Mobile application analytics tools. Comparison of mobile analytics systems: Mixpanel, Amplitude, Flurry, Localytics Analytics application

World market

Demographics play a significant role in the development of mobile analytics: more and more users within companies are familiar with mobile devices and the Internet from school, so as the staff of organizations gets younger, they will be the ones who will manage the further consumerization of IT. Gartner predicts that 33% of analytics will be used in the pocket by 2013. By the end of 2012, 55% of organizations using BI had already started or planned to implement mobile data analytics. As a result, by 2015 the ratio of mobile and traditional BI projects will be 4:1.

According to a 2012 Dresner Advisory Services study for MicroStrategy, the role of mobile analytics has grown exponentially from 2010 to 2012. Thus, the number of study participants who are confident in the exceptional importance of mobile analysis tools increased from 52% to 61% over the indicated two years.

Mobile analytics penetration is uneven across vertical markets. Retailers and other trading companies give maximum importance to mobile BI; it is in this industry that there are already many positive examples of the implementation of mobile analysis tools. On the contrary, mobile analytics has a fairly low priority in the public sector, although there is successful experience here too.

Most study participants noted the penetration level of mobile analytics in their organizations at 10%, however, it should be noted here that in 2010, for the majority, this figure tended to zero. Companies have quite aggressive plans for the implementation of mobile BI solutions until 2015 - they are shown in the diagram below.

The study also looked at 25 key BI vendors offering mobile analytics solutions. The results of the analysis of these systems are presented below; out of a maximum of 36.5 points, the highest results were shown by MicroStrategy, Yellowfin, SAP and QlikTech.

Dresner Advisory Services, 2013

According to research company Ovum as of January 2013, the most functional solutions on the market are offered by MicroStrategy, Information Builders and Yellowfin, while mega-vendors like SAP provide the best mobile analytics systems for iOS devices.

As for mobile platforms, all BI vendors, without exception, support the Apple iOS platform; Android is in second place in terms of vendor support (90%), followed by the BlackBerry (50%) and Windows (33%) mobile operating systems. The Ovum report does not take into account systems running Windows 8, RT and Phone 8.

According to Frederica Tanwalla(Fredrik Tunvall), Ovum analyst, there is no doubt that the success of the iPad in the market has a significant impact on the product strategies of BI developers. Meanwhile, for mobile BI solutions to reach a larger audience, vendors will also have to create applications for the Windows and Android platforms, which are most widely used, especially in emerging markets.

While SAP has a head start in the mobile analytics market thanks to the company's acquisitions of Afaria, Syclo and Sybase, other vendors such as IBM, Oracle and SAS are also gradually gaining market share. To maintain their solutions at a competitive level, vendors will actively use an M&A strategy, says Tanvall. He also believes that innovation in the BI and analytics market will peak in the next few years.

Russian market

The growth of mobile analytics in Russia is accompanied by growing user interest in mobile versions of other classes of business systems. As the penetration of related hardware in the corporate segment increases, this phenomenon will eventually become widespread.

Thus, according to IDC forecasts from February 20013, in 2013 the Russian smartphone market will grow by approximately 15%. IDC believes that the Russian smartphone market is still far from saturation: the growth trend will continue for another two to three years, but by 2015 smartphones will not occupy 50% of the total phone market. The share of smartphones in the Russian phone market in 2012 increased from 25% to 33%.

In unit terms, at the end of 2012, 12.3 million smartphones were sold in Russia, reports J’son&Partners Consulting. Compared to 2011, sales growth reached 52%. The share of smartphones in total sales was 30%.

The development of the Russian tablet computer market also plays an important role. According to Euroset, in 2012 it showed the maximum increase among all portable electronics - 366% and 256% in quantitative and monetary terms. According to various estimates, in 2013 tablet sales may reach 2-3 million units.

Today, almost all major BI vendors have mobile BI products in their product lines. Some vendors, for example, Oracle, have all the functionality of the main version of the analytical product available in the mobile version. Most of these solutions and tools are already available to Russian customers, which is an additional incentive for the development of mobile analytics. Unlike, for example, open source BI, where many serious solutions are still not presented in Russia and therefore cannot be evaluated by potential customers.

Market participants note that if in 2011 mobile BI was more of a “nice addition” to the main project than an essential tool, then in 2012 the presence of such an application increasingly became a mandatory requirement for customers. Moreover, if at first the main users of mobile analytics were top management, investors and company owners, now projects are being implemented in Russia where mobile analytics works “in the field” - for example, with representatives of trading companies.

Vladimir Itkin, director of development of the Qlik partner network (QlikTech) Russia, confirmed:

“Today, every second client when choosing a BI platform and solution has a mandatory criterion - work on mobile devices. Just two or three years ago this was more of a bonus to the platform than a necessity. But mobile technologies have firmly entered our lives, and now a tablet in the hands of an ordinary employee is not a luxury.”

According to him, the modern business user can no longer wait for the report to be prepared and sent to him by the IT service staff. He is already accustomed to receiving information at the right time for him, using the Internet and search tools, for example, Google. Requests are starting to grow, and mobile access to aggregated reports is also not enough; there is a growing need to see the detailed level of data. The speed of decision-making in a competitive environment depends on this, which will be especially important after Russia’s accession to the WTO.

Vladimir Itkin noted another interesting fact: recently, more than half of the pilot projects of QlikTech partners are tested by clients on mobile devices. “That is, before the final decision to purchase a platform is made, the client wants to make sure that the solution being created will not limit its capabilities in analysis when used on mobile devices,” the expert explained.

Sergey Shestakov, deputy general director for business development at Prognoz, told TAdviser that the company is very actively using mobile tools in its projects. In particular, Prognoz creates mobile versions of situation centers, program-targeted control systems, modeling and forecasting systems. At the same time, solutions based on the mobile version of Prognoz Platform allow you to work with information both online and offline: for example, on an airplane. All this functionality is implemented in the standard software product “Electronic Executive Office”, which the vendor brought to the market. So not only Western platforms have mobile BI tools.

On the other hand, added Sergey Shestakov, mobile analytics for mass consumption is practically not represented on the Russian market. “We regularly create and develop similar solutions for international organizations that use mobile applications to widely disseminate macroeconomic, environmental and other statistical data,” he explained. The IMF eLibrary, World Bank DataFinder applications and their various modifications are available for free download in the App Store, including in Russia.

The Forecast expert believes that similar analytical products, designed for an unlimited number of end users and aimed directly at the Russian market, should appear in the near future.

The widespread use of mobile applications in corporate environments is hampered by security policies and company concerns related to the transfer of their data over mobile networks.

Examples in action

An example of the implementation of mobile BI is a mobile application for iPad based on the Prognoz Platform, which is used by the management of the Federal Tax Service of Russia. With its help, detailed information is displayed on the activities of the departments of the Federal Tax Service, operational indicators on tax revenues, debts that are formed in the department’s information system. The head of the Russian tax service, in particular, used this application when reporting to the Prime Minister of the Russian Federation.

Also, Dmitry Medvedev was recently presented with a solution created by Prognoz for monitoring investment activities in the North Caucasus Federal District. It includes, among other things, the investment portal of the North Caucasus Federal District and its mobile version for iPad, which contains information for investors about investment opportunities and sites in the region, as well as about existing government support measures and investment projects already being implemented in the North Caucasus.

In regional administration, the use of iPad applications is very common - the number of administrations of federal subjects that order similar solutions from us is in the dozens. For example, the leadership of the Lipetsk region was one of the first to use mobile technologies in daily work, using them to access a single repository of all regional information. Today, the Information Information System of the Lipetsk Region operates more than 20 thematic subsystems (on territorial development, agriculture, social support, energy saving, etc.), on the basis of which summary analytical materials are provided to regional leaders via electronic tablets.

Tracker— a mobile application analytics system with the ability to track the source of installations.

There are many things to consider when choosing a tracker. First of all, it’s worth understanding that there are two types of tracking: a system for tracking the install channel, as well as in-App analytics for analyzing demographics, geolocation or the length of a user’s session. The principle of tracking is simple. The user clicks on an advertising banner (video, text ad), a redirect occurs, and the tracker records the click in the system.

Screenshot from Yandex's presentation on the work of AppMetrics

What is an SDK?

Regardless of which tracker you choose, you need to implement it using the SDK.

SDK(software development kit) is a set of development tools that allows software specialists to create applications for a specific software package, core development software, hardware platform, computer system, game consoles, operating systems and other platforms.

Simply put, code is generated inside each tracking system that will transfer data between the store, the application and the tracker itself. If you work only with the App Store and your application is paid, you can limit yourself to the built-in analytics of iTunes Connect. Then there is no need to implement anything additional. You can create tracking links right there. The algorithm is simple. 1. Go to “application analytics”.
2. Select the desired application. 3. In the “sources” section, select “create a link to the campaign.” 4. The system will transfer you to a new tab, where you:

select the application for which you are creating tracking;
indicate the name of the advertising campaign;

copy the link that the system generates.

On the “Sources” tab, in the “Most Popular Campaigns” section, statistics on advertising campaigns will be displayed. Important:

App Store analytics collects data only for devices with OS 8 and higher;
The attribution window is 24 hours for campaigns;
To display statistics there must be at least five unique application downloads.

Attribution— a method of transmitting the source of a click or conversion. Attribution Window— time interval between clicking on an ad and installing the application.

Which tracker should you choose?

Free solutions

1.

We recommend that all clients install a free solution from Google - Google Analytics for mobile applications. If the client’s activities are presented not only on mobile, but also on the desktop web, then this is generally a prerequisite. First, GA provides a comprehensive view of traffic distribution. Secondly, you can create Google Ads remarketing lists, which is still the leader in Runet as a source of traffic. The standard attribution window, as for web versions, is 30 days. Initially, the system is not designed to work with mobile applications, so it cannot calculate full-fledged cohort analysis or funnels. However, Google strives to improve its products, and this is noticeable in the number of updates over the past year. It is worth noting that Google Analytics has a powerful set of APIs, which allows you to receive much more information than when working in the interface.

This tracking method is available for: Android, iOS, Windows, Unity.

2. Yandex.Metrica for applications or AppMetrica

A relatively new product on the mobile market, but at the same time quite functional. AppMetrica provides a symbiosis of functions that is convenient to work with - the system can send postbacks and at the same time analyze user behavior. Analytics is integrated with many advertising networks, which greatly simplifies tracking when working with a large number of sources. The system uses several attribution models: Device Identifier Matching (48-hour window) and Device Fingerprint Matching (click attribution can take up to 10 days). The first model involves integration with advertising networks and adds two parameters to the tracking link, which allow you to record the installation only after it is opened. And Device Fingerprint Matching links ad clicks with installs. Tracking is available for: Android, iOS, Windows, Unity.

3. Yahoo Flurry

Flurry and Google Analytics are competitors. Often developers are faced with a choice between these two solutions. Flurry allows you to track everything that GA does. One of the advantages of the system is that it works with events, which are quite easy to set up, and the ability to set the period for the implementation of a step when setting up a funnel. The reports inside the analytics are well organized, you can create the necessary segments and funnels. The disadvantages include the lack of calculation of financial indicators and the ability to conduct A/B testing. We can conclude that the system is not at all suitable for those who have in-app purchases in the application. The attribution window is 48 or 72 hours, depending on whether the system receives the device ID or not. Tracking is available for: Android, iOS, Unity (difficult to integrate).

4. Facebook Analytics

The system allows you to track the effectiveness of advertising campaigns on Facebook. Installs and even in-app conversions are tracked. The standard attribution window on Facebook is 28 days. This means that conversions will be credited to the ad within 28 days of the ad being clicked. This method of tracking installs can significantly distort the real picture, so it is better to change the attribution window in conversion reports. Select “customize columns”:

Tracking is available for: Android, iOS, Unity.

Review of paid trackers

A high-quality tracker is quite expensive. And all because paid solutions analyze traffic sources more accurately.

1. AppsFlyer

Paid tracker with a good reputation in the market. You can download a trial for 5,000 inorganic installs, more than 10,000 integrated partners, which makes it easier to work with paid traffic sources. Recently, AppsFlyer has learned to compare data on user activity during TV campaigns. The latter is a clear advantage for large brands that run national campaigns and want to measure their effectiveness. In addition, AppsFlyer makes it easy to measure conversions and in-app purchases, LTV, and even ROI. The cost calculator for a paid package for the Russian version of the site looks like this:
The system calculates settings differently for different channels. The attribution window for Twitter is 14 days, Facebook is 28 days (read more about the difference in counting on Facebook), other advertising systems are 7 days. Trial is present. Available for: Android, iOS, Windows Phone, Amazon Apps, Unity.

2.Adjust

If you buy a paid tracker package, you get standard reports, API access, and organic install tracking as a bonus. When analyzing conversions in the application, please note that the number of installations has absolutely no correlation with the number of conversions themselves during this period. Event attribution is tied to the installation source throughout the entire lifecycle of a given user. Use to view statistics for a specific user. Plus - flexible payment systems that depend on the number of attributions recorded by the system. For large players there is an unlimited package. The standard attribution window is 7 days. To receive a trial version for 30 days, you need to send an application by email. Available for: Android, iOS, Windows, Unity.

3. Mobile App Tracking or Tune

Satisfies almost any reporting request. Gives a clear understanding of how much income the installation brought, and also knows how to calculate the user’s LTV. The system has powerful API functionality that allows you to generate reports of any complexity. The system is so serious and extensive that it allows you to configure the attribution window for each partner separately. It is worth noting that such analytics are only suitable for advanced application owners. Otherwise, the amount of information received will simply be useless. Tune's functionality is extremely powerful, so its cost is even higher than AppsFlyer or Adjust. In addition, the tracker does not track installations from Facebook, so you won’t see the full picture. You can ask for a trial tracker on the website. Available for: Android, iOS, Windows, Unity.

4.

Like Tune, Mixpanel is a serious mobile analytics solution that is quite expensive. The site offers several tariff plans from $150 to unlimited (over 50 million events). The high cost of the tracker (for an application with volumes of 20 million events or more is $2000 monthly) is due to its capabilities:

financial analytics;
in-App analytics;
detailed segmentation;
classic funnel;
well-designed cohort analysis.

The standard attribution window is 30 days and, just like in Tune, it can be adjusted. Among the advantages of Mixpanel is the ability to send push notifications and targeted messages to any application audience. Trial is present. Available for: Android, iOS, Windows, Unity.

conclusions

We have compiled a table comparing tracker functions:

In my opinion, with the classic option of app promotion, the best option is free Google Analytics and a paid tracker. This combination will allow you to accurately track the source of installations and optimize advertising campaigns, as well as provide an understanding of user behavior after installing the application. We will be glad to hear your opinion in the comments. Write which tracker you are working with and why.

Overview of other tracking services and general summary table -

Collecting data for mobile analytics is impossible without special tools. In addition to accumulating information, we also need its systematization and the ability to operate with data. There are special analytical services for this.

There are two main areas in mobile analytics: tracking installations and events that occur before them (external analytics), and monitoring events within the application, that is, user behavior (internal analytics). In addition to them, you can also highlight application store analytics.

Install tracking

It is necessary to track application installations primarily so that you can evaluate the effectiveness of advertising. In other words, it allows you to recognize traffic sources. Tracking provides all the necessary information about the user’s path from click to launch of the installed application.

The application integrates the tracking service SDK, which allows you to collect information about installations and a number of other data.

The tracking scheme works like this:

Thus, tracking allows you to collect almost the entire pool of data that is needed to evaluate the effectiveness of advertising.

Internal analytics

Another direction in mobile analytics is collecting data on user behavior within the application. Such product analysis helps to obtain statistics about any user actions. Special services through the SDK integrated into the application record all events, be it screen scrolling or a transaction. In this case, other important parameters must be taken into account, for example, device model, operating system version, lifetime in the application and a number of others. In addition, it is possible to segment the audience according to almost any parameter.

ASO analytics

As you know, an impressive share of the audience comes to the application from internal store searches. Therefore, another type of mobile analytics is in demand on the market - collecting application store statistics. This mainly concerns search queries.

Such services help you understand which queries are most important for promotion. In addition, they make it easy to monitor keywords and monitor any changes in query rankings and even the actions of competitors.

Mobile analytics market

Previously, there were separate services for install tracking and internal analytics, but recently there has been a trend of combination: many systems combine both areas of mobile analytics. Trackers expand their functionality to track internal events. In this way, companies expand their presence in the market, and it becomes more convenient for their users to conduct analytics, since all the tools are collected in one place. In addition, analytical systems have made significant strides. They don’t just give out dry statistics, but interpret the data, which simplifies the final assessment of advertising effectiveness. If earlier analytical systems were purely technical services, now they have turned into a marketing tool. They can give users tips on or optimization of the application. For example, ASO services recommend keywords, and analytics systems provide optimization tips to improve user experience.

There is also a tendency for analytical services to converge with systems that sell advertising. Already, many of them have direct integration with advertising networks. In addition to simplifying the work, this allows you to receive some statistics directly from advertising networks - mainly the cost of impressions and clicks. Thus, complex metrics, for example, ROI, for the calculation of which the same indicators are needed, can be immediately available in the service panel.

On the other hand, advertising networks themselves have begun to enter the analytics market. If previously almost the only such service was the Google Analytics system, now AppMetrica from Yandex and MyTracker from MyTarget have appeared.

Services overview

TUNE

1 /1

TUNE - this service was previously called MAT (Mobile App Tracking), and, as the name suggests, its main purpose was to track installations. Now it has functionality for both external and internal analytics. In addition, it includes tools for business analytics and obtaining statistics from stores.

As for ease of use, the huge TUNE toolkit is redundant for typical tasks. This is where its main drawback arises: simple settings require too many operations and it is difficult to navigate the confusing interface.

AppsFlyer

2 /1

3 /1

AdJust also combines both areas of analytics and its tools are enough for a full range of work on collecting and analyzing statistics. At the same time, the service has a rather complex interface, which usually scares beginners. Indeed, sometimes even experienced developers and marketers have questions, especially when it comes to setting up event tracking within an application. But there is a plus for agencies - AdJust has convenient integration with partners.

4 /1

Kochava is a service with an impressive set of tools for tracking installations and tracking events. It is convenient to work with postbacks, it implements functionality for agencies, but there are also disadvantages. The most significant disadvantage is the very complex interface, in the wilds of which even a pro can get lost. The service also lags behind its competitors in technical terms: it has a rather low speed of transferring statistics.

5 /1

Firebase is a product of Google. Initially, this service was conceived as a backend platform for applications (as a real-time database, hosting and authorization system), but then it began to develop, acquiring new modules. Including the Firebase Analytics tool. Its capabilities are quite enough for most users. This service is very convenient to monitor user actions and analyze the activity of different audience segments. In addition, Firebase Analytics is free, which is a clear plus for indie developers and small studios.

6 /1

AppMetrica is a Yandex analytical system for mobile applications. The service has a very simple and logical interface. Perhaps AppMetrica is the most convenient to use; this service has everything you need for full analytics of mobile applications and nothing superfluous. In addition, it has another undeniable advantage - it is free, unlike the vast majority of other services. AppMetrica also has a simple API, but there is no separate partner interface.

7 /1

Flurry is a convenient service; its tools are suitable for solving almost all basic tasks of mobile analytics within applications. But at the same time, it does not implement cohort analysis and some functions are too deeply hidden in the interface.

8 /1

Mixpanel is a very powerful tool that provides the widest opportunities for internal analytics. It allows you to create different event funnels, segment your audience based on any criteria, and create complex events based on actions and demographics. Mixpanel is mainly aimed at large developers and publishers, as its functionality is redundant for beginners. In addition, the service is considered one of the most expensive.

Universal Analytics

9 /1

Universal Analytics is the next round of evolution of the Google Analytics system (it was originally intended only for web analytics). Therefore, the service is primarily intended for cross-platform products. But many people successfully use it exclusively for mobile applications. It allows you to solve common analytics problems and is absolutely free for users, but the interface is inconvenient and confusing.

MyTracker

10 /1

Services for ASO

The main task of ASO services is to study search within stores. These tools allow developers to understand how users search for apps and what they type into the search bar. Thus, ASO analytics helps you select the right keywords and ultimately make it easy for users to find your application. Such keyword optimization sometimes gives a 20-30% increase in traffic!

The essence of special ASO tools comes down to collecting search statistics. They show the search queries that users enter and their frequency. Next, the services select the most effective words for the search. They also allow you to monitor the dynamics of the frequency of your words.

One of the most popular services -Sensor Tower. With its help, you can track search statistics and easily select new words. It also allows you to spy on competitors - look at their keywords. ServiceMobile ActionIn addition to the main set of functions, it has a convenient recommendation module, which significantly saves time and simplifies work. There are also recommendations for choosing words in the service.MobileDevHq.

Analytics services are one of the mobile marketing tools. Without them, it is impossible to promote mobile applications. Skillful use of analytical systems saves time and provides a huge amount of useful information. Don't neglect analytics, improve your application!

At the moment, there are a sufficient number of solutions for mobile application analytics. Their functionality is, in most cases, similar. There are only a few services with broader functionality. In this article we will look at such a function as video recording of user behavior.

To bookmarks

What benefits does video recording of behavior provide?

Video recording of user behavior allows you to determine how clear the application interface is to the user, and how the user navigates through the application itself. By receiving such data, it is possible to determine not only the suitability of the interface for use, but also how the user generally perceives the application. How he registers, what functionality he uses, what interests him more and what interests him less. Video recording allows you to see the application through the user's eyes. This functionality will be useful for both the ux/ui designer and the application owner or analyst.

What services make it possible to record and view videos of behavior?

At the moment, there are three most common services with normal video recording functionality. We don’t take standard metrics into account; they are mostly the same. These are: English-language uxcam.com, appsee.com and the Russian service userx.pro.

Based on personal experience of using all platforms, I can highlight several main differences between all three services:

1) For uxcam and appsee, all standard metrics, including the video recording function, are paid and limited in quantity. Userx has all the basic functionality for free. You only need to pay for video session packages.

2) uxcam and appsee have one important difference from userx. When purchasing service packages from uxcam and appsee, recording management is not possible. All sessions are recorded until the package of purchased services ends. At userx you can buy a package for recording video sessions and record exactly the number that is needed, for example, for initial testing of the interface. For example, we buy 1500 sessions and can record 300 sessions each time it is necessary. Moreover, the package of services is not limited in time and expires only after reaching the purchased number of sessions. This solution is what distinguishes userx from its competitors.

3) Also, there is such an important point as the price. Experience has shown that userx has the best pricing solution.

For all three services, we can say the following: if you are just faced with the need to integrate analytics into your application, then at the initial stage, the minimum userx package will be sufficient. It will allow you to solve the problem with analytics. For a more in-depth analysis, it is enough to expand the package of video sessions.

The userx console is much simpler in interface, unlike uxcam and appsee. This makes it possible to learn basic tools without having special skills. In appsee and uxcam this will take much longer.

It is also very important that the Russian platform has Russian-language support. Its specialists quickly resolve most user problems.

Having used all three analytics services at the same time at one time, I chose userx.

But, as they say, there are no comrades for taste and color.

Find out how users who install your app find your app's page in the app store.

Install tracking allows you to collect and transfer information from mobile applications to your Google Analytics account, after which it appears in the report on mobile application traffic sources.

To set up tracking of mobile app installs, you must follow the steps below.

Content

Step 1: Enable tracking of app installs in your account

For Android

Android app install tracking is automatically included in Google Analytics. No action is required on your part.

For iOS

To set up install tracking in your Google Analytics account, follow these steps:

Click Administrator and navigate to the required resource.
In column RESOURCE click Resource settings.
In chapter Campaign tracking for iOS apps set the switch to position ON.
Click the button Save.

Step 2: Update the Google Analytics Developer Kit

For Android

Android developers need to change a few lines in the manifest file. An example is provided in the Android Developer Guide.

For iOS

Make the necessary changes to your app to use the Google Analytics for iOS Developer Kit. To track installs in Google Analytics, your app must have access to the iOS Identifier for Advertising (IDFA). To do this, you need to add the libAdIdAccess.a library to your XCode project. For details and examples, see the SDK Developer's Guide and README file.

Step 3: Set up custom campaigns

You need to set up specific campaigns for each platform you use. To make this task easier, we created a URL Builder. For more information about setting up custom campaigns, see the developer guides for your operating system.

For Android

For iOS

Further actions

Create and manage campaigns in the selected advertising network

Once you've set up mobile app install tracking in Google Analytics, you can begin developing your advertising strategy and ad content. How effectively you manage your campaigns will determine the number of users who download your app. You need to determine your target audience, your ad message, and the ad network you want to use.

Be sure to include a dedicated URL to the download page for each ad or campaign, and add tags to all ad URLs. This will help you figure out which ads are most effective at driving traffic to your download page.

It might be useful to read: