A comprehensive guide to App promotion and operation knowledge: [Data Analysis]

Part 1 How to evaluate channel quality from data Part 2 Introduction to domestic and foreign data analysis tools Part 3 How to identify channel fraud from data Part 1: How to evaluate channel quality from data There are too many channels and they are mixed. It is very difficult to choose suitable high-quality channels. The case where a company spends tens of thousands of yuan on promotion but obtains tens of thousands of activations and has only a few real users is by no means an isolated case. An excellent promotion channel must not only have a stable number of new and active users, but also reflect the real user behavior data in the data so that we can continuously optimize the product. Here I will share with you some methods to evaluate channel quality through statistical analysis tools. First, let’s clarify the definitions of several indicators. 1. Add new users, also known as activated users, that is, users download the application and start it. 2. Active users refer to the number of users who have opened the app at least once during the selected time period. 3. Composition of active users, which can clearly reflect the proportion of new and old users among active users. 4. Usage duration, that is, the time distribution of a user launching an application at a time. 5. Retention rate: Users who start using an app within a certain period of time and continue to use the app after a period of time are considered to be retained; the proportion of these users to the new users at that time is the retention rate. For example: On January 1, channel A added 1,000 new users. On January 2, 500 of these 1,000 users launched the app again, and on January 3, 40 of them used it again. Then the next-day retention rate of the new users added to channel A on January 1 was 50%, and the retention rate after two days was 55%. 6. Custom events can record an action or operation of an end user. For example, user downloading, sharing, payment and other behaviors can all be monitored through custom events. Different user behaviors may have different indicators that need to be paid attention to, which can be divided into numerical events and counting events. Some other common indicators such as device model, network type, geographical distribution and other information also serve as a reference when judging channel quality. Let’s talk about how to use these indicators to evaluate channel user quality. Retention Analysis First of all, it should be said that nowadays, whether it is next-day retention, seven-day retention, or 30-day retention, they can all be faked, and can even be calculated to be very natural. Therefore, it is meaningless to look at retention alone. It must be combined with other indicators such as usage duration, custom data, etc. Assuming that the data is not adulterated, developers can evaluate the quality of channel users by comparing user retention rates of different channels after a fixed length of time. For example, the monthly retention rate of new users of channel A after three months remains at around 30%, while the monthly retention rate of new users of channel B after three months is about 20%. The number of new users in the two channels is similar, but the user quality of channel A is better than that of channel B. Duration of use There must be a difference in the usage time of data generated by real user behavior and fake data. Operations staff usually have an overall grasp of the distribution of application usage time. If the data shows that the usage time of a certain channel is significantly different from that of other channels, the authenticity of the data of this channel is questionable. Terminal properties The terminal attributes I am talking about here include device model, network and operator, and geographical distribution. There was a case where an operations colleague of a well-known company had always suspected that the channel dealers were falsifying data, but had been struggling to find clear evidence to prove it. After comparing multiple data indicators, he finally found clues in the terminal attribute data: the models of normal channels were mostly Samsung, Xiaomi and other devices, while more than 90% of the models of this channel were cheap Android phones with a market price of around 400 yuan, and some of the models were actually Android development boards! In short, making good use of statistical analysis tools can help us identify general data falsification and evaluate channel quality. On the other hand, data fraud is becoming more and more professional and industrialized. How to fight against cheating masters and find clues of fraud behind seemingly beautiful promotional data will be further discussed in the third part of this article [How to identify channel cheating from data]. Part 2: Data Analysis Tools  16 local data analysis tools  30 foreign data analysis tools 16 domestic data analysis tools: Most domestic data analysis tools are a combination of a small number of free functions and advanced paid function services. It is estimated that the early free ones will gradually shift to a paid model. 1. Umeng http://www.umeng.com/ It supports statistical analysis of iOS and Android application data, and basically covers the data analysis functions required by APP. On January 26, 2016, Umeng, global Chinese website statistics and analysis platform "CNZZ", and domestic Internet data service platform Diyuanxin Network Data, three companies jointly invested by Alibaba, announced their merger into "Umeng+". "Umeng+" can reach more than 900 million independent active Internet users worldwide every day. 2. Cobub Razor http://www.cobub.com/ is the open source version of Umeng. Supports iOS and Android application data statistical analysis. It mainly provides multiple data statistics such as channels, versions, usage frequency, usage time, page access paths, user retention, terminals and networks, operator distribution, events and conversion rates, error analysis and automatic updates, and provides viewing by region and by time period. 3. Qingyuan Huoyan (huoyanapp) http://www.huoyanapp.com/ You can view the daily, weekly, and monthly activity of any APP user. In addition, application comparison can be performed. Its functions are mainly divided into three parts: channel monitoring, user analysis, and crowd analysis. User analysis mainly includes new users, active users, time period analysis, and usage frequency.　　 4. AYL Aiyingli App ranking monitoring http://rank.aiyingli.com/ Monitor your app's ranking and download volume changes in major domestic app markets, covering both iOS and Android. Historically, there have been many tools for iOS but few for Android. Previously, statistics on Android download data, search rankings, overall rankings, and category rankings all required manual methods. This method is much better. It is said that many product transfer transactions, financing, and CPT statistics will refer to their data. 5. Application Radar http://www.ann9.com/ For iOS only, view the App Store overall and category rankings. Check the product’s search score in the App Store, which is one of the criteria for judging ASO effectiveness. 6. Zhuge iO https://zhugeio.com/ features insight analysis of user behavior through attribute filtering-selecting conditions-selecting time filtering bars, more intuitive user portraits based on each step of the user's operation, and different messages can be pushed to different users! And after the push, there is also effect monitoring. 7. TalkingData http://www.talkingdata.com/ The statistical categories include users and usage (new, active, regional distribution, device models, etc.), channel statistics, events and conversions, and analysis tools. We can use channel data to view the source of users and evaluate the effectiveness of promotion. 8. Baidu Mobile Statistics http://mtj.baidu.com Support iOS and Android platforms. In addition, after embedding the statistics SDK, developers can conduct more comprehensive monitoring of their own products, including user behavior, user attributes, geographical distribution, terminal analysis, etc. There are too many Android channels, and it is impossible to maintain every channel in an all-round way, so we have to rely on analysis. We can monitor the promotion effects of different channels through Baidu Mobile Statistics, so as to eliminate inferior channels.　　 9. ASOU http://www.asou.com/ Through popular searches, we can analyze the user's search behavior in a certain period of time. However, considering that brushes are very powerful nowadays, this piece of data is no longer of much reference value. We can focus on checking whether there are behavioral words in it. In addition, developers with iPad business can also use asou tools for analysis.　　 10. Vtool ASO http://vtool.cc/ It can help developers to conduct ASO keyword sharing, keyword restoration, keyword expansion, keyword 24-hour ranking, download volume estimation and other keyword analysis and statistics.　　 11. APPDUU http://www.appduu.com/ Only supports iOS. You can view APP weight, ASO keyword coverage, etc. However, this is a paid APP statistics tool, and there are many restrictions for ordinary users. You can see more comprehensive data after upgrading to VIP. Another highlight of keyword analysis is the word segmentation tool.　　 12. ASO100 http://aso100.com/ It supports querying the top 1,500 apps in Apple's overall and sub-lists, and developers can check their app rankings at any time. You can search the keyword index ranking and query the ranking of each keyword in each category list. It can also perform competitive product analysis. Developers can learn from the keywords used by competitors to fill in the gaps for their own apps. 13. Coolchuan http://www.coolchuan.com/ Only supports Android platform application monitoring. Developers can view data such as app downloads, rankings, ratings and comments, keyword rankings, etc. in mainstream markets, and can also systematically compare data with similar competing products. 14. Play Data http://www.playdata.cn Supports statistical analysis of application data on mainstream platforms such as iOS, Android, and WP. Functions include operational analysis, user usage, channel analysis, user terminals, events and conversions, error analysis, and ad space monitoring. 15. Haidu Cloud Analysis https://m.hiido.com/ It supports mainstream smartphone platforms, claims to be permanently free, and can help APP entrepreneurs count and analyze active devices, user sources, user attributes, channel data and retention rates, etc. 16. Chandashi http://www.chandashi.com/ supports statistical analysis of iOS and Android application data, and can track the performance of applications in various market channels. It also provides ASO keyword diagnostic tools, expansion tools and intelligent recommendation tools. 30 foreign data analysis tools User group segmentation 1. Upsight (including paid items) Upsight is an analysis tool for mobile application developers. Its features include: user segmentation, funnel analysis, retention analysis, in-app purchase components and unlimited data storage space. Upsight supports almost all mobile platforms, including iOS, Android, Java Script, Adobe Air, etc. 2. Tap stream (free) The highlight of Tap stream is the analysis of user life cycle. If you want to know where users search for information about your app every day, or how often they actually download it from a certain channel, Tap stream will become a source of information you can trust. Tap stream supports iOS, Android, Windows, and Mac apps. 3. Flurry Analytics (Free) Flurry is almost the "industry standard" for mobile app analytics. Flurry helps you track user sessions so you can see where users are having trouble using your app. You can also create custom segments to better understand your app’s user base. 4. Capptain (paid items included) Capptain is a real-time analysis tool that looks like a set of data dashboards. Not only can it track real-time user behavior, but it can also monitor user usage feedback, and even group user groups in real time, send them instant messages based on their geographic location, and so on. Capptain is available for iOS, Android, HTML 5, Blackberry, Windows and other platforms. 5. Followapps – App Refined Analysis Platform 6. MobileAppTracking – User data tracking and prediction model Touch screen hotspot analysis 7. HeatMa.ps heat map (paid) Heat map is one of the few App hot zone tracking tools. Heatmaps help app developers record all screen touches, gestures (expand/zoom/swipe), and device positioning. You can even get detailed hotspot maps of user touch screens. The only regret is that the heat map is only supported on iOS App. 8. Heat Data (paid) Heat Data is another heat zone tool for mobile apps and websites. You can track all the actions that occur when your users touch the screen: clicks, swipes, zooms, etc., and get detailed visual analysis reports. Heat data is cross-platform, all you have to do is copy one line of JS code to embed it in your App and use it. But if you don’t want to embed JS in your app, then you need to use another tool. In-app purchase tracking 9. Appsflyer (including paid items) Appsflyer is an all-in-one marketing tool with built-in analytical functions. You can track in-app purchases, software installations, and user usage in the same tool. In addition to supporting mainstream iOS, Android and Windows systems, Appsflyer also supports other platforms and engines, including: Unity, Marmalade, Appcelerator, etc. It can be said that it truly achieves full platform support. 10. Appfigures (including paid items) Appfigures can monitor in-app sales related to events while tracking them. Appfigures collects and presents app ratings, downloads, and payment amounts from different channels. Appfigures is also available for iOS, Android, and Mac platforms. They also provide an API so you can use it and get anything else you want. 11. Swrve – In-app purchase analysis platform channel tracking, advertising placement in app ratings 12. Apsalar (including paid items) Apsalar is a data analysis tool specifically designed for large application stores. In addition to basic user analysis functions, Apsalar also has a powerful advertising management component. 13. App Annie (including paid items) App Annie is a very unique analysis tool. It no longer analyzes user activities, but only tracks app downloads and sales. Whether it's iTunes, Google Play or Amazon store, you can directly understand the app's downloads, ratings, reviews and rankings through App Annie. 14. Askingpoint (including paid items) The highlight of Askingpoint is also the tracking of App ratings. In fact, its main function is to prompt more users to review your app. Although I don’t think this is the best way to improve user reviews, this tool can still help developers obtain and track reviews more easily. 15. Distimo's AppLink is a cross-platform channel distribution and conversion rate tracking tool. They also have their own App to help you monitor App operation data anytime and anywhere. 16. Trademob – Mobile Marketing Analytics 17. Adxtracking – In-app advertising operation, optimization and analysis tool Basic statistics 18. Amazon Mobile Analytics (Free) Mobile data analysis is just a part of Amazon’s huge ecosystem and is a basic cross-platform analysis tool. You can use it to track the apps you publish on iOS, Android, and of course Amazon. It has all the typical data analysis features you could think of. At the same time, it also has A/B Testing function to help operators test different operating modes on one application. 19. Roambi (paid) Roambi focuses on serving large R&D teams. This is a 3-in-1 analysis tool that integrates three major functions: basic data analysis, BI reporting for mobile applications, and program anomaly warning. Roambi also allows you to push data back into its Box component, generating reports that are easy for team members to read. 20. App celerator (including paid projects) App celerator's main business is the integrated marketing component of mobile applications, but their application analysis tools are also sufficient to stand on their own. In the App celerator tool, you can track session duration for new users and custom events. 21. Countly (including paid items) Countly is an open source mobile app analytics tool. One thing that's different from most open source projects is that Countly is actually pretty nice. With Countly you can easily see the distribution of your app on different platforms, screen sizes and devices. 22. Kontagent – Mobile application data analysis component 23. Claritics – App BI Data Analysis 24. Appsee – Visual mobile app analytics 25. Yozio – Mobile App Data Tracking 26. AppsFlyer – Mobile App Instrumentation and Data Tracking 27. Telerik – Mobile App Analytics Focus on mobile game analysis 28. Honey tracks (including paid items) The difference of Honey tracks is that it focuses on mobile application analysis of games. Honey tracks are configured to help game studios track over 90 metrics, including engagement and retention analytics for mobile game users. 29. Playtomatic (Free) Playtomatic is also an open source app analysis tool, but it focuses more on the mobile gaming field. Playtomatic helps game developers track players’ location and achievements within mobile games, supporting multiple platforms including iOS, Android, JavaScript, HTML 5, Unity 3D engine, etc. 30. Applicasa – Mobile game management platform Part 3 How to identify channel cheating from data How to judge whether the user is real, whether he is from a firewall, whether he is a machine-generated or human-generated user? At present, it is difficult to distinguish based on the user attributes tracked and counted. Even if user retention, model, region, and Internet connection are normal, it cannot be said that there is no cheating. People are too sophisticated nowadays and can do all these things. Even mobile phone number registration can be faked! The ultimate judgment method is to look at the user's contribution to the content: First of all, you have to look at the type of your application. Here are three examples. For example, if you are an e-commerce application, the most direct way is to look at the order consumption of users in this channel - the conversion rate from activation to order creation. If this fraud can generate revenue contribution for you, that's not bad, haha, so this can't be false. For example, if your application is a life-related application that can generate information, such as the Ask a Doctor series, then you should look at the number of questions asked by users of this channel. The cost of faking this is high, and it is not cost-effective for them to falsify this. For example, if your product is a game, then it is similar to e-commerce. You look at the proportion of users who purchase value-added props and the proportion of users who complete a level. They won’t hire people to cheat in playing games, right? ! From these three examples, we can see that it is much easier to look at your own product first and judge based on the details of the product content. Cheating data that is relatively easy to reveal: 1. Retention rate The channel will choose to import user data at important time points such as the next day, 7 days, and 30 days. Then we found that the data of APP at key time points such as the next day, 7th day and 30th day were significantly higher than those at other time points. The retention curve of real users is a smooth exponential decay curve. If they find that the retention curve has abnormal fluctuations with sharp rises and falls, it basically means that the channel has intervened in the data. (II) User terminal information 1. Ranking of low-priced devices: The ranking of devices of new users or startup users in the channel is analyzed based on experience. If they find that a low-priced device is ranked abnormally high, they will regard it as abnormal and start to report it. 2. The proportion of new versions of operating systems: After years of channel devastation, operators finally discovered that many channel-boosting studios have delays in adapting operating system versions. When viewing the operating systems of channel users, you can compare them with the distribution of operating systems of all mobile Internet users. 3. Usage of Wi-Fi network: for example, whether the usage ratio of 2G, 3G, and 4G is normal, etc. (III) Extended information 1. The distribution and regularity of registered nicknames. Many low-end fake registered nicknames have strong regularity. All operators must have encountered such a situation. 2. The distribution of the registered mobile phone numbers by location. I think you have all encountered this before. The mobile phone numbers of users coming from a certain channel are not only from a certain city of a certain operator, but they are even consecutive mobile phone numbers. (IV) Single indicator 1. IP: whether it is a blacklist IP or a proxy IP, compare it with a huge blacklist database; 2. IMEI: Is it a blacklist IP? 3. Mobile phone number: whether the number is illegal or on the blacklist. 5. Group indicators 1. IP: Whether the geographical distribution of user IP is consistent with the distribution of prior data, including the distribution of domestic provinces and overseas markets; 2. IMEI: Whether the geographical distribution of user IMEI numbers conforms to the distribution of prior data, and whether the distribution of manufacturers represented by IMEI is random; 3. OS: Whether the distribution of the operating system version of the channel conforms to a certain degree of randomness and statistics, and is compared with the previous prior data; 4. Model: whether the model distribution is consistent with the prior data and the proportion of the latest smartphone shipments; 5) Location information: whether the ratio of location information opening and the ratio of geographical distribution of location information obtained are consistent with the distribution of prior data, the geographical situation promised by the channel, and the actual distribution of the application; 6) Operator: whether the data distribution of the operator is random, whether it is consistent with the normal distribution of domestic operators, and the random distribution of overseas operators; 7) Network access method: Whether the distribution ratio of WiFi, 2G, 3G, and 4G maintains the same trend and data characteristics as the prior data. (VI) Information consistency Verification of device consistency, including: CPU, manufacturer, Mac address, IMEI, model, and operating system consistency verification;

<<: I am so impressed by how information flow advertising is done!

>>: After refuting the relationship rumors, Zhang Ming En appeared in Hu Bingqing's residential area and left immediately when he saw the candid photos!

Zhang Jianman's 30 Relationship Management Lessons All Smart Women Need Watch Online

A comprehensive guide to App promotion and operation knowledge: [Data Analysis]

Zhang Jianman's 30 Relationship Management Lessons All Smart Women Need Watch Online

How to join Jack Ma's facial recognition payment, and how much is the facial recognition payment agent fee?

Is it true that there will be no Qingming Festival holiday in 2022? Why not take a holiday?

How to create an operational plan for a product from scratch. The focus is on ideas, not writing plans

Case review + fission methodology | Why is your fission activity ineffective?

Five major factors and optimization methods for Toutiao information flow advertising conversion

How to plan an attractive live broadcast?

Talk about user cognitive cost!

Chen Wen's top private equity trading skills secret training camp: call auction + time-sharing decision-making

With these tools, you can easily create high-conversion landing pages even if you don’t know design or code!

Recommend

The secrets of Douyin App's operation and promotion routines

How to enter the Daily Specials Factory Store?

The golden combination strategy for live streaming sales

Getting Started Guide to Short Video Promotion!

What is news media outreach? What are the characteristics and advantages of news media marketing?

Notice! Douyin, Baidu, and Toutiao's future marketing guidelines revealed~

How to make a WeChat red envelope cover? WeChat Red Packet Cover Making Tutorial

Is there any difference between Douyin Feed and Dou+?

How to quickly attract traffic and promote? Which platform is better at attracting traffic now?

How should brands measure the effectiveness of advertising on TikTok?

Do I need a business license to register Blue V on Douyin? What is required for Blue V certification?

7 essential skills for operations - marketing promotion steps!

Service Mesh Istio Basics and Practice

How to place KOLs on Weibo and WeChat to achieve the maximum effect?

Li Xiaoyue's "Little Red Book Full-Level Short Video Training Camp" creates an exclusive IP in 30 days