Data analysis thinking-business indicators

Preface

For people in the data industry, two important abilities are required:

1. Understand the data, know how to find business indicators from the data
2. Use relevant indicators to analyze data, use multiple indicators to analyze a problem

1. How to understand the data?

When you get the data, you can understand the data according to the following steps:
(1) The first step: understand the meaning of each column of data
(2) the second step: classify the data, usually the data is divided into three categories

• User data (the basic information of the user)
• Behavioral data (that is, recording what the user did)
• Product data (what is sold, that is, everything in a platform can be regarded as a product)

2. What are the commonly used indicators?

Indicator: It is to use a unified standard to measure the business, and this unified standard is the indicator.

1. User data indicators

Users generally include new users, active users, and retained users. Among them, active users correspond to inactive users, and retained users correspond to lost users.
User data indicators include:

1. Indicator used for new users: number of new users per day
2. Indicators used for active users: active rate
3. Indicators used for retention of users: retention rate

(1) Daily number of new users
means the number of new users every day
(2) Active rate
means the proportion of active users in total users, which can be divided into daily active rate (DAV), weekly active rate (WAV), monthly Active rate (MAV)
(3) Retention rate
retention : New users promoted through channels may gradually lose some users over a period of time, and the remaining users are retained users.
The role of retention : You can evaluate the stickiness of product features to users.
Retention rate = Among the new users on the first day, the number of users who have used the product on the N day / the number of new users on the first day

2. Behavioral data indicators

Behavioral data related indicators include: PV, UV, forwarding rate, conversion rate, K factor
(1) PV and UV

PV: indicates the number of visits, that is, the number of times a webpage is viewed within a certain period of time
UV: indicates the number of visitors, that is, the number of people who visit a certain webpage in a certain period of time

Function: Through the size of PV and UV, you can see what functions the user likes and dislike, and always optimize the product.

(2) Forwarding rate
Forwarding rate = the number of users who forward a function / the number of users who see the function

(3) Conversion rate
The calculation of conversion rate is based on specific business scenarios.
For example: shop conversion rate = number of people buying products / number of people going to the store

(4) K factor is
used to measure the effect of recommendation, that is, how many new users can be brought by initiating a user.
K factor = average number of invitations sent by each user * the conversion rate of those who received the invitation to convert into users
K>1 means that the number of new users will increase like a snowball, less than 1 when the number of users reaches a certain scale Will stop growing through self-propagation

3. Product data indicators

Product data indicators include:

1. Indicators used to measure the total amount of business: such as total transaction volume, transaction volume
2. An indicator used to measure the per capita situation: such as the unit price per customer
3. Indicators used to measure payment: such as payment rate, repurchase rate
4. Product-related indicators

(1) Total
amount The total amount includes the total amount of transactions (GMV) , the number of transactions and the length of the visit

The total transaction amount includes sales, canceled order amount, rejected order amount, and return order amount. For e-commerce products, the
transaction amount is the number of products ordered . For the education industry, it is the number of courses ordered.
Visit time refers to the total time the user uses the app or website

(2) Per
capita The indicators used to measure the per capita situation are:

• Per capita payment = total revenue / total users. Per capita payment is also called ARPU in the gaming industry, and it is also called the customer unit price in the e-commerce industry.
• Per capita payment of paying users (ARPPU) = total income / number of paying users, this indicator is used to calculate the average income of paying users.
• Per capita visit time = total time / total number of users, used to count the average time that each person uses the product

(3) Payment The
relevant indicators of
payment are: payment rate = number of paying people / total number of users, the payment rate can reflect the product's monetization ability and user quality.
Repurchase rate = the number of people who have spent more than two times/the number of people who have paid. It is the frequency of repeated purchases and is used to reflect the frequency of users' payment.

(4) Product
Product: refers to measuring which goods are good and which goods are not good from the perspective of goods. Focus on selling by finding good products, and analyze the reasons for bad products.
Several common indicators are: what are the top-selling products, good products, and bad products? Here according to specific business needs, flexible expansion of use.

4. Promote paid metrics

In paid advertising and promotion, it involves examining the promotion effect indicators, which can be divided into different payment channels:

channel	Promotional payment metrics	meaning
Display ad	Pay per impression (CPM)	How many people saw the ad
Search ads	Pay per click (CPC)	How many people clicked on the ad
Information flow advertising	Pay per click (CPC) or actual delivery effect (CPA)	How's the effect

3. How to choose indicators?

When selecting indicators, two points need to be paid attention to:
(1) A good data indicator should be a ratio
(2) According to the current business focus, find the North Star indicator

4. Index system and reports

1. What is an indicator system

The indicator system sorts out the business from different dimensions and organizes the indicators systematically. In short, index system = index + system, so an index cannot be called an index system, and several unrelated indexes cannot be called an index system.

2. What is the use of the indicator system

Functions include:

• Monitor business conditions
• Find current business problems through dismantling indicators
• Evaluate the areas where the business can be improved and find out the direction of the next step

3. How to establish an indicator system

Process of establishing index system:

(1) First, clarify the KPI of the department and find the first-level indicators. The first-level indicators are the most core indicators used to evaluate the company or operation.
(2) Second, understand business operations and find secondary indicators.
(3) Third, sort out the business process and find the three-level indicators
(4) Fourth, continuously update the indicator system through report monitoring indicators

Report preparation steps:

1. Demand analysis: What is the purpose of making reports?
2. Establish an index system: What kind of index system can meet business needs?
3. Design the presentation form: How to display the data so that the business can understand the data more easily?
4. Prepare requirements documents: explain the report prototype, update frequency and other information
5. Report development: communicate and schedule with development, follow up the development progress, and verify the data quality of the report.

Precautions for establishing the index system:

1. Without first-level indicators, can not grasp the key points
2. There is no logical relationship between the indicators. When there is a problem, if the corresponding business node is not found, there is no way to solve the problem.
3. The dismantled indicators have no business meaning
4. One person completes the indicator system and reports, and does not communicate with the business department