Business data analysis of CDA level1 knowledge point summary

Data driven business management method

Generation and application of data ： Remember that the purpose of a business is to make money 、 To earn money , All data analysis 、 Experience reference produces decisions for this purpose . Data analysis is icing on the cake , And personnel experience 、 Executive ability is the timely help . To put it bluntly , Talent is fundamental .

Application and design of indicators

The role of indicators – Quantify business strength with aggregated data
Understanding of indicators – Summary of metrics , Summary rules of data
Characteristics of indicators ： The index is Free state Of , Such as total sales , need Combined with statistical dimension , Clarify the object of statistical indicators , For example, under the product time dimension ,2021 Total sales of a product in . Indicators can be freely combined to form new indicators , To adapt to the changing business environment .
Basic indicators ： Summation class （ Conventional summation 、 Cumulative sum 、 Cumulative indicators in the time dimension ）、 Count class 、 Comparative （ Percentage difference 、 Average ratio 、 Fixed base growth rate 、 Month on month growth rate ）

Common scenario indicators

1. Flow related indicators

The amount
CPM– Every thousand exposures 、CPC– Charge per click 、CPS– Charge according to actual sales
UV Number of visitors ： The number of non duplicate users accessing the corresponding dimension
PV Browse volume ： Total number of page views under dimension
Visits Number of visits ： Count the number of sessions
Number of new visitors

quality
Average access depth ： Browse volume / Number of visits
Jump loss rate ： Number of jumps / Number of visits （ Skipping refers to leaving the page without clicking ）
Proportion of new visitors ： Number of new visitors / Number of visitors

2. Transform relevant indicators

Conversion rate
Golden three swordsman ： Traffic 、 Conversion rate 、 Customer unit price

3. Operation 、 Sales related indicators

turnover ：GMV、 Actual sales 、 After tax sales 、 Refund amount
volume ： Actual order quantity 、 sales 、 Refund order quantity 、 Number on shelves
completion ： Target achievement rate 、 Return rate
effect ： Screen effect 、 Commodity relevance

4. Inventory related indicators ( Difficult to understand )

Days of inventory turnover = Inventory / lately n Daily average sales
Library age = Storage duration
Number of inventory turns = Average inventory / Total stock out
Order fulfillment rate = Normal supply times according to reasonable demand / The total number of times the product is required to be supplied
Out of stock rate = Number of out of stock / Total order times
Sold out rate =1- Inventory amount / Purchase amount

5. Performance indicators

6. Customer related indicators

The amount ：
Number of registered users 、 Number of login users 、 Number of browsing users

quality ：
Number of active users 、 Number of re purchase users 、 Retention 、 Online hours

Design new indicators

Index analysis method – Longitudinal analysis （ The change law of the index under the time cycle ）、 Horizontal analysis （ Specify the difference of indicator values of different items under the dimension ）、 Early warning analysis （ The cumulative value or instantaneous value of the indicator under the corresponding dimension of the current time node , Compare the reference value ）

Customer analysis

User life cycle

Customer source analysis
Channel user value contribution ：a. Instantaneous value – User volume ;b. Long term value – User quality
Retention analysis
Retain users ： After the user registers, he starts to contact the product , Users who use the product without leaving .

User churn analysis
Loss definition ：
Not clearly defined , according to APP Specific business behavior . Such as user 30 No purchase occurred within days ; If the user uninstalls APP ; As recently 2 The user has not logged in for months
User behavior value analysis
Online retailers app For example , Number of pages viewed 、 Browsing duration 、 Number of logins 、 Number of items collected 、 Collection and purchase times . Number of additional purchases 、 Pay the amount 、 Days between payments 、 Number of sharing products

Behavior effect analysis

Business analysis model

RFM Model
Important value users – Keep the status quo
Important development users – Increase frequency
It's important to keep users – User reflow
Important to keep users – Key recall

 Customer loyalty model 
![ Insert picture description here ](https://img-blog.csdnimg.cn/40a7c2ae5c434642b5898c30d9060396.png?x-oss-process=image/watermark,type_ZHJvaWRzYW5zZmFsbGJhY2s,shadow_50,text_Q1NETiBA54uQ54u455qE5bi95a2Q,size_20,color_FFFFFF,t_70,g_se,x_16)

 Funnel model 

Business analysis method

Tree structure analysis – Drill down 、 The pyramid

28 analysis –80% The problem is that 20% Caused by , Used to find the core problem .
Four quadrant analysis – Boston matrix （ For commodity analysis ）

Simultaneous group analysis – Quantitative behavioral indicators , Analyze the changes of this index over time in different groups
Contemporaneous group – Within the specified time period or at the specified geographical location , Users with the same characteristic attributes



other ：LTV User life value ：LTV=LT*ARPU, LT It's the user life cycle ,APRU Is the average user income .

1. When viewing the transaction data of China's e-commerce market , You will find that due to double 11 The existence of , The trading volume of China's e-commerce market will increase significantly in the fourth quarter of each year . This feature reflects the transaction data of e-commerce market （）
A. Long term trends
B. Seasonal changes
C. Cyclical changes
D. All of these are
answer ：B.
Wrongly selected as C. But cycle change generally refers to 2-3 A cycle of more than years .

2. The annual profits of three enterprises are 5000 Ten thousand yuan 、8000 Million and $ 3900 Ten thousand yuan , Then there is （ ） A variable
A.0 individual
B.2 individual
C.1 individual
D.3 individual
answer ：B. Enterprise and annual profit are taken as variables respectively .（ A silly question ）

3. The following statement about the data analysis process is wrong （）
A. Classical statistical analysis focuses on theoretical testing of the applicability of statistical models
B. The problem of cost and benefit runs through the data analysis
C. The same data analysis methodology applies to all areas of analysis
D. The idea of confirmatory data analysis is conducive to better guide the practice of data analysis
answer ：A. Wrongly selected as C, Because I think C All of this word is too absolute , I didn't expect it to be right again .A What's wrong ？