How to choose H5 multi-channel App promotion?

Nowadays, there are more and more channels for App promotion. In addition to Wap sites and third-party apps, HTML5 has become the most important way of channel promotion, and channel statistics are an indispensable part of App operation and promotion.

Perhaps you have all encountered this problem: the App has been developed, the market has been researched, the channels have been established, and the advertisements have been placed, but the quality of users brought in is very poor. At this point we will all start to think back and ask ourselves, is there something wrong with the product? Or is it because the promotion channel was not chosen well? Or is the copy not attractive enough? Maybe all of these are correct, but we need someone to answer, and that is - channel statistics. The purpose of channel statistics is not only for settlement, but more importantly to improve App promotion efficiency and operational conversion level.

Channels (which can be advertisements, QR codes, links, etc.) are the last link facing users, and also the most important one. If you cannot trace the effect of the delivery in real time, then no matter how much effort you have made before, it is possible that it will be completely wasted in this link without you realizing it.

1. Measure the value of traffic and avoid extensive promotion

For App promotion, each App first needs to gain and buy traffic through various channels, and then convert it. Among them, the link of acquiring traffic is the most important. Don’t forget that the distribution channels and middlemen who provide you with traffic are not responsible for the quality of the traffic. If you unfortunately obtain a large amount of low-quality traffic, then no matter how hard you operate, you will not be able to convert and retain users.

If you want to avoid this kind of extensive promotion, you must first answer an important question: where does the traffic come from?

First of all, it is clear that the substantive questions that cannot be answered by extensive promotion are:

• The feedback we get from the advertising platform is that the user clicked on the ad, but did the user install the App after clicking it?
• We have promoted through many channels at the same time, but how many users have come from each channel? What motivated them to download? Which one should I invest in next time?
• How many conversions did new users contribute after installing through different channels?

Even if you don’t ask yourself these questions now, your boss will ask you at some point in time. When promoting an App, we should not only focus on quantity but also quality. We should not only promote but also evaluate. Money must be spent wisely and with good reasons so that product promotion can be targeted.

To understand the answers to these questions, we need to explain them in combination with actual promotion scenarios. You should know that channel statistics are quite complicated. For example, an App company has hired 30 sales representatives, each of whom is holding a download QR code to promote the app, and they have brought in a total of 10,000 downloads. Then, it is necessary to know how many installed users each person has brought in order to make an assessment and settlement.

Traditional field marketing solutions require each field marketing staff to allocate a field marketing code/registration code and ask users to fill it out, or manually register the user's mobile phone number and report it for statistics. However, this process often results in inefficiency, manual statistical errors, and false reporting. Similarly, inviting friends to install and placing advertisements will face similar problems. This is the traditional extensive statistics, false data, leaks, information asymmetry, opacity, and human errors, any one of which will bring huge losses to the development of promotion business.

However, if the App uses some platform-based programmatic channel statistics services, we can reasonably obtain objective channel effect data. For example, tell the field sales manager that there are 30 field salesmen distributed in different regions, how many installed users each of them brought, as well as the subsequent payment status and activity status of these users. Through the channel statistics plan, the entire promotion and operation process can be connected, making more scientific judgments and retrospection possible.

At the same time, we now have a basis for independent judgment and no longer have to listen to the opinions of advertising platforms and channel dealers. For example, if we place advertisements on TikTok and attract a large number of new users, but their willingness to pay and activity levels are very low, we can conclude that TikTok is not our target channel. The number of users is large, but the quality is not good. We can choose to reduce or abandon the placement next time. Traceable channel statistics can provide a strong basis for scientific optimization of App launch and operation.

2. What are the current best channel statistics for App promotion?

Given the complexity of App promotion channels, it is difficult to integrate multiple channels in a detailed manner, and there are not many feasible statistical solutions currently. I will talk about two common solutions:

1. Analysis of App Store Connect sources provided by Apple to developers

Log in to the Developer Connect Center - App Analysis - Source Analysis to obtain the download source. The main statistical dimensions are as follows:

• Contains the overview, quantity, proportion and trend of the four major sections: "App Store browsing", "App Store search", "App referral source" and "Web referral source". The referral source can be used to obtain data about the users who were redirected to the App Store from the App.
• Marketing Campaigns link statistics. You can customize parameters on the download link. When the user clicks the link to enter the App Store download page, this data will be collected and counted.

But unfortunately, this API is also extensive. Due to official restrictions from Apple, data obtained through this method is usually delayed by more than one day and requires at least 5 channels to be installed before it will be displayed. More importantly, it can only trace download statistics, but it cannot obtain conversion data such as registration, retention, and purchase after users open the App. Simply put, it only does the first half of the work and is not suitable for channel quality assessment under large-scale promotion.

2. Professional third-party channel statistics tools

Third-party channel statistics tools mainly obtain data through SDK methods. The general logic is: distribute promotional pages (H5 pages) with different parameters to various delivery channels. When users open the promotional page, obtain external data such as click conversions. When users download the App and open it, collect internal data through the SDK. The final attribution match is to determine which page (channel) brought the user, his subsequent activity, payment rate, and whether there are other in-depth operations. Third-party channel statistics tools play a connecting role, connecting all channel data to obtain more accurate analytical data.

Of course, you can’t just blindly choose a third-party tool. App installation is an extremely important step. If the service provided for installation is unstable, then when there are too many channels or large-scale promotion, the following problems may occur:

• The most serious consequence is that the installation fails when downloading the App, and the user is directly lost;
• Slows down the download process. According to Google data, 53% of users will abandon a mobile site if it takes more than 3 seconds to load.
• The statistical accuracy decreases, which may result in some installations not being counted, and eventually lead to errors in the statistical results and settlement disputes in the business.
• A large amount of server resources are consumed during operation, forcing service interruption and excessively high maintenance costs.

Imagine if the installation fails because of a statistical function. That would be a catastrophic blow to the user experience. Insufficient statistical accuracy is actually equivalent to traffic loss, which will also waste the company's budget.

How to choose a third-party channel statistics tool? First of all, start with the quality. For example, first investigate which customers are served (especially large enterprise applications, many of which have tens of millions of installations), how many installations have been served, and then whether the SDK documentation and API design are complete, and whether more value-added services can be provided. These are all key factors in judging whether the service is stable and accurate.

Take openinstall as an example. The SDK focuses on installation without filling out an invitation code and core services such as channel statistics. Since it started in this field relatively early, it has accumulated some typical customers and its product line has also expanded to businesses such as private deployment and advertising platform monitoring. You can use it as a criterion when selecting channel statistics services.

3. Extension: Channel parameters can be used for scenario optimization

In addition to judging channel quality, we can also use channel statistics tools to obtain the initial motivations of new users and achieve scenario optimization. By making good use of the custom parameters carried by the H5 page and identifying the channel from which the user downloaded the application, you can restore the parameters after opening the App (for example, define the parameters as user ID, content ID, room ID), which can provide new users with a better experience or content cold start.

Let me give you a few scenarios:

• In the invitation channel, users install and open the App through the H5 page. After opening it for the first time, they do not need to fill in the invitation code. The system can automatically obtain this channel parameter and automatically distribute corresponding rewards to users and inviters. The user conversion rate is bound to be higher;
• A user downloads the App through an article introducing Hainan. After opening it for the first time, the page parameters will be obtained and the page will automatically jump to the article. At this time, Hainan hotels, related attractions, etc. will be recommended to him below. The user retention rate, activity and even purchase rate will be improved.

According to the conclusion drawn after multiple rounds of channel testing: after these improvements to the App, the conversion effect is much higher than doing nothing. In different business scenarios of each App, we can use our imagination to design different functions based on channel parameters, which will produce unexpected results.

IV. Conclusion

In general, channel statistics are the necessary foundation for App promotion. Today, when the traffic dividend has gradually disappeared, the only thing you can compete on is refined operational capabilities. If you don’t do channel tracing, you will never know where your budget is wasted, and you will never think of improving user conversions based on channel parameters.

Evaluating channel quality is essentially to improve operational conversion and revenue growth, weaken ineffective channels, strengthen effective channels, and enable operators to further achieve accurate recommendations, accurate guidance, and accurate delivery based on big data, thereby fully unleashing the value of refined operations.

Author: Big City Small Things

Source: Big City Small Things