The first version of Taxi Explorer was a free app for calculating taxi trip costs in 10 cities in Asia, Europe and North America. There were no in-app purchases or ads or any other kind of monetization built in. I named this version “1.2” to make it look a bit more mature.
I was impressed by the results during the first month. Even after the usual spike the first week after release, the installs kept in the range 250 – 300 each day. I didn’t expect daily rate in the hundreds for such a simple app. If the app regularly reached 300 installs per day for mere 10 supported cities, I was wondering how many installs would be for 100 or 1000 cities. At this point I firmly decided it was worth developing the app further.
I had built a simple feedback menu item in the app – it opened the default mail application. At the third day I started receiving support emails. The very first one was a request to add Jakarta – the capital of Indonesia. I put Jakarta on top of the to do list to honor the request. The second email was especially pleasing – it contained just two words “Brilliant app”. I replied to those emails the same day personally. Till present day I try to answer emails as fast as I can and never leave requests without reply.
I felt encouraged and enthusiastic. What does an enthusiastic developer do? He starts developing! I had cut some corners to ship the version sooner. Now, given my plans to support 1000 cities, I had to make things scalable. I put up a simple system to monitor taxi tariffs for changes. I made tariffs and fare calculation algorithms serializable. I configured a simple web interface to enter taxi service data. I added automated tests for each taxi service to verify rate integrity.
All these features were not easy to implement. I took some time off for Christmas and New Year, then I continued working on them. I was ready in March. During all these months I kept an eye on the analytics. Install rate continued in the range 200 – 300 per day. In February it was more often around and under 200 and rarely reaching 300. In March it felt considerably to 150 – 200. It was high time for update.
This time I wanted to test if the app would earn money. I added 18 new cities (including Jakarta), but I locked 13 of them behind an in-app purchase. The 10 cities from the first version and 5 new were available with the free download. The user had to buy single one-time in-app purchase to be able to calculate fare estimates and monitor trip costs in Live mode for the rest 13 cities. Since I still considered the app too simple, I priced the in-app at the lowest possible tier available – $0.99, with even lower cost in China, India, etc. At this early stage I didn’t think about profits, my point was to test monetization scheme.
With 200 installs per day and conversion rate 0.5%, this would mean 1 sale a day. $20 a month (after Apple’s cut) is not much at all, but it’s a starting point. I put this value as my baseline expectation. And I hoped more cities would drive more installs.
Taxi Explorer 2.0 is out: 28 cities supported
— Taxi Explorer App (@taxiexplorerapp) April 10, 2016
I named the new version “2.0”. It went live in the App Store on April 9. As with the previous release, the first days looked promising. The app started generating 1-3 sales each day. The installs went back up above 200 per day. A steady drop below 150 occurred in the last days of the month though. The sales stopped too. All in all, the app made 22 sales in April. Given that it missed the first 9 days of the month, it met my target precisely.
In May and June my analytics graphs went south. Installs dropped to 50 – 100 a day. There were 18 sales for the two months combined – well below target. As I said, I considered the app too simple and I thought supporting more cities would fix sales. I had received lots of requests about booking through the app. I didn’t have the time nor resources to implement a proper booking mechanism, but I added shortcuts to Uber and Lyft apps and taxi companies phone numbers. I added 23 new cities of whom 2 free, the rest joined the locked group. I increased the price a little bit to the standard Tier 1 – $0.99 worldwide.
The new version 2.1 was released on June 29. It was the same scenario all over again. A slight bump up in installs in July, followed by a drop afterwards. Total sales for version 2.1 till yesterday – 22. There was a modest spike on Aug 5, but what had caused it remained mystery to me.
In September I finally realized they were not the features that were lacking. My monetization scheme was broken – it was getting in the way of the adoption rate. 99% of all estimated routes were for free cities. I expected free cities to be popular, but I assumed more users would unlock the extra cities. Obviously a new user looks for a single city and if it is locked or not supported, this results in immediate uninstall.
Here is the time to give credit to my unbelievable marketing efforts. During the whole past year I was so obsessed over development that I didn’t spend a second for marketing. I announced the first version to the world by a single post in Facebook and one tweet. That was followed by silence between updates, when I also published few Facebook posts and tweets. Though I had a landing page, I did no publicity for it. In fact it didn’t occur to me even once to check my own page traffic analytics. I’m looking at them now and they seem pretty poor as expected. If CIA or KGB or any other secret service would like to release a secret app in the App Store in a way that no one discovers it – I’m the man who knows how.
Enough talking, here are the raw statistics after one year in the App Store (Oct 1 2015 – Oct 10 2016):
- App Units: 62 441
- Sales: 62
- Revenue: $53.80
- Rating: 3.8 based on 11 ratings
- Support emails received: 76
- Facebook posts: 12
- Facebook page likes: 7
- Tweets: 8
- Twitter followers: 5
- Landing page views: 8667
- Landing page visitors: 1778
Yesterday version 3.0 went live. As tradition goes, I published one post in Facebook and one tweet. No more locked cities, all cities are free now. I put ads inside the app to compensate. The in-app purchase will remain for the ability to remove the ads, this time it’s priced at $1.99. I want to say thank you to everyone of these 62 people for buying my app. These people are upgraded with the new in-app item too, so they will not see ads.
My main goal is to increase daily installs. I don’t care about revenue, but I’m curious what the ads will bring. This time I’m monitoring app and website analytics closely. I’ll analyze the stats thoroughly to measure my effectiveness. I hope my graphs turn north. Whatever the result, I’ll keep you posted here.
If you want to see version 3.0 stats and revenue, see the next post from Taxi Explorer series: How Much Do Users Like Ads.