Introduction to say that Quick Audience is a global consumer operation platform that integrates data asset construction, user analysis, precision marketing, cross-terminal social interaction, and global membership management. One of the big reasons is the integration of classic marketing Models, such as the RFM model and the AIPL model, support consumer operations by methodology to achieve efficient growth and new innovation.
-For more information about the transformation of digital intelligence and the content of the data center, please join the Cloud Data Center Exchange Group-Digital Intelligence Club and follow the official WeChat official account (scan the QR code at the end of the article or click here to join )
Cloud Data Center official website 16110d97f4e705 https://dp.alibaba.com/index
background
In the marketing process, we need to think about how to analyze data and how to present data, because this is a very important link to play the value of data. Through data analysis and visualization, more intuitive insights can be found, and the value behind the data can be discovered, thereby assisting business decision-making and realizing true data-enabled business.
Quick Audience is a global consumer operation platform that integrates data asset construction, user analysis, precision marketing, cross-terminal social interaction and global membership management. Quick Audience incorporates classic marketing models: RFM model and AIPL model, which supports consumer operations by methodology to achieve efficient growth and new innovation.
RFM model
Basic concepts of RFM model:
The RFM model is a means to measure customer value through three indicators: the customer's R consumption interval (Recency), F consumption frequency (Frequency), and M consumption amount (Monetary).
The RFM model quantifies and scores the values of the three indicators of the customer, and then compares the score of a single customer with the comparative value to obtain the relative value level of the customer in the group, and then integrates the three indicators to divide the customer group into multiple Types, so as to facilitate the adoption of targeted operating methods for different types of customers.
RFM user type classification rules
Compare the user's RS, FS, and MS scores with the RS comparison value, FS comparison value, and MS comparison value, respectively, to get the user's relative value level in the group:
- The user score is greater than the comparison value, and the value is higher.
- The user score is less than the comparison value, and the value is low.
description:
RS, FS, and MS are respectively the user's consumption interval, consumption frequency, and consumption amount score.
The RS comparison value, FS comparison value, and MS comparison value are respectively the average of the consumption interval, consumption frequency, and consumption amount scores of all users in the RFM model (that is, the weighted average in statistics), or a custom value.
The value of users in any of R, F, and M can be divided into high and low categories. The performance of R, F, and M can be integrated into 8 types. The detailed types and classification rules are as follows Shown.
| the RFM user type | RS | the FS | the MS | described |
| <span class = "Lake-fontSize-. 11"> high value customer </ span> | <span> greater than or equal RS | value (high)</span>greater than or equal to FS contrast value <span>(High)</span> | <span class="lake-fontsize-11">greater than or equal to MS contrast value</span>< span>(High)</span> | <span class="lake-fontsize-11">Defines people with a relatively recent consumption date, high consumption frequency, and high consumption amount as high-value people</span> |
| <span class="lake-fontsize-11">focus on keeping customers</span> | <span class="lake-fontsize-11">less than RS contrast value</span><span>(low )</span> | <span class="lake-fontsize-11">is greater than or equal to the FS | <span class="lake-fontsize-11 | value</span><span>(high)</span> 16110d<span class="lake-fontsize-11"> defines the population with a relatively long recent consumption date, but high consumption frequency and high consumption amount as the key to keep customers. </span> |
| <span class="lake-fontsize-11">Key development customers</span> | <span>is greater than or equal to RS contrast value</span><span>(high)</span> > | <span class="lake-fontsize-11">Less than FS | <span class="lake-fontsize-11"> greater than or equal to MS Comparative value</span><span>(high)</span> | <span class="lake-fontsize-11">The most recent consumption date is closer, the consumption amount is higher, but the consumption frequency is not high. Focus on developing customers. </span> |
| <span class="lake-fontsize-11">Key customer retention</span> | <span>Less than the RS comparison value</span><span>(low)</span> | <span class="lake-fontsize-11">less than FS | <span class="lake-fontsize-11"> greater than or equal to MS comparison Value</span><span>(High)</span> | <span class="lake-fontsize-11">The most recent consumption date is farther, the consumption frequency is lower, but the population with higher consumption amount is defined as Focus on retaining customers. </span> |
| <span class="lake-fontsize-11">General value customers</span> | <span> greater than or equal to RSspan> (higher than or equal to RSspan)</span> > | <span class="lake-fontsize-11">is greater than or equal to the FS | <span class="lake-fontsize-11">less than MS | value</span><span>(low)</span><span class="lake-fontsize-11">The most recent consumption date is closer, the consumption frequency is higher, but the consumption amount is not high. For general value customers. </span> |
| <span class="lake-fontsize-11">Maintain customers</span> | <span class="lake-fontsize-11">less than RS<span class="lake-fontsize-11"> span>(low)</span> | <span class="lake-fontsize-11">greater than or equal to the FS | <span class="lake -fontsize-11">Less than the MS comparison value</span><span>(low)</span> | <span class="lake-fontsize-11">The most recent consumption date is far away, and the consumption amount is not high. However, the population with a higher consumption frequency is defined as general retention customers. </span> |
| <span class="lake-fontsize-11">General development customers</span> | <span class="lake-fontsize-11">greater than or equal to RS comparison value</span> <span>(High)</span> | <span class="lake-fontsize-11">Less than the FS | <span class="lake -fontsize-11">Less than the MS comparison value</span><span>(low)</span> | <span class="lake-fontsize-11">The latest consumption date will be closer, but the consumption frequency and consumption The population with a small amount of money is defined as general development customers. </span> |
| <span class="lake-fontsize-11">Potential customers</span> | <span class="lake-fontsize-11">less than RS contrast value</span> >(Low)</span> | <span class="lake-fontsize-11">Less than the FS | <span class="lake-fontsize | value</span><span>(low)</span><span class="lake-fontsize-11">The latest consumption date is farther, the consumption frequency is not high, and the consumption amount The low population is defined as potential customers. </span> |
### RFM model application:
In Quick Audience, the RFM model can be used to analyze the core indicators of users and the proportion of users classified. According to different user types, different marketing activities are delivered.
#### RFM model transaction data analysis
core indicators
Display specific values and trend graphs of the number of transaction users, transaction amount, per capita transaction amount, and per capita transaction frequency. The effects of the RFM analysis interface in Quick Audience are as follows:
RFM user composition (user type)
According to the user classification definition of the RFM model, the distribution of the audience's user types is displayed. The effects of the RFM analysis interface in Quick Audience are as follows:
RFM user composition (consumption distribution)
According to the user's consumption interval, consumption frequency, consumption amount, display the consumption potential distribution. The effects of the RFM analysis interface in Quick Audience are as follows:
Consumption Ability Distribution (MF-R) : The abscissa is F transaction frequency, the ordinate is M transaction amount, and the point size is R last transaction interval. Through the MF distribution, we can intuitively see the distribution of customers' spending power, and then use the size of R to lock which customers are more loyal. The larger the point, the higher the customer loyalty.
Consumption Potential Distribution (MR-F) : The abscissa is the last transaction interval of R, the ordinate is the transaction amount of M, and the point size is the frequency of F transactions. Through the MR distribution, we can visually see the customer's consumption potential, and then use the size of F to find more valuable customers. The bigger the point, the more valuable the customer is.
consumption distribution (RF-M) : The abscissa is F transaction frequency, the ordinate is R last transaction interval, and the point size is M transaction amount. Through the RF distribution, we can intuitively see the customer's consumption changes, and then use the size of M to determine which customers are more necessary to recover. The larger the point, the more necessary it is for the customer to recover.
Through the above analysis, the corresponding customers can be labeled with customer characteristics, so that different marketing strategies can be specified for certain types of customers.
## Problems in the RFM model building process:
#### 1. The analysis data loading is too slow for customers with tens of millions of data
Solution: Perform pre-calculation when creating the RFM model, and use the model id as the cache key.
#### 2. The processing of rank-level authority control and analysis for different department authorities for the same RFM model
Solution: Generate SQL when creating RFM pre-calculation and row-column-level configuration, and perform md5 calculation according to SQL plus model id as the cache key
# AIPL model
## Basic concepts of AIPL model:
The AIPL model is a means of quantifying and linking brand crowd assets. in:
* A (Awareness): Brand awareness people, generally refers to people who passively come into contact with the brand, such as people who are reached by brand advertisements and searched for category words.
* I (Interest): Brand interest groups, generally refers to people who actively come into contact with the brand, such as advertising clicks, browsing the brand/store homepage, participating in brand interaction, browsing product detail pages, brand word search, receiving trial, subscribing/following/ Those who join the club and purchase additional collections.
* P (Purchase): Brand purchasers, including people who have made purchases.
* L (Loyalty): Brand loyal people, such as people who have repurchase behaviors or have positive reviews and sharing of the brand among the purchasers.
## AIPL model construction process and application:
In Quick Audience, the AIPL model divides the brand crowd into cognitive crowd, interest crowd, purchase crowd, and loyal crowd. You can view the total number of consumers and consumer trends. Launch marketing activities according to different groups of people, and check the marketing effects of different groups of people with the returned data.
### AIPL model construction process:
### AIPL model application:
#### AIPL user analysis:
Based on the specific number of users in the four categories of cognition, interest, purchase, and loyalty based on the calculation base date, and the changing trend of consumers. For example, AIPL user analysis interface effect in Quick Audience:
#### AIPL circulation analysis:
Number of users
Shows the number of users of the four categories of users of cognition, interest, purchase, and loyalty at the end of the date interval and the chain difference. For example, the AIPL flow interface effect in Quick Audience:
In the marketing process, users can view the conversion of users according to different marketing results. View the conversion status of users at each level.
User conversion
User conversion volume refers to the number of users converted from a certain type of users into other types of users.
Take the cognitive crowd in the following figure as an example, the user conversion volume is 1, which means that 1 person in the cognitive crowd is converted into an interested, loyal or buying crowd. In the picture below, 1 person is converted into interest.
The number of cognitions on the start date is equal to the number of cognitions, plus the number of cognition user conversions (that is, the number of people who convert from cognition to interest, loyalty, and purchase), plus the sum of the number of cognition lost .
## Issues in the AIPL model construction process:
#### 1. How to prevent data expansion and reduce data storage
Solution: Users who have calculated A, I, P, and L every day only keep incremental data, such as 500w of data on the first day, 501w of data on the second day, of which 2w is added and 1w is lost, and then Based on the original data, there is only a 3w data volume change.
In commercial activities, the actual marketing has precipitated a variety of classic methodologies, guiding the development of corporate business. With the accelerating process of digital intelligence, Quick Audience products incorporate methodology in addition to packaging technology capabilities to help companies make better use of data and achieve continuous growth.
*
Data center is the only way for enterprises to achieve digital intelligence. Alibaba believes that data center is a combination of methodology, tools, and organization, which is "fast", "quasi", "full", "unified", and "passed". Smart big data system.
A series of solutions are currently being exported through , including 16110d97f4fe5f 16110d97f4fe63 general data middle-station solution , retail data middle-stage solution , financial data middle-stage solution , , , Subdivision scenarios such as and other subdivision scenarios for government data middle-station solutions.
Among them, the Alibaba Cloud Data Center product matrix is based on Dataphin, and the Quick series is used as a business scenario cut-in, including:
* -Dataphin, a one-stop, intelligent data construction and management platform ;
* -Quick BI, intelligent decision-making anytime, anywhere;
* -Quick Audience, comprehensive insight, global marketing, intelligent growth ;
* -Quick A+, a one-stop data-based operation platform cross-multi-terminal global application experience analysis and insight;
* -Quick Stock, an intelligent goods operation platform ;
* -Quick Decision, an intelligent decision platform ;
official site:
Data Zhongtai official website https://dp.alibaba.com
Dingding Communication Group and WeChat Official Account
> Copyright Statement: content of this article is contributed spontaneously by Alibaba Cloud real-name registered users. The copyright belongs to the original author. The Alibaba Cloud Developer Community does not own its copyright and does not assume corresponding legal responsibilities. For specific rules, please refer to the "Alibaba Cloud Developer Community User Service Agreement" and the "Alibaba Cloud Developer Community Intellectual Property Protection Guidelines". If you find suspected plagiarism in this community, fill in the infringement complaint form to report it. Once verified, the community will immediately delete the suspected infringing content.
**粗体** _斜体_ [链接](http://example.com) `代码` - 列表 > 引用。你还可以使用@来通知其他用户。