The Ukrainian has created a weather forecast service. It was downloaded 3 million times.

Ukrainian developer Oleksiy Schastlivy worked for MacPaw, and in his spare time created the RainViewer weather forecast application. It predicts precipitation with 90% accuracy based on data from weather radars, it has been downloaded 3 million times since the beginning. AIN.UA editor tells the story of the project.

How RainViewer began

Olexiy started working on RainViewer back in 2012. The first trial application showed data from the radar in Boryspil.

“On this everything died out. The next time I returned to the iOS application, I came up with a way to show a lot of radars on the map at the same time”, – says the developer.

He spent five months working on the application in the evenings and nights, after the main job at MacPaw. The first full version of RainViewer for iOS appeared in 2014, it included a map with radar readings.

How the application developed

On the first day, the app had 250 downloads, but their number quickly dropped to 10-20 per day. This did not prevent Olexiy from continuing to work on the application in his free time. From the very beginning, his wife helped him: she was engaged in marketing and promotion in social networks, support and localization of the service.

Within 9 months, the application had its first fans: the developer began to receive 1-2 letters a day. He was asked to write an Android version.

“Me and Android are two opposites at that time, but I pulled myself together and over the summer I wrote a simple application”, – says Olexiy.

At that time, the peak downloads of the app rose to 100 per day. It earned money through advertising and paid ad rejection.

A year later, in May 2017, there was a turning point. The author of the app translated it into 12 languages and saw good growth in downloads and ad payments as a result. The app brought in $ 500 a month for the first time. At that moment, Alexey decided that he had to leave his main job to deal with RainViewer.

Downloads increased to 1000-2000 per day, the developer found among his acquaintances those who were interested in the project: so in almost three years the team grew to 10 people.

How RainViewer works now

Over the entire period of its operation, the application has been downloaded about 3 million times. It is installed on 700,000 devices, and active users (depending on the weather) – from 70,000 to 200,000.

The app has a free API and is used by two big, well-known weather forecasting companies: Windy.com and Tiempo. Besides them, about 30 other services use the RainViewer API, including Flowx, Appy Weather, Weather NOW, World Weather, WeatherKitty and others. There is even an application that completely copies it: RainRadar.

On June 30, 2020, the second version of the application was released – RainViewer 2. It provides a minute-by-minute weather forecast for 60 minutes for all 90 countries supported by the service. “The closest competitors either work at 30-minute intervals, or work only in the US,” says the developer.

For forecasts, the service uses the PySTEPS system, it provides an accuracy of 90% (for now), which decreases to 55% in the 60th minute.

On the first day of launch alone, the application gained 20,000 new users (according to the App Store and Google Play stores). According to Alexey, there is a demand for the application all over the world, from Thailand to Brazil. In Thailand, the app is still in the top 10 weather apps.

Interestingly, the application is gaining popularity in those countries and in those periods when large-scale precipitation occurs:

“We shot hard in Saudi Arabia in their rainy season. Local weather photographers were delighted to finally have an app that shows the nearest hail on Google maps. Then there was Brazil, with its afternoon showers, this spring and summer we gained popularity in Spain and Poland”, – says Olexiy.

Plans: to further develop the project, improving the weather forecasting capabilities. The application grows organically: the team hardly invests in advertising, only tests Apple Search Ads.

“Our main idea is to teach people to use maps from weather radars, to make them available to everyone, because it is from these data that you can understand whether it will rain at a certain point or not”, – says the author of the project.

Source: ain.ua