APP channel promotion statistics: application market analysis and multi-channel statistical methods

With the rapid development of mobile Internet, the era of fragmented mobile entrances has arrived. Only by mastering the user's content channels can we have the ability of mobile entrances. If an APP wants to acquire users, it needs to rely on the power of channels. The following are some mainstream promotion channels and statistical methods.

1. What are the common APP application store channels?

Mobile app stores are divided into two major camps: iOS and Android, and have experienced different stages of development in different historical periods.

Google Android and Apple iOS, as mainstream smartphone operating systems, lead the development trend of mobile APPs. iOS pursues perfection and has built its own independent kingdom around the iPhone. With AppStore as the only official download channel, the jailbreaking channel is becoming increasingly niche. Google chose another stance: openness and open source, so a hundred flowers bloomed and various third-party Android application markets emerged around Android.

Compared with foreign App Store and Google Play, the domestic Android application market has a more competitive force.

2. Traffic sources of app stores

Mainly:

​Android phone market share: App Store first, Huawei second, OPPO third; Huawei, OPPO, etc.

1. App Bao is an application acquisition platform created by Tencent App Center specifically for smart phone users. App Bao launched the "only" search for application searches, which can effectively help users solve the problem of application downloads, download applications safely and securely, and avoid downloading counterfeit applications.
2. As a representative player in the domestic Android application market, Huawei has been constantly active in the application market in recent years, and its achievements are obvious to all. As the official application distribution platform for Huawei terminals, Huawei AppGallery claims to be the world's first application market to implement "developer real-name authentication". All developers who enter the market must undergo strict real-name authentication review to filter out third-party applications of unknown origin in the Android ecosystem and ensure the safety of user use.
3. OPPO, ranked third, is not to be outdone. OPPO App Store is available in all walks of life, and video playback apps are the most commonly installed software.

Third-party app stores:

The old brands Wandoujia and Anzhi have gradually faded out of everyone's sight. However, they remain neutral and are also constantly improving and developing themselves. In particular, Wandoujia has gradually transitioned from an early application market to a humane multi-content search.

At present, the domestic Android market is divided into the third-party market and the mobile phone manufacturer market. The third-party market mainly includes App Store, 360 Mobile Assistant, Baidu, etc., and the manufacturer market includes Huawei, Xiaomi, VIVO, OPPO, Meizu, Samsung, etc.

Among these app stores, in addition to the size of traffic and the activity of channel users, the attributes of channel users are also different because the sources of channel traffic are different.

3. What are the APP promotion channels?

1) App store promotion

App store promotion is currently one of the most important APP promotion channels and also the channel with the largest number of user downloads. The more apps are released in third-party application markets and mobile phone manufacturer markets, the better. Major stores also have free activities that can be applied for.

2) New media promotion

New media can be divided into:

• Brand promotion: When the APP is first launched, there are no users and no brand awareness, and there is no relevant introduction to the brand on the website. It is necessary to first establish relevant introduction to the brand on the website
• Question and answer promotion: mainly publish various related posts on major question and answer platforms based on product-related keywords, or answer other people's questions, answer users' questions in a timely manner and collect users' questions to optimize products and user experience in a timely manner
• Social platform promotion: At present, some APP operation and promotion personnel have begun to use this method to promote, mainly through word-of-mouth communication by users, and can also plan forms and forwarding methods to drive product exposure

3) Exchange volume

Traffic exchange can be divided into free and paid. Free means that you can exchange resources with other APPs, or you can join the traffic exchange alliance. Paid cooperation can include content cooperation, opening the APP to see pop-ups, advertising images, in-app recommendations, etc.

4) Local marketing

Field promotion is the most direct way of communication with users. It can eliminate some potential users' concerns to a certain extent and increase potential users' trust in the product. Even during the field promotion process, we can interact with users and obtain the most direct user feedback, thereby optimizing the product.

4. Screening and analysis of APP channel effect data

APP promotion must open up channels, such as advertising. Simply put, mobile Internet advertising can be divided into two types: brand advertising and performance advertising.

As the name suggests, brand advertising pursues brand exposure, and advertisers pursue the shaping of brand image, so the effect is difficult to measure. Performance advertising aims to convert investment costs, such as how many times the ad is shown, how many clicks it converts to, and how many installations/sales it ultimately brings. In other words, advertisers will calculate how many actual users or revenues are converted from each advertising fee. This is the legendary ROI.

When the economy is in a downturn, advertisers will naturally cut back on brand advertising, whose effectiveness is difficult to measure, and turn to performance advertising with clear ROI indicators and data. And this is what has been happening in the vast Chinese advertising industry in recent years.

The ROI indicator for measuring promotion effectiveness is derived through data statistics. Therefore, with the rapid development of promotion forms such as performance advertising, the promotion effect is guaranteed, and multi-channel promotion has become the norm. The adoption of a set of accurate and effective statistical plans has become a very critical link.

Most operations and promotion personnel will use a multi- channel promotion approach to quantify the promotion effect, but the quality of each channel is uneven, so we need to evaluate each channel. We can select suitable and efficient promotion channels for products according to different needs. We can select channels by comparing various data indicators.

The following user data needs to be obtained:

• Download volume: refers to the number of users who download and install through a specified channel.
• Registration volume: refers to the number of users who have registered among those who downloaded and installed the app.
• Active users: Active users refer to the number of users who have launched the app within a certain statistical period. Active users are an important data used to measure the scale of app users and product status.
• User retention rate: refers to the ratio of new users who remain after a certain period of time.
• Number of new users (activated users): New users refer to users who launch the app for the first time after installing it, that is, activated users. It is meaningless for users to download the app but not use it. Therefore, when evaluating the user quality of a channel, we should not only focus on the number of users' downloads.

Where do these data indicators come from and how do multiple channels conduct effect statistics?

Multi-channel promotion is a promotion strategy commonly used by marketers, which can achieve high frequency and comprehensive growth in a short period of time. For example, the most common forms of promoting apps are used: channel links, QR codes, filling in channel codes, etc.

The most direct method is to use a third-party channel statistics tool. There are many statistics tools for APP applications. By integrating the SDK method, you can directly access the third-party statistics tool, so that you can quickly obtain channel data.

A good channel statistics tool must ensure stable services, comprehensive and complete data indicators and dimensions, and accurate data. Introducing several commonly used statistical tools in China: Umeng and talkingdata.

• Umeng: It is the earliest platform in China to enter the field of statistical analysis. It has a rich product line and performs relatively well in data stability and functional completeness.
• TalkingData: It started out mainly as a game statistics analyst, and in the field of Internet big data, its products tend to be digital services.

How do third-party statistical tools track multi-channel data sources?

During this process, channel monitoring and performance management are carried out, that is, the promotion effect of each advertisement, each invited user, each field sales staff, each platform, and each region. As we all know, iOS usually only has one download address, the App Store, while Android is more complicated, with multiple app stores and direct installation via APK. So, facing the traffic pouring in from all directions, how can we identify the sources one by one?

This involves the tracking and attribution of App channel sources, in order to determine which promotion channel each new user is attributed to. At the same time, it is also necessary to know the subsequent user behavior in order to obtain a complete ROI model.

How to count channel sources?

Based on different promotion scenarios, the traditional practices adopted will be different.

Android generally uses channel packages for statistics. After the user installs the package of this channel and activates it, the package is registered in a certain channel number for monitoring. Users from different channels are naturally distinguished.

It is still difficult to truly detect the specific quantity on iOS. Generally, each company generates an iTunes marketing link that links to the App Store download statistics.

If you need to collect statistics for tens of thousands of channels (such as user invitations, offline promotions, etc.), then channel packages, iTunes links, etc. are obviously not flexible enough, after all, they all need to be generated manually. Here, users are required to fill in the channel code (invitation code/local promotion code/activation code) to realize the binding of superior-subordinate relationships (such as invitation relationship, local promotion object, etc.).

However, the issues involved are very complicated, mainly including the following points:

• When faced with too many promotion channels, Android needs to package each channel once, which results in high labor costs and low work efficiency.
• iOS iTunes statistics also have limitations, such as data has a certain delay, channel links need to be manually created, and statistical data will not be displayed if the number of installations is less than 5.
• Since Android and iOS use different statistical channels, their data will inevitably be inconsistent, which will cause a certain degree of data trouble for market operations.
• Filling in the channel identification code is a cumbersome operation, and it also causes considerable damage to the user experience. Missing, filling in incorrectly, or abandoning the code may lead to interruptions in the invitation process and errors in local sales performance.

To avoid these situations and perfectly calculate ROI, we need to use third-party channel statistics tools.

There are many third-party platforms for statistical data. Most of them in China use Umeng, which was also acquired by Alibaba. It is currently the most widely used statistical platform for domestic APPs. There is also Google Analytics from abroad. These two are currently the mainstream statistics, and the data statistics, event analysis, retention rate, etc. of the APP data tools are very comprehensive.

However, there is a lack of APP installation statistical channel data.

Channel statistics:

There are not many third-party platforms for statistical channels on the market. The main reason is that the statistical accuracy has not reached 100%.

• H5 cannot obtain IDFA, and IDFA is disabled by default after iOS 14, which leads to a sharp drop in accuracy.
• The method of injecting cookies depends on SFSafriaViewController, which is not very versatile.
• By judging by the device feature set, the split ratio, the phone model, the current IP, etc., the error rate can be reduced to a certain extent, but it is disastrous for the same IP as the company network.
• Through Apple's official Adnetwork ID, but only supports app-to-app jump, and only supports attribution, and cannot pass parameters. The attribution number is limited to 100

Small companies do not want to spend too much time on this technology stack, so they mainly look for companies like Umeng that already have a certain amount of technology accumulation online. After comparing with several existing companies, most of them have been achieved, but the accuracy rate is a little bit biased.

Taking the domestic third-party channel statistics Xinstall as an example, the advantage of this is that it provides relevant APIs, and you can directly call the APIs to obtain relevant data statistics to your own background.

First of all, we need to clarify the needs. We mainly need to know the effects of two aspects: one is the source channel of promotion, not just the delivery platform, but also the specific number of users that each QR code, each promoter, and each advertising display can bring; the second is to know the operating behavior of each new user after installing the App, as well as the user portrait. At this time, more professional and flexible third-party channel statistics tools are needed to meet these two requirements.

Based on the channel link format, a channel link/QR code can be generated for each channel for statistics. Each user can even be treated as an independent channel. There is no limit on the number of channels, and any amount of channel data can be displayed.

It is compatible with both Android and iOS systems, and both can be promoted directly in the form of links/QR codes, so data consistency is guaranteed and there is no need to manually merge download addresses.

This method can directly eliminate manual operations such as packaging channel packages and filling in channel identification codes. User invitation relationships are automatically bound, and data statistics are reported in real time.

Subsequent user installation volume, activation volume, retention, activity, and even payment, check-in, upgrade and other behavioral data can be obtained through the docking API.

From the perspective of channel management, a good statistical method can be used to judge channel quality, rationally optimize channel delivery strategies, save promotion costs, and improve channel conversion. In terms of user growth, it can be used to judge the quality and effectiveness of a promotion channel, thereby optimizing the delivery strategy and increasing user growth.

If you don’t do data analysis, you won’t know at which stage the advertising money is wasted and which channel brings the best data results. You can focus on tracking and analyzing and increasing investment in effective channels. Today, as big data technology is gradually improving, there is a soil for channel analysis. APP promotion can play a greater value and is crucial to improving ROI.

Author: master

Source: master