May 7, 2010 at 10:43Godville on the iPhone - how it was
In this article I want to talk about our experience in creating and promoting an iPhone client for the Godville browser game . The game itself is a text-entertaining-parody self-game and has already been covered on the pages of the Habr. Actually, it was she who owed her existence to the habr , for which we once again thank from the bottom of our hearts and bow to the belt.
Godville has been developed for more than two years, so it managed to grow fairly interesting. But you won’t take a browser with you everywhere, and sometimes you want to visit a hero right here and now. We needed a mobile client, and after reading success-stories about successes in the Appstore, we decided to do something for the iPhone. We already had an iPod and a poppy, however, like registering with the iPhone Developer Program. The only thing left was to write.
They began to write in a team of two and a half people: one was directly involved in the client (seeing MacOS and Objective-C for the first time in his life), the second worked on the server API, and the author of this article pushed brilliant ideas and otherwise interfered with the process. Having started work in September last year, we planned to get the first viable version by the beginning of November, which could be released to the people and observe the effect. The client and the game itself are free, and the development was supposed to pay off and bring millions at the expense of those who wanted to buy in-game currency.
One of the key features of the game is the hero living on the server, who independently gets into adventures at unpredictable times. Therefore, the client really wanted to organize a data update close to realtime, so that possible news would come to the player as quickly as possible. The first experience with such things we got last year, adding Comet in the main version, so it was a sin not to use existing developments in the native client, simultaneously getting rid of browser “crutches”. The basic idea here is very simple - the client keeps a constant socket connection to the server, and that, in turn, sends updates exactly when they are, while at the same time trying to save traffic. As a safety option, there is a regular polling along with a tricky mechanism for switching from it to push and vice versa.
The convenience of the client and the interface have also been given special attention. Take the registration window as an example. The screenshot of the original version was not preserved, but initially it looked like a more or less standard registration form for six fields (two names, the hero's motto, password with confirmation, and mail). It seems to be nothing fatal, moreover, in the app, we often came across similar forms. But after the release of the first version, we found that many players register attempts from 6-8 (so many times the form was sent to the server to return with an error from the category "this name is already taken"), and some, apparently, just left, failing to register immediately. I had to deal with simplification - realtime-validation of the entered names was added, optional fields were transferred to the profile, and the default hash of the device id began to be used, which generally eliminated the need for the player to enter it. It all took about two weeks, but then the registration window began to please the eye and whistle through even the most lazy user (you can watch him live in the video below).
Finally, from the browser version, one of our favorite inventions, the “lower box”, was transferred to the client. It is a built-in form for feedbacks of various types, directly sent to our table. Generally speaking, the lack of contact between developers and users is a big problem for the app store (until recently, developers really couldn’t even read reviews, not to mention the answers). And direct contact with players through a convenient form allows you to significantly improve the customer experience, as evidenced by the five updates since the initial release.
The client’s release date was constantly postponed due to the desire to “tighten this useful little thing, too,” so we spent almost four instead of the originally expected two to three months (3-4 hours a day). At the end of December, by a strong-willed decision, they somehow put an end to it and finally sent the application for review to appstore, having previously read about the horrors of reviewing and expecting a refusal for some wild reason after two weeks of leisurely review. However, review passed on the first try and was surprisingly fast in just two days. It's time to reap the benefits of work.
The initial idea was a quick start due to those already registered in the game.owners of iPhones, getting into the general “Top-100 Games” and the subsequent raging there in the region of 50s with a small stable influx of new players. But here, we missed a lot by overestimating the number of iPhones in our community. Less than a hundred downloads on the first day, and on the strength of a dozen on the following, brought the application to the local “Top-100 RPG Games”, but potential players could not see it there. The updates and topic on the developer forum did not help, and there was simply no one to evaluate the months of hard work.
The option with advertising was not even considered. Firstly, we did not have sufficient finances for a noticeable advertising campaign. Secondly, Godville himself was never advertised anywhere either, and our only movements in the direction of promotion were the two aforementioned posts on the hub - the word of mouth worked further. Breaking a good tradition and luring people with banners was not interesting.
The last chance was to publish on some popular and thematic iPhone portal, and here the contest of developer reviews held at www.iphones.ru turned up very handy . We approached the matter in detail - it took several days to create a video review, the text itself was written and licked a couple of days. Finally a review has been published.on the iphones.ru main page, where it was read by a relatively small number of people for the site - about 3 thousand. About a thousand of them installed the application and registered in the game, watching this video review here:
And then an unexpected event occurred for us. Such a sharp wave of installations of the free application brought him to the main page of the Russian app, where it was fixed in third place in the global Top-10. The people tumbled down in shoals - about 1200-1500 downloads per day. Oddly enough, nothing fell, and, with the exception of a couple of minor minor bugs that crawled out, there were no special problems. Every day we went to appstore to look at our brainchild on the main page with tears of emotion and be happy for ourselves. But after five days, the number of downloads began to decrease (there was a steady feeling that all interested visitors to the appstore had already downloaded the game, and the people simply ended. Although who knows, maybe the name, description or icon is to blame?), And gradually we fell out of the Top- 10, and then from the Top-100.
Despite returning to almost starting positions, things are now much more fun - there are still a couple of thousand regular players from the iPhone, so complaining about life would be bad. But for us it remains a mystery what exactly keeps some applications in the top for months. It’s clear why Opera or Yandex.Maps are located there - but often frank junk with a rating of one and a half stars hangs in the top for weeks, and what can hold it there? Apparently, “nothing good, but everyone wants to see this personally” (c).
And finally - about these same ratings. After the first release, we found a powerful plug in the ratings: the application scores were completely polar, which is quite rare. The explanation is simple: the game, generally speaking, is not for everyone and requires some tension to get used to, so a rocker who does not understand right away with anger removes it with a minimum rating, and players who break through unusual and grasped the essence usually put the highest score. Everything would be fine, but the ingenious feature of iPhoneOS suggests rating the application only when it is removed (and what score can be expected?), But if it is installed and actively used, then there is no easy way to evaluate it. Having decided to correct the injustice, in the latest version we added a quick link to the client’s review page in the appstore, as well as a one-time request crawling out after a week of the game to rate it. The result is enchanting and shown in the figure below. Not that the rating of the application was of particular importance (unlike the paid versions, the free audience has much more and rarely knows what they want), but 90% of five-point reviews pleasantly warm the soul.
We summarize. The first experience of crawling onto a mobile platform with some stretch can be considered successful. Players are slowly coming, reviews are being collected, millions are accumulating, the client is regularly updated (next week it is planned to release the next version with a bunch of new amenities). Along the way, from this whole venture a full-fledged API was born, with the help of which especially advanced players make clients for other platforms. Well, in the meantime, we continue to bear plans for world domination.
Godville has been developed for more than two years, so it managed to grow fairly interesting. But you won’t take a browser with you everywhere, and sometimes you want to visit a hero right here and now. We needed a mobile client, and after reading success-stories about successes in the Appstore, we decided to do something for the iPhone. We already had an iPod and a poppy, however, like registering with the iPhone Developer Program. The only thing left was to write.
They began to write in a team of two and a half people: one was directly involved in the client (seeing MacOS and Objective-C for the first time in his life), the second worked on the server API, and the author of this article pushed brilliant ideas and otherwise interfered with the process. Having started work in September last year, we planned to get the first viable version by the beginning of November, which could be released to the people and observe the effect. The client and the game itself are free, and the development was supposed to pay off and bring millions at the expense of those who wanted to buy in-game currency.
One of the key features of the game is the hero living on the server, who independently gets into adventures at unpredictable times. Therefore, the client really wanted to organize a data update close to realtime, so that possible news would come to the player as quickly as possible. The first experience with such things we got last year, adding Comet in the main version, so it was a sin not to use existing developments in the native client, simultaneously getting rid of browser “crutches”. The basic idea here is very simple - the client keeps a constant socket connection to the server, and that, in turn, sends updates exactly when they are, while at the same time trying to save traffic. As a safety option, there is a regular polling along with a tricky mechanism for switching from it to push and vice versa.
The convenience of the client and the interface have also been given special attention. Take the registration window as an example. The screenshot of the original version was not preserved, but initially it looked like a more or less standard registration form for six fields (two names, the hero's motto, password with confirmation, and mail). It seems to be nothing fatal, moreover, in the app, we often came across similar forms. But after the release of the first version, we found that many players register attempts from 6-8 (so many times the form was sent to the server to return with an error from the category "this name is already taken"), and some, apparently, just left, failing to register immediately. I had to deal with simplification - realtime-validation of the entered names was added, optional fields were transferred to the profile, and the default hash of the device id began to be used, which generally eliminated the need for the player to enter it. It all took about two weeks, but then the registration window began to please the eye and whistle through even the most lazy user (you can watch him live in the video below).
Finally, from the browser version, one of our favorite inventions, the “lower box”, was transferred to the client. It is a built-in form for feedbacks of various types, directly sent to our table. Generally speaking, the lack of contact between developers and users is a big problem for the app store (until recently, developers really couldn’t even read reviews, not to mention the answers). And direct contact with players through a convenient form allows you to significantly improve the customer experience, as evidenced by the five updates since the initial release.
The client’s release date was constantly postponed due to the desire to “tighten this useful little thing, too,” so we spent almost four instead of the originally expected two to three months (3-4 hours a day). At the end of December, by a strong-willed decision, they somehow put an end to it and finally sent the application for review to appstore, having previously read about the horrors of reviewing and expecting a refusal for some wild reason after two weeks of leisurely review. However, review passed on the first try and was surprisingly fast in just two days. It's time to reap the benefits of work.
The initial idea was a quick start due to those already registered in the game.owners of iPhones, getting into the general “Top-100 Games” and the subsequent raging there in the region of 50s with a small stable influx of new players. But here, we missed a lot by overestimating the number of iPhones in our community. Less than a hundred downloads on the first day, and on the strength of a dozen on the following, brought the application to the local “Top-100 RPG Games”, but potential players could not see it there. The updates and topic on the developer forum did not help, and there was simply no one to evaluate the months of hard work.
The option with advertising was not even considered. Firstly, we did not have sufficient finances for a noticeable advertising campaign. Secondly, Godville himself was never advertised anywhere either, and our only movements in the direction of promotion were the two aforementioned posts on the hub - the word of mouth worked further. Breaking a good tradition and luring people with banners was not interesting.
The last chance was to publish on some popular and thematic iPhone portal, and here the contest of developer reviews held at www.iphones.ru turned up very handy . We approached the matter in detail - it took several days to create a video review, the text itself was written and licked a couple of days. Finally a review has been published.on the iphones.ru main page, where it was read by a relatively small number of people for the site - about 3 thousand. About a thousand of them installed the application and registered in the game, watching this video review here:
And then an unexpected event occurred for us. Such a sharp wave of installations of the free application brought him to the main page of the Russian app, where it was fixed in third place in the global Top-10. The people tumbled down in shoals - about 1200-1500 downloads per day. Oddly enough, nothing fell, and, with the exception of a couple of minor minor bugs that crawled out, there were no special problems. Every day we went to appstore to look at our brainchild on the main page with tears of emotion and be happy for ourselves. But after five days, the number of downloads began to decrease (there was a steady feeling that all interested visitors to the appstore had already downloaded the game, and the people simply ended. Although who knows, maybe the name, description or icon is to blame?), And gradually we fell out of the Top- 10, and then from the Top-100.
Despite returning to almost starting positions, things are now much more fun - there are still a couple of thousand regular players from the iPhone, so complaining about life would be bad. But for us it remains a mystery what exactly keeps some applications in the top for months. It’s clear why Opera or Yandex.Maps are located there - but often frank junk with a rating of one and a half stars hangs in the top for weeks, and what can hold it there? Apparently, “nothing good, but everyone wants to see this personally” (c).
And finally - about these same ratings. After the first release, we found a powerful plug in the ratings: the application scores were completely polar, which is quite rare. The explanation is simple: the game, generally speaking, is not for everyone and requires some tension to get used to, so a rocker who does not understand right away with anger removes it with a minimum rating, and players who break through unusual and grasped the essence usually put the highest score. Everything would be fine, but the ingenious feature of iPhoneOS suggests rating the application only when it is removed (and what score can be expected?), But if it is installed and actively used, then there is no easy way to evaluate it. Having decided to correct the injustice, in the latest version we added a quick link to the client’s review page in the appstore, as well as a one-time request crawling out after a week of the game to rate it. The result is enchanting and shown in the figure below. Not that the rating of the application was of particular importance (unlike the paid versions, the free audience has much more and rarely knows what they want), but 90% of five-point reviews pleasantly warm the soul.
We summarize. The first experience of crawling onto a mobile platform with some stretch can be considered successful. Players are slowly coming, reviews are being collected, millions are accumulating, the client is regularly updated (next week it is planned to release the next version with a bunch of new amenities). Along the way, from this whole venture a full-fledged API was born, with the help of which especially advanced players make clients for other platforms. Well, in the meantime, we continue to bear plans for world domination.