Announcing Countly services - or, “Release the Kraken!”

Hello Countly fans!

Today we are very proud that open source, real-time mobile analytics platform Countly is announcing general availability of analytics services and products, providing valuable insights of how users interact with mobile analytics and games. Our service is now out of beta, offering affordable, advanced analytics for companies requiring a stable and feature rich solution for mobile, as well as desktop and Facebook apps. 

As you all know, we have been providing sophisticated web based analytics capabilities for apps, through an easy-to-use graphical interface that greatly simplifies data collection and analysis. By tracking in-app activity, developers can focus on issues, improving in-app usability, increasing user experience hence retention and in-app purchase sales.



As of today, we renewed our web page (which in fact caused a lot of vibe among graphic design sites), offering quite an extensive list of services, including training, different editions, paid services and partnership opportunities. We extended our documentation, added datasheets andpresentations, and many screenshots who love to watch visuals.

Until now, we have received hundreds of testimonials, some of which came from Twitter directly, and some directly dropped in our mailbox. One of them is from Ygor Lemos, CTO of famous AirportChatter app: “Countly enlightened our path to knowing our users behavior and gave us mobile metrics that we simply didn’t have before. We chose Countly not only for its beautiful UI or the very flexible API, but specially because the awesome support and the startup spirit of the company that surely is in tune with our own.”. So refreshing.

Working well over a year, drinking 300+ liters of coffee, having 200+sleepless nights, our Community Edition is now trusted by 750 system administrators on 65 countries, collecting analytics data from 60M devices. An incredible curve from a minimum viable team!

Together, let’s build world’s most extensive, user-centric and powerful mobile analytics platform. Join us now and share the love.

— Countly Team

Hacking with Countly, Raspberry Pi and a servo motor

Countly is a perfect tool for hacking. This time Paul Brown, a veteran Countly hacker, combined his Raspberry Pi hardware with Countly to build a cool toy - an in-app purchase flag. 

Paul used a small Ubuntu server, Python as the programming language and a simple servo motor from Sparkfun and a lot of coffee to make it happen. Now, every customer of his Easy Grade Free app who prompted to buy an asset from in-app purchase, but didn’t, makes the orange flag raise.

Paul also provided the source code to pull events from Countly server in Python language. Head over to Github to grab your code. You can also watch the video of a real-world servo running as it catches an in-app purchase attempt:

This nice example shows how Countly API can be used to make fun projects. It’s really a brilliant hack, and gives us lots of clues in Countly HQ to make ourselves (we might even copy his idea, shamelessly!) :-)

Building a mobile application engagement scorecard

Many companies have their own mobile application development teams, working on several projects at the same time. Sometimes outsourcing the mobile development needs can also be an option, decreasing the operational expenses at the expense of depending on third parties. Mobile development is becoming a requested skill and due to abundance of development platforms, operating systems and programming languages, it may be hard to find a specific, focused skillset for a project.

Whatever the solution is, companies are looking for a good performance model that can be used as a mobile application success index. C-level execs are depending on certain metrics like downloads and installs, amount of usage, in-app purchases and other numbers that are simple to read and interpret. However some statistics hide crucial information about app’s performance and user experience, but they are neglected due to lack of information and awareness. This hinders potential issues found in user experience, backend, graphics and applications purpose for a use in general. 

Below you’ll find a scorecard listing most important engagement metrics for an application. Use it as a basis to calculate the overall performance of your mobile applications. You can add or remove extra metrics, e.g amount of in app purchases, or assign weighted numbers in order to calculate the total score for a given application. 

1. Number of downloads: This is by far the most important metric you need to analyze. If possible, try to track the referrers for your application. Under Android, campaign tracking is straightforward,  however Apple doesn’t allow third parties to track referrals (but there are different methods for this such as IP matching).
2. Number of uninstalls: We’ve seen apps with more than 160% uninstall rate - every 10 times the app is installed, users uninstall it 16 times. Uninstallation can be tracked from push notification feedbacks, that is, if an application enables push and if it’s uninstalled, then you can count it. 
3. Retention: Retention simply tells how many users come back to open your app just because they embraced it. Just like repeating customers, retained users are one of the most critical metrics in application engagement. With retention screen you get an insight of percentage of users coming back to use your app again. 
4. Store comments and ratings: Though not directly available in Countly dashboard, there are several online services providing this information by scraping application store web. Honest opinions from end user should be taken seriously by the product manager and considered as an improvement area.
5. Number of opens: Statistics show that one fourth of all apps are opened only once. A small frustration, an issue with login, a typo in help screen will force user press home key, never returning back to that app again. Fun stops. Thanks to Countly dashboard, you can see how many times your app is launched, and take necessary precautions in case average number of opens is alarming.
6. Time spent: This metric can easily be misleading, since time spent in a simple utility (e.g weather app), a casual game or a video streaming app can vary greatly. Therefore do not simply compare your “time spent” data with other apps, but rather benefit from custom events (e.g number of channels watched in a session) to correlate both data. 

These are the most important insights you can note if you want to build a scorecard for your application set. This list is no means exhaustive, and you are encouraged to add or remove other parameters you see fit (e.g in-app purchase). 

Make sure you assign weighted points (e.g 15 for number of downloads and 20 for retention), summing 100 in total. Provide a simple formula for each item and define how you’ll score them. For example, getting 1 star is 4 points, 2 stars is 8 points etc. 

Finally, fill this scorecard periodically (e.g every week or month), so you’ll get a consistent sampling for each application. You’ll be amazed with the insights gained after noting down a few samples already.

You are what makes Countly successful

Thank you.

Because you believed in us, our mission and vision, to make mobile analytics truly open and private for everyone.

You’ve given us inspiration - to build a community around Countly, a project that acts autonomously and socially responsible.

You supported us with our key values - freedom, usability, quality and business readiness.

You made Countly possible - with your feedback, trust and collaboration. 

We are grateful for your help.