How to release a great iOS application that someone needs

The motive for writing this article in something was served by another habratopic "On how to release a great iOS application that nobody needs." I will try to talk about how to release a great application that someone still needs, but which did not work.

Another motive is to sort everything out for yourself. Yet the story is rather voluminous and until you systematize everything on paper, it is very difficult to realize all your mistakes. Constantly you will think that you did everything correctly, and it will seem that negligent users who feel sorry for the dollar are to blame for everything. And the brain will push its miscalculations to the far shelf. Well, some less obvious things may be prompted by the habrowade.

It’s not for me to judge whether this story can be considered a success-story. Probably not yet. This project is my success as a programmer and a project manager in one person and at the same time it’s my failure as a marketer and a person who does not mind making some money.
And, of course, the unequivocal success for my wife Alexandra and her brother - the design of the application in question, this is their first mobile application design, which gave way to further interesting work.
')
Also for fans of numbers I will say - all the numbers in the article will be . I’ll tell you how many downloads, how much is earned, how many inapps have been bought and how many times they have been tried to be broken.

Development

There is no definitive answer to the question “why the converter”. For example, at the time of the decision to write an application, I did not like any of the paid or free currency converters. But besides this, there were, of course, "hidden" reasons. For example, I really wanted to have my app in appstore. It so happened that under the iPhone, I started writing from the very release of the SDK (and I can’t stop everything), in the middle of 2008, but I did it mostly to order and “for my uncle”. And I wanted to do my own right from the very beginning to the very end, to bring my product to the end user and see that the application was used by real people.

One way or another - currency converter. Initially, it was decided to focus on usability - yes, there are a lot of converters in the store that do about the same thing - they suck in courses from yahoo.finance and give something that can be used exclusively as “well, probably, about that’s what happens” . Differences in the exchange rate of the ruble to the dollar relative to the rate of the CBRF can reach one ruble, which, of course, cannot please the user.



In terms of features, everything was easy: they took the most popular currency converters, wrote out the killer-features of each of them and tried to implement them all in their own way. In the first version, this turned out only in part. So, the main features of our application were:

• nice and localized interface (now - 19 languages);
• iPhone and iPad support;
• selection of currency exchange sources (currently more than 20 different sources, including the Central Bank of Russia and Sberbank);
• multiple currency conversion at the same time (4 for iphone / ipad and 6 for iphone 5);
• simple calculator;

From what the application clearly gave way to competitors, there were two points. The first is graphics. With the graphs in the first version did not work for one reason only - because of the support of different sources of courses. It is unlikely that the user would be interested in charts from yahoo.finance, if the source of the courses he has selected the European central bank. And if, for example, for the CBRF, the European Central Bank and several other sources, it was possible to obtain exchange rates for a certain period, then the vast majority of banks had APIs only for current rates, no history. Thus, the solution to this problem was postponed until the second version of the application (and solved it).

The second point that was lost to competitors is portrait support for iPad and landscape for iPhone. It seemed to us not very important, but some users complained. This problem has not been solved so far and, apparently, it will not be solved already.

"Promotion"

Marketer of me useless :). However, I described some “strategy” of the launch for myself in my brain: the very first month we are distributing the application for free, dialing the user base, and in version 1.1 we add support for the iPad and we will ask for it $ 0.99.
For the month that has passed since the release of version 1.0 before the release of version 1.1, there were about 3,000 free downloads. I agreed on a free review on AppleInsider.ru, created a post on leprosy, but in addition to this, the release was almost unnoticed.



The paid version was tried to “unleash” in some way: I wrote devstory for iphones.ru (for free, I had to tinker, but it was worth it). Two days of sales after the publication devstori were the most profitable days in the history of the 1st version of the application. The application has risen to the top 50 in the Russian appstore, which brought us a total of about 150 euros for these 2 days, and for another week there was some kind of fading effect.

The review on iPadStory.ru was also very helpful, also free of charge (the owner Dmitry was a very pleasant person to talk to). In return, the publication went even better than iphones.ru: it brought about 200 euros. 50 immediately after the review, and at the end of the year, the application fell into the list of the best applications of 2012 and immediately after the publication of this list, sales grew quite well - another 120-150 euro was earned on this.

A good effect was from the publication on Deepapple.com and AppStudio.org. And again for free.


Points on the graph from left to right:

• release 1.1 (the leftmost point);
• another review on AppleInsider.ru (the first and last paid review of the application);
• iPadStory review;
• devstory on iPhones.ru;
• review on DeepApple.com;
• release 1.2 (the rightmost point);

As you may have noticed, all sites are Russian. Not a single owner of English (Italian, French, etc.) linguistic sites has not published a review at my request. Probably, I somehow didn’t write to them wrongly, I went in at the wrong door, but, most likely, currency conversion there is just a less popular thing.

There was also a handful of incomprehensible Russian sites that wanted some money for a review (range from $ 20 to $ 150), but I decided that it was necessary.

Competitors

At the same time, another currency converter, Vkurse, began to unwind. Very nice, with a denim interface. At first, they tried to “spud” the user with in-built purchases of skins - jeans of different colors, the application itself was free, but later they refused this idea and began to ask for a dollar for the application. The idea was not very clear to me - the application was very poor in features, but people liked it - often in the ranking of financial applications they turned out to be higher.

Of the “successful” competitors, I can also single out the “Currency!” Application, which started a few weeks after us, immediately went to the main section of New & Noteworthy and, judging by the appannie statistics , it still sells better.

Sales

In the summer of 2012, two sales occurred. The June sale for the day doubled the number of users - from 3.5 to almost 8 thousand, but I, of course, did not receive a single cent from it, and a month later I dropped the price to zero for three days.

I will dwell on this point in more detail, because the effect was very good: in three days the application was downloaded more than 70 thousand times, several reviews were written in Polish, Portuguese, English, French. In many ways, lucky, of course - I got a representative from the site appgratis.com and offered to place the application on their sites (all but American) in exchange for what I write in the application description “today for free thanks to appgratis.com”.
Unfortunately, Apple at that time changed the application ranking algorithms in the appstore and we did not get into the tops of free applications anywhere. That is, they disappeared from everywhere — even in the financial applications section, we were not. Although 10K + downloads in Brazil, Spain, Germany, this is at least the top 3 free in these countries. But not lucky.



The effect of the sale was very good: if before it, on average, the application was bought 12-15 times a day, then immediately after over two three-days sales were at the level of 80-100 copies, and then the decaying went up to 30-40 copies day. Something like this remains to this day.

Thus, by the end of the summer, the application had 80,000 downloads in the asset (~ 10K sessions every day), about 2,500 euros earned and very vague prospects.

2.0

Surprisingly, I decided to try to “finish the” application. Yet it is very difficult to surrender and put aside something, on which I spent a lot of time, energy, and that I started to grow from the very "int main ()". It was decided to try to monetize “run-by” users on sale with in-app'ov.

They chose a simple strategy: we release a major update (version 2.0) with a store, where we will add a number of in-app bucks and one “cumulative” purchase, which will be cheaper for a dollar or another than all the others. The features for sale are as follows:

• graphics;
• date selection for conversion;
• "Additional" sources of courses (PayPal, Mastercard, Visa, Sberbank, Bitcoin);
• push notifications when updating courses on the server;
• the ability to set currency rates manually;

A bot for collecting currency rates was written and set up, a server providing history rates for a given period of time.
In general, it was very happy to break away from Objective-C, which had already become bored, and pee a little on favorite “non-browser-based” javascript. I decided to make maximum use of technologies unknown to me until now, so the choice fell on node.js and mongodb.

Another interesting idea came (and most likely, I read it somewhere, but now I can’t remember exactly where) to me about a month and a half before the release. One of the built-in purchases to give for free, provided that the user places a similarity of the advertising message in his twitter, facebook, VKontakte and a couple of other social. networks to choose from. In my case, there were no two opinions on which of the in-apps to choose: 4 out of 5 used and, accordingly, loaded my server, and I have to pay some money to Amazon for it. One of the purchases — user exchange rates — did not create a load. ShareKit was quickly connected to the application, I found a girl on leprosy who wrote me a dozen different phrases and the question was closed.

Adding in-app'ov solved another problem - the problem of subsequent sales. Understandably, in-apps are not needed by everyone, but a certain percentage of users who get the program for free will buy them. Everything is nicer.

Apple's “spoiled” mood a little bit - it turned out that you cannot take money from the user for using push notifications, so it was decided to refuse this built-in purchase in order not to delay the release.

In addition to in-apps, the application has added support for a different buying and selling rate (i.e., $ 1 in a bank will give you 30 rubles, at the same time you buy it there for 31), various cosmetic trifles that make life more pleasant.

And now...

So, in the middle of March 2013th year, exactly one year after the release of version 1.0, version 2.0 was released. The first version for the year of existence earned 6 thousand euros. The total costs of transfers / devaccounts / sites (for both versions) amounted to about 2.5 thousand euros.



The results of the first and second day of sales of the second version of the application were encouraging, but to millions, of course, far away. Profits increased 5-6 times: 90 and 125 euros per day, respectively. It was made quite a lot of built-in purchases and, which is nice, mostly bought the very “cumulative”.

However, the effect quickly faded away. In the comments to the application began to receive complaints about in-app'y: they say, paid for the application itself dollar, and here also require money inside; and also the fact that perfect in-apps are not given (I made in-app checks on the server on the recommendation of Apple and did not miss invalid purchases). Apparently, someone really got used to the fact that any in-app on a hacked device can be obtained for free and the reverse situation angered the person to the depths of his soul so much that he found it necessary to complain about this in the comments.

If you look at the conversion, it is very good - about 5-7% of new users who run the application make a purchase inside the application. Another thing is that the number of new users is quite small and there is no certainty that with more figures, the conversion will not fall.

60-70 percent of the total number of purchases within the application - an attempt to steal. And this is taking into account the fact that the application itself can not be stolen . That is, the percentage of jailbroken devices on which the application runs on average less than others. I don’t miss similar in-apps on my server, but I think that the figure itself may be interesting to someone.

Hopes for "word of mouth" also did not materialize. Many smart people (~ 60%) after sending a message to a social network immediately delete it. In principle, this behavior is understandable, but this is no less unpleasant.

At the time of this writing, there was one untested option: to make the application free and see whether the influx of users who buy in-app will cover losses from reducing the cost of the application. I will definitely try this option early next week. But in general, I do not think that it will give any tangible positive effect.

Conclusion

Application support, of course, will be implemented as long as it is profitable. Nevertheless, 1000 euros per month is a good passive income, and it is quite possible to do something for this. At the same time, if you take a sober look at the costs, you can easily scream - a year of programming (even in the evenings, weekends and fragments), a little less time is invested in the design, but this is still not a pound of raisins. And although I didn’t count the total labor costs exactly to the penny, I would rate them by the eye of thousands at forty euros. The bottom line is a hard minus.

However, not everything is so sad. First, a number of new technologies have been studied. For example, I wrote my first Python script to communicate with iTunesConnect via the command line. If only you knew how many mats I folded while I poured pens on localized screenshots for all devices and 20 different languages. And the script is fyut, and in 15 minutes everything is ready without my participation.

Secondly, there is such a moment as the satisfaction of the finished product that people use. Yes, I already had one (written from the very first line of the code, counting personal finances), but this is not my own, and now I don’t have any relation to it.

Thirdly, the project, which was originally written for fun brought, and continues to bring some profit. Sitting and playing in the evenings at kongregate is not the best alternative.



Well, fourthly, of course, it is very nice that it turned out to realize what was originally set for itself as the main task: to make a high-quality application that people would like. 88% of reviews to the application - five stars, another 8.5% - four. Unfortunately, in-apps have slightly spoiled this statistics: users cannot yet realize that good is rarely given for free.


In parallel with the development, I made a lot of new things for myself in terms of application localization: how best to do it, which languages ​​to support in the first place, and most importantly, how not to do it. Well, I did not miss the chance to share my ideas with Habr.


If we talk exclusively about the financial side of the matter, then, of course, I would have considered success to be everything that would bring more than 200 euros per day on average. In this case, I probably would have tried to continue to improve the application and to do something. And so ...

It's time to put a fat point. Here it is .

I would like to thank hackrauser jeck_landin , twitteruser @ivanovaos and @ 5g6r for their help in preparing and reading the article.