Solve the three major problems of App channel tracking and increase the efficiency of attracting new users by 200%

Everyone is promoting apps, so why is your average order value so high?

Why do others spend the same 100 yuan and get the same effect as 500 yuan?

What methods do you have to increase the number and quality of new app users by 200%?

······

This article starts with the three major problems of App channel tracking, and introduces in detail 5 methods of channel tracking and two major ideas of channel data analysis . I believe there is always one method that you will find useful.

1. Common App Channel Tracking Methods

Because app stores do not provide channel source data, monitoring the effectiveness of app promotions has always been a difficult problem. In order to track the source of App channels, people have come up with various methods. Below we will give you a systematic introduction.

1. Android channel tracking method

As we all know, Google Play cannot be used in China, so the domestic Android market is occupied by dozens of app stores ( Wandoujia , Baidu Assistant, Cool Market, 360 Mobile Assistant , etc.). Android channel tracking mainly revolves around the above channels.

Method 1: Package each channel

Specifically, the developer generates a channel installation package for each channel, and different channel packages are identified by different Channel IDs (channel identifiers); after the user downloads the App, the operator can view the data of each channel through the channel ID.

Android channel packaging mechanism

Although this can count the source data from different channels, when the number of channels increases, or the same channel is promoted on multiple platforms, the channel packaging approach will become inadequate.

Method 2: Use data provided by the platform

Some third-party promotion platforms provide channel data, but it is difficult to find truly high-quality channels by relying solely on the platform's "one-sided statement".

2. iOS channel tracking method

Unlike the open ecosystem of Android, iOS is a completely closed system; except for a few jailbroken machines, most apps are downloaded from the App Store. With Apple's dominance and strict review system, Android packaging is simply not feasible here.

In order to track iOS channel data, developers have come up with a lot of black technologies. Below I will introduce three common methods.

Method 1: Tracking channels through IDFA

The full name of IDFA is Identifier for Advertisers, which means advertising identifier. This is an identifier specially set by Apple for advertising providers to track users.

Tracking channels through IDFA

As an advertising provider , Toutiao can obtain the user's IDFA. When the App you place on it is downloaded and activated by the user, your App can also obtain the user's IDFA. By matching the IDFA provided by the advertising provider with the IDFA you obtain, you can track the source of the channel.

The disadvantage is that IDFA can only be used in App-type channels. It is not supported if you are placing advertisements on a web page. At the same time, users can choose to turn off IDFA access permissions in iPhone settings.

Method 2: Tracking channels through cookies

The SFSafariViewController class was introduced in iOS 9. On the one hand, it provides a better user experience, and on the other hand, it can share cookies with Safari across apps.

Tracking channels via cookies

When a user clicks on an ad link, the monitoring server can receive the channel information contained in the cookie; when the user downloads and activates the App in the App Store, the monitoring server receives the cookie information again. The system matches the two cookies before and after to track the channel.

The disadvantage is that tracking based on SFsafariviewcontroller is only effective in iOS 9 and above, and WeChat public account ads and Moments ads still cannot be tracked.

The above method can achieve tracking and monitoring of some platforms and some channels, but the three major shortcomings are also obvious:

(1) The channel data of Android and iOS platforms are separated, making it difficult to integrate and analyze;

(2) Android delivery requires repeated packaging, which is inefficient;

(3) The iOS channel has many limitations and cannot be promoted on a large scale.

2. Solutions based on user device tags

Below we introduce a fast and flexible solution – a tracking method based on user device identification, which is compatible with both Android and iOS platforms and applicable to most delivery channels.

1. Tracking principle based on user device tags

The two methods based on IDFA and SFSafariViewController described above are both restricted by iOS, while the user's device tag is not affected by the system. In the GrowingIO [Channel Source] solution, we set the "IP+UserAgent+Device ID" combination as the user's device tag.

Tracking channels by user device tagging

When a user clicks an ad link containing UTM tracking parameters, the GrowingIO server detects the user’s device tag and UTM channel parameters. After the link jumps to the app store (both Android and iOS), the user downloads, installs and activates the App. At this time, the GrowingIO server receives the user's device token for the second time.

The system matches the two tags before and after to determine the user's channel source, and at the same time presents the detailed channel information contained in the UTM parameters.

2. Characteristics of user equipment marking methods

Of course, the method based on user device tagging also has certain shortcomings. When the network environment of a small number of users changes (such as switching from WiFi to 4G), the inconsistency of IP will cause matching failure.

However, compared with the previous four methods, the channel tracking method based on user device tags is obviously more advantageous:

First, by connecting the channels of iOS and Android, the [operating system] can be added to the user attribute integrated analysis;

Second point: It avoids the duplication of channel packaging on the Android platform;

Third point: It avoids many original restrictions of iOS and is applicable to a wider range of promotion channels;

Fourth point: You only need to modify the parameters in the promotion link without changing the installation package. It is suitable for large-scale, multi-channel, and agile promotion needs.

What does the UTM parameter mean?

At the same time, the channel parameters contained in the advertising link (ad source, advertising media, advertising name, advertising content, advertising keywords) can be added to the user attribute data to facilitate multi-dimensional comparison and cross-analysis of user data in the future.

3. Two major approaches to App channel data analysis

With the App channel tracking data, we can use the five parameters of UTM as dimensions to analyze the App channel data from the perspectives of quantity and quality.

1. Quantity: Find the channel with the lowest customer acquisition cost

According to business needs, we selected two dimensions: advertising source (utm_source) and advertising keyword (utm_term), calculated the number of customers acquired through different channels and evaluated the customer acquisition cost.

Evaluate customer acquisition costs

An O2O app was launched twice on three channels, with the contents being "food" and "takeout" respectively. Through UTM, we monitor the number of new App users for each channel and each launch, and then calculate the average customer acquisition cost.

From the perspective of advertising sources, Channel 1 has the lowest average customer acquisition cost; from the perspective of advertising keywords, ads with the theme of "takeout" have the lowest average customer acquisition cost. From the perspective of average order value, we can then optimize the delivery channels and content in a targeted manner to significantly reduce delivery costs and improve the efficiency of attracting new customers.

2. Quality: Find the channel with the highest customer acquisition value

The two indicators "number of newly activated App users" and "customer acquisition cost" are analyzed from a quantitative perspective, but large quantities and low prices do not necessarily mean high quality of channel users. We also need to comprehensively consider the performance of new users in the future and the value they can bring.

Method 1: User behavior data analysis

In this process, we focus on user retention indicators, including next-day retention rate , three-day retention rate, seven-day retention rate, thirty-day retention rate, etc.

New user retention curves from different channels

We analyzed the retention of new users by visit source (utm_source) and found that the 30-day retention rate of Channel 2 was as high as 14%, while Channel 1 was 8% and Channel 3 was 6%. From the perspective of retention, the value of new users acquired through Channel 2 is significantly higher.

Method 2: User Value Analysis

In addition to user behavior indicators, financial indicators are also very useful for reference. According to the advertising source (utm_source), we calculate the financial value of new users acquired through different channels, such as the monthly paying rate (MPR) and average revenue per user ( AR PU) of new users in the first month.

The value of new users from different channels

Through analysis, it was found that the first-month payment rate (42%) of new users acquired by Channel 2 was the highest, and the average user revenue (30 yuan) was also the highest. Although the customer acquisition cost of Channel 2 is slightly higher than that of Channel 1, from the perspective of returns, investing in Channel 2 is obviously a wiser choice.

Based on the above indicators, this O2O App will focus its resources on Channel 2 in the market launch next month, while focusing on the theme of "takeout". It is still the same marketing budget as last month, but the number of new users has increased by 150% and the retention rate of new users has increased by 240%, which is a huge growth.

The traffic dividend of mobile Internet is gradually fading, and millions of apps are competing for users in a stock market; whoever can improve the efficiency of customer acquisition will have the chance to win in the fierce competition. Building a complete App channel tracking system and conducting lean analysis on the quantity and quality of each channel and each delivery is a way to get twice the result with half the effort. Don’t you want to try it?

Mobile application product promotion service: APP promotion service Qinggua Media advertising

This article was compiled and published by @GrowingIO Growth Team (Qinggua Media). Please indicate the author information and source when reprinting!