Spending Is Technology! Teach You To Pay For Channels

First,

To do master level user data analysis, you need to have detailed records of users' data before promotion.

This is the least bit lazy.

How to record data? The key is to distinguish, track and record.

What are the tools?

Tool 1:

Android platform and jailbreak channel are applicable.

According to the promotion channels, different channels are identified (Qu Daobao), and each channel package tracks downloading, activation, pformation, deep participation (purchase, recharge, etc.).

Tool two: links

IOS platform can not be differentiated according to the channel.

Then set up different links. If APP itself embeds Talkingdata or SDK, then you will have mysterious weapons. If you join Track or TD Ad Tracking, you can track the click, activation and retention of each channel.

Note: APP is required to be embedded in SDK between TD and EU.

SDK of TD is said to be more accurate.

Tool three: two dimensional code

Offline advertising placed two-dimensional code, two-dimensional code link do not be silly to take the App Store link directly, remember to turn a short chain.

If you do not generate short chain background partners, Mai Jie recommends two tools: Baidu statistics.

You can see the PV/UV of each short chain every day.

Tool four: the last artifact.

At this time, there should be a small partner to raise their hands. I think the top is not easy to use. What should I do?

The magic weapon of Mai Jie's bottom of the box is coming - the code number, customer service QQ, WeChat public number, customer service telephone.

If you are promoting online and offline, you can consider issuing different discount codes from different channels.

Like Uber, it is very smart. The discount code is divided according to every city, every quarter and every event - uberyouthbj, 54 Youth Festival Beijing exclusive discount code.

By the same token, you can set up customer service QQ, telephone, micro signal, etc., to set up a variety of platforms to communicate with users.

In addition to using the data that can be obtained by itself, direct communication with users can get more user behavior information, such as users' habits, preferences, cognition of products, and understanding of activities. It is very constructive as a reference.

Two. After the promotion.

After the promotion period, the most important thing is

Analyze the hidden truth behind the data, which channel is a slag and which channel is too valuable?

The yardstick of this value is ROI (return on investment), the rate of return on investment.

I is easy to understand, cost, how much money did you spend on this round of promotion?

R is different.

For game APP, it can be understood as the conversion rate of recharge. Besides the regular activities such as user activity, long usage time and participation activities, repeat purchase behavior also needs special records. This is of great reference value for analyzing the purchasing power of users and pricing products.

For APP, it can be understood as buying conversion rate.

For example, user consumption frequency, customer price and repeat purchase rate;

For service class APP, it can be understood as service utilization.

For example, users hook up several younger sister paper and pass several photos.

Three, before the promotion of the next round.

Through this launch, we not only know the traffic introduced by various channels, but also get more detailed information about traffic pformation, pformation of users' activity, behavior, purchasing power and so on, and we can achieve the full cycle ROI analysis, which creates very favorable conditions for our precision marketing.

First of all, we need to comprehensively analyze the contribution value of the channel and adjust the input ratio of each channel.

It is far from enough to consider the flow of light. Conversion rate, purchasing power, activity and so on should be included in order to fully judge the channel value.

Secondly, according to the continuous collection of user information, more in-depth mining of user characteristics, and then mining more corresponding channels, supplemented to consider the launch list, so as to expand the larger scope of the launch.

Finally, if we can optimize the input ratio of each channel, we may gain higher profits.