launch and the month that followed were very interesting times, as we gathered information and lessons about the App Store market that we hope will be of use to other developers.
We found that the average iPhone/iPad gamer is more interested in pleasantly passing time than being intellectually engaged or challenged, at least when compared to the average PC or console player. Overall, World of Goo has been receiving very favorable reviews in the App Store, averaging 4.5 stars. But of the most critical consumer reviews, the majority express their frustration with the difficulty of the game (price too, but that will be discussed later). A sampling of our favorite 1-star reviews in the App Store:
“I don’t know how the heck you do this!!!!!!!!!! “Drag and drop to build the pipe”? WHAT???? Somebody please tell me how to do this!”
“I’m only on the 6th level and I hate this game. Levels are ridiculously hard from the start and are just stupid. I spent an hour on one level and still cannot beat it. Screw this crap. Worst. Purchase. Ever.”
“Don’t get it, it will get you very frustrated if you don’t beat a level bottom line don’t get it”
We playtested World of Goo extensively on six year olds, parents, grandmothers, and random people at coffee shops who either don’t play games or actively avoid playing them, so we know that the game is intuitive and self explanatory. We’ve also never received this type of complaint for either the PC or Wii version. What we neglected to consider is that the iOS audience might be looking for a different kind of fast-fun entertainment, where punishment for failure, no matter how slight, is not an option, and no matter how badly you play the game you always feel you have a reasonable chance of success. To address this, an updated version included a more prominently featured “skip level” button, and allows an unlimited number of skips. Another option we jokingly considered is to ask players to pay to auto-complete levels by purchasing a Mighty Goo Ball.
Even after making it easy to skip levels, the game will likely still be too challenging for some players’ tastes, but we’re OK with that. It’s a puzzle game, not a scenic tour of Kyle’s artwork.
For exactly these cases of false expectations and disappointment, we wish Apple offered a mechanism by which a developer can issue a refund to unsatisfied customers. We sell the PC version of World of Goo on our website directly, and any time we get an email from someone who was not pleased with their purchase, we offer them a refund. We’d much rather have a happy non-customer than an unhappy customer.
As far as we could tell, there are three ways an app can be effectively promoted:
1. Get it featured by Apple,
2. Get press to write about it, and
3. Be in the top selling / top grossing charts.
It was intuitively clear to us that the most important promotion we could get is to get the game featured by Apple. Promotion inside the sales channel is effective in both retail (the reason publishers buy end-cap space for their games and why impulse items are shoved in your face at the checkout line at the grocery store) and in digital (which is one of the reasons Steam promotions are so incredibly effective and why publishers buy dashboard placements for their XBLA games).
We assumed the same is true for the App Store, and this was confirmed by more experienced iOS devs we talked to. The only thing we could think of to increase the chances of getting featured was to build awareness of the the upcoming release within Apple. We emailed the one person we knew there and were fortunate that the news was received with excitement. We sent them a pre-release build, got some great feedback, and continued to work closely and seek advice up until release.
Considering both the quality and quantity of games being released around the holidays, we operated under the assumption that World of Goo would NOT be featured and did everything we could to generate buzz for the launch by working with the press to get the word out.
Our general approach to PR was the same as for every other platform we launched on: build up buzz by gradually releasing more specific/exciting information leading up to the release date. We asked experienced iOS devs for press contacts at the larger iOS review sites and sent them pre-release versions of the game. Because World of Goo has become fairly well known, we were fortunate to have a relatively easy time getting coverage.
The big difference this time around though, was that it had been two years since the game’s first release, and a large part of the audience that would read this news had already heard of the game, if not played it. Was there really a reason for them to play it again? We thought there was. On the large and beautiful iPad touch screen, the feeling of having your fingers dipping directly into the goo made this one feel like the Definitive Version of World of Goo. We emphasized this when writing about the iPad release and when speaking with the press.
As for getting the game in the top selling / top grossing charts, we didn’t really give this factor the attention it deserved. At some point shortly after the game was released, Jamie Swirsky (of Indie Game: The Movie) said to us something along the lines of “This is going to be the next Angry Birds” to which we instinctively replied “There’s no way this is going to be the next Angry Birds at $10”. And right then and there it became clear that $10 was not the right price point. Which brings us to the next point of discussion…
As proud parents, we wanted the perceived value of World of Goo to be high, and we wanted to combat the so called “price erosion” of the App Store. Plants vs Zombies was selling for $10 at the time and we thought World of Goo, being on par with with PvZ could support the same price. In a funny turn of events, PopCap dropped the price of PvZ to $7 less than a week after World of Goo was launched.
Sales held up nicely while the game was prominently featured in the App Store but started to decline pretty rapidly afterwards, as you can see here:
We were convinced that without intervention, World of Goo would fall off the charts and lose all visibility, so we decided to experiment with a price cut. We dropped the price from $10 to $5 about a month after release. We emailed press folks to notify them of the upcoming price change ahead of time, and over the course of the next 24 hours World of Goo shot up the top grossing charts from #51 to #2. We were hoping that this new price point would find a relatively high equilibrium point on the charts, but so far it seems that it provided a temporary (but very significant) sales boost without preventing long term decline.
One unfortunate outcome of any sale in any market is the possibility of alienating early adopters, a group which likely includes the most loyal fans. Before launch, we thought it would be better to start high and lower the price, as a pleasant surprise, than to start low and then anger people by raising it. We suspect that a vocal minority would have been angered either way, but if we could do it over again we might have launched at a lower price point and said it was a temporary promotion, essentially reserving the right to raise the price, but without angering early adopters.
It’s possible that $5 might have been a better price point to begin with. While $10 is less than most people pay for a movie ticket, or lunch, it’s still seen as a very high price for a game on the App Store and turns many people off. As you can see from the daily revenue chart below, World of Goo generates significantly more revenue at a $5 price point than it did at $10 (price was halved on January 14). The counter-argument would be that the second round of press we got when the price dropped could have more than made up for the missed sales at the $10 price point.
The higher revenue at the lower price point brought about a realization about “price erosion”. The notion that “App Store price erosion is bad for developers” could be a backwards way of looking at things. What is generally referred to as price erosion occurs because developers are optimizing their revenue. If a game earns 50% more revenue at a lower price point, it’s a pure win situation as the developer makes more money AND more people get to enjoy the game. And if those two things are true, does it really matter what the sale price is? If we all charged double for our games we might all earn more money, but we could also end up earning less money because people would buy much fewer apps.
That, along with examining the top selling games on the iPhone App Store made us realize that if we end up releasing an iPhone version we would need to sell it at a very low price point if we want it to be a top seller in the long run. At the time of writing, 18 out of the top 20 selling iPhone apps are priced at 99 cents. Of the top 20 grossing apps, 15 are either free or cost 99 cents.
In early November we finalized the details of our development agreement with Gil and decided to crunch in order to get the game out in time for the holidays. We worked on it full time along with Gil and two other programmers from Page 44 Studios, Brian Morishita and Nick Tourte.
The rationale for the crunch was that if World of Goo was going to get featured, releasing the game on December 16 would mean that the game would remain featured for an extra week during Apple’s holiday freeze of the App Store (December 23-28).
This is the same App Store “loophole” that EA exploited by dropping the price of all their games to 99 cents the week before the holiday freeze in order to gain chart position, and the increased visibility that comes with it.
The timing worked out very well. World of Goo was featured as iPad game of the week for two weeks straight, and sales were great.
The rush to submit the game in time for a Christmas release did have a couple of downsides. There were some bugs that we may have found in testing if we had more time, and performance wasn’t as good as it could have been. Both issues were addressed in an update we released shortly after. Overall, we feel this was a good trade-off.
Rank And Revenue
Having obsessively monitored World of Goo’s App Store ranking and sales numbers after launch, one of the things we found surprising is that when World of Goo was hovering near the top of the charts we saw that the #1 app was selling about twice as much as the #2 app. This drove home the point that it’s dangerous to judge the health of a distribution channel by how much the top selling game makes. If you’re lucky enough to reach the top of the charts, unless you’re Castle Crashers or Angry Birds, you’re not going to stay there for very long. So what matters more than how much the top dog makes is sales distribution. How much, on average, do the top 20 make? Top 50?
To begin answering this question, we took the sales and rank data we collected since the release, and created a scatter plot. On the vertical axis is the net revenue generated on a given day, and on the horizontal axis is the rank of the game on the iPad top grossing chart:
As you can clearly see, once a game breaks into the top 10, the amount of revenue it generates skyrockets. It’s the other way around, really, but you get the picture. It is encouraging to see that when a game drops off the top 10, revenue declines fairly slowly. Even the lowest data point in this scatter plot still represents daily revenue measured in thousands of USD.
One Last Observation…
In 2008, with the successful releases of Castle Crashers, Braid, and World of Goo, it became fairly clear that consoles were “where it’s at” for independent developers, and a lot of attention was given to which console provided the best distribution opportunities. Nintendo had the largest install base, XBLA had the largest number of registered users, and PSN had the strongest growth momentum. This discussion is still going on today and the landscape is constantly shifting.
World of Goo’s launch on iPad gave us a new perspective on that discussion. In the first month of sales on the iPad App Store, World of Goo sold 125k copies (thanks to being prominently featured by Apple). In comparison, World of Goo’s best 31 day period on WiiWare was 68k copies (thanks to a mass mailing by Nintendo), and on Steam it was 97k copies (thanks to two promotions at discounted prices). So far, the iPad version is by far the fastest selling version of the game, both in terms of number of units sold and in revenue generated.
What makes this even more amazing is that this is a two year old game released on a platform that is less than a year old. The iPad doesn’t have the benefit of an install base built up over several years.
In the short term, we still think that if an independent developer can get their game on a console it’s a safer bet than playing the App Store lottery, but one might wonder whether, in the long run, it even matters who wins the PSN / WiiWare / XBLA race.