Teach you step by step how to judge the fraud and cheating of channels based on data (real examples)
|
Many times, if you don’t push yourself, you don’t know that you still have the ability to mess things up. I finally managed to deduct some money from the captain's budget for promotion, but ended up being cheated by the channel and given a cold shoulder by Pia. It can only be said that the means of cheating are becoming more and more sophisticated, from high click-through rates but extremely low activation rates, to high activation rates but low retention rates , and even now the retention rates can be very high. So in this case, how should we judge whether it is a high-quality channel or a cheating channel? Today I will teach you how to use data to determine whether a channel is cheating. Digging for clues under the veil of cheating layer by layer: The first step, the most basic judgment: retention rate Retention rate: It is basically the standard by which every company judges the quality of its channels. Under normal circumstances, it is generally divided into indicators such as next-day retention rate, 2-day retention rate, 3-day retention rate, weekly retention rate, 14-day retention rate, and monthly retention rate. The retention curve of real users is a smooth exponential decay curve, with basically no abnormal fluctuations of sudden rise and fall and no obvious faults. Once such fluctuation occurs, it means that the channel has intervened in the data (proportional machine brushing, point wall , etc.). Here are some pitfalls that my product [Youwo] encountered during the promotion process (as shown in the picture). The quality of such users is very poor and they have no commercial value. The second step is to combine other data to make judgments For example, the above picture of machine brushing behavior. Now machine brushing behavior has become more and more intelligent, and it can even simulate a perfect retention curve for you. How do we judge and identify it at this time? Below we combine several aspects to conduct a deeper level of screening: 1. Make judgments based on conversion rates Under normal circumstances, the next-day retention rate is lower than the registration conversion rate. Put yourself in their shoes. If normal users are unwilling to register, will they still be willing to open the APP on the second day? 2. Combined with average startup data Under normal circumstances, the number of times users open the APP every day also shows an exponential downward trend. When most of the users of a channel only open the APP 1-2 times a day, then there must be something wrong with this channel. It is recommended to stop the delivery immediately and observe the subsequent user behavior. You know the reason~duang~ 3. Make judgments based on the user’s behavior within the app When encountering some more unscrupulous companies that fake and inflate traffic, the above data indicators are difficult to distinguish between true and false. To deal with them, it is necessary to track and monitor the subsequent performance of users from different channels within the app, such as basic performance such as visited pages, usage time, visit intervals, interaction frequency, and compare them with overall user data. For example, in Youwo, we monitor the number of orders placed and received by each user within the app, the number of clicks on each page, the length of stay, the number of people participating in the interaction, etc. We use this data to determine whether the channel is fraudulent or of high quality. Generally, app-boosting robots can simulate seemingly real user behavior, but it is difficult to make it completely consistent with the daily data of your app (I won’t show an example picture here, the captain is already standing behind you with a knife). Third, observe long-term performance Short-term data is easy to falsify, so we can use long-term data, such as monthly retention, to measure how many users stay within a month and how they perform within the app. By making a comprehensive comparison, we can still rationally judge the quality of the channel. There are also some average level indicators in each industry, and you can also refer to these indicators to judge the quality of the channel. In short, it is not recommended to use CPA promotion method when your own data monitoring is not perfect. Cheating and anti-cheating, this is a game where the devil is always one step ahead of the good, and it must be constantly optimized and keep pace with the times. At the same time, for new feasible channels, it is best for marketing personnel to first conduct a small trial investment, observe the results after the investment, select the appropriate channels, and increase the benefits of the channel investment. APP Top Promotion (www.opp2.com) is the top mobile APP promotion platform in China, focusing on mobile APP promotion operation methods, experience and skills, channel ASO optimization ranking, and sharing APP marketing information. Welcome to follow the official WeChat public account: appganhuo [Scan the top APP promotion WeChat QR code to get more dry goods and explosive materials] |
<<: 5 ways to screen high-quality channels in App promotion
Recommend
Watsons crosses borders to attract fans, teach you how to play fan marketing
In brand marketing , fans are not only high-quali...
Tips for promoting and attracting traffic on Weibo!
Weibo can be said to be a big brother-level platf...
Hema Fresh Product Analysis
Hema Fresh is an important part of Alibaba's ...
Xiaomu C4D Product Rendering 2021 Redshift Course [HD Quality with Materials]
Xiaomu C4D Product Rendering 2021 Redshift Course...
Kuaishou information flow advertising marketing method!
A detailed introduction to Kuaishou's informa...
Hangzhou SEO Training: What should you pay attention to when positioning your online marketing promotion?
With the continuous development of the Internet e...
Monthly turnover of 60 million was selected into Tencent's social advertising game white paper, reviewing the advertising case of "Huluwa"
The buying volume market has exploded in the past...
Core strategies for new media operations
Many students working in new media often have the...
up to date! Traffic rankings of 45 information flow platforms across the entire network!
The latest traffic rankings of major information ...
The iOS promotion and operation master gave me a sudden enlightenment: how to tap into iOS channels, etc.
Question 1: I don’t know where to tap into the iO...
Count down what you need to do to promote your app! Understand channels, make plans, have connections, analyze competitive products...
1. Learn to analyze competitive products. Competi...
Form costs reduced by 30%, home decoration advertising case!
With the further deepening of regulation and cont...
How to plan a successful and beautiful event?
As an operator, event promotion (event operation)...
Recommended places for drinking tea with studios in Nanjing, which you must not miss when traveling on business
Recommendations for places to taste tea in Jiangn...
A practical guide to high conversion rates with KOL campaigns!
In the era of traffic anxiety, companies are incr...