15 essential data indicators for product operations

This is the first lesson I shared with the team on a complete set of design dataization, and it is also the most basic lesson in the entire series.

This is the only lesson that does not involve project content, so I share it with you, hoping to help you quickly understand some basic indicators of data and open the door to data-based design.

This article mainly talks about some basic data indicators of general products to help designers who are not familiar with data get started quickly.

Visualization: User behavior visualization can clearly understand user behavior.

Traceability: By comparing data over a period of time, you can track product data all the way, understand product changes, and track product design issues through abnormal data.

Verifiable: Provide data support and verification of subsequent plans.

Predictable: Through data changes, the direction and trend of the product can be predicted.

Definition: Daily Active User / Weekly Active Users / Monthly Active Users correspond to the number of daily / weekly / monthly active users respectively

For example: On January 17, a total of 3 million users entered the Zhihu page (including those who entered through Zhihu links from other channels), so the DAU on January 17 was 3 million; similarly, WAU, the users who entered Zhihu within 7 days (deduplicated), is the WAU data.

Data usage: Measure the activeness of product usage. It is convenient for product designers to understand the daily user situation of the product and understand the user growth or decrease trend of the product.

Note: If a user accesses Zhihu three times through a channel in one day, only one DAU will be counted.

Definition: The number of users who visit again within a period of time. Retention includes next-day retention, 7-day retention, 30-day retention, etc.

For example: In January, there were 80 new users. In February, 75 of the 80 new users in January visited the product again, so the retention rate in the second month was 75.

Data usage: used to measure the user stickiness of a product and the scale of its retained users.

Note: The number of retained users can well demonstrate the scale of the number of retained users. And understand the stickiness of new users to the product.

Definition: Number of retained users within a certain period/number of visiting users within a certain period.

For example: It is used to measure user stickiness. The general cycle is the next day, 7 days or 30 days. The algorithm for the 7-day retention rate is: the number of deduplicated users who visit again within 7 days/the number of users who visited on the same day 7 days ago. The daily time node is 23:59:59.

Data usage: It is used to measure user stickiness. It can also be used as an important indicator after a product revision. The retention rate has increased, which means that the design revision is successful without changing the core functions.

Note: The monthly retention rate can be used to determine whether the user stickiness of the product is increasing or decreasing at a macro level. This is also the most intuitive data of the product experience. The stronger the product's demand for users and the better the experience, the higher the retention rate.

Definition: Number of lost users/total number of users.

For example: Based on the retention rate, the churn rate for the next month can be obtained through the data of the previous month.

Data usage: Through the churn rate, you can see all churn situations globally, find abnormal churn data, and track what situations caused the churn data in the past.

Note: The abnormal data loss rate can be used to locate the cause of the loss and fix product problems.

Definition: The number of times a product/page/feature is exposed in the field of view.

For example: When a user enters the homepage, it is counted as one display PV, and when the user refreshes the page/exits and comes back again, the PV is accumulated.

Data usage: To determine the number of times this feature/product has been read.

Note: When a user is browsing Zhihu and refreshes the homepage once, PV+1 will be displayed.

Definition: The number of users who are exposed to the operation activities/pages/functions. One terminal is counted as only one UV.

For example: When a user enters the Zhihu interface, it is counted as one display UV, and the UV will not be accumulated if the user exits and re-enters.

Data usage: Determine how many users read this function/interface.

Note: If a user accesses Zhihu 3 times a day, only one UV is counted.

Definition: The number of users who clicked.

For example: If there are 4 million impressions on the homepage, and 100,000 of them click on the question button, then the click UV is 100,000.

Data usage: Understand the number of clicks by users on functional interaction events, and understand user usage behavior through the number of clicks.

Note: Click-through rate can more vividly demonstrate the attractiveness of functional/interactive elements.

Definition: Click PV/display PV.

For example, if the homepage shows 4 million PVs on the day, and 50,000 people click on the question button 100,000 times, then the click rate is 10/400=2.5%

Data usage: Used to measure the attractiveness of the content in a product/page/feature to users, and to compare different features on the same page.

Note: Click-through rate can be divided into PV click-through rate and UV click-through rate.

Definition: Click UV/Show UV

For example, if the number of UVs displayed on the homepage on that day is 4 million, and 50,000 users click on the question button, then the click-through rate is 5/400 = 1.25%.

Data usage: Used to measure the attractiveness of the content in a product/page/feature to users, and to compare different features on the same page.

Note: The difference between UV click rate and PV click rate is whether to deduplicate multiple clicks by the same user.

Definition: Click PV/Click UV

For example, on the Zhihu homepage, on January 16, 100,000 people clicked on the question, and there were 120,000 clicks in total. So the average number of clicks per person is 12/10=1.2 times.

Data usage: Used to measure the attractiveness of the content in a product/page/feature to users, and to compare different features on the same page.

Note: The average number of clicks per person can be used to determine whether users have a strong demand for this function. The value of the average number of clicks per person is always greater than or equal to 1.

Definition: The length of stay of all users and/the number of users

For example, if the total length of time that all users spend on the Zhihu homepage is 1 million hours, and there are 2 million users in total who stay on the homepage, then the average length of time that they spend on the homepage is 0.5 hours.

Data usage: Used to measure page attractiveness. Generally speaking, the longer the stay time, the stronger the user stickiness.

Note: The length of time users stay on a page can be measured for either a page or the entire product. It is not always the case that the longer the stay time, the better. For example, when filling out a form, the longer the stay time, the worse the experience.

Definition: The average time a user spends on a product per day.

For example, on February 20, Zhihu had 1 million users who spent a total of 500,000 hours on Zhihu products. The average usage time per person on February 20 was 0.5 hours.

Data usage: used to measure the depth of user product usage and determine the user's stickiness and dependence on the product.

Note: The longer users use a product, the more dependent they are on the product and the higher its commercial value.

Definition: (number of recommenders - number of detractors) / total number of testers

For example: the number of people who recommend refers to those who give a score of 9-10; the number of people who disparage refers to those who give a score of 0-6.

Data usage: As an important indicator of product word-of-mouth communication.

Definition: The number of users who exited the app from the current page and did not open it again within 30 minutes / the total number of people on the current page

For example: There are 10,000 users who enter the question page, then jump out and do not open it again within 30 minutes. The number of UVs through users on the question page is 50,000, so the bounce rate is 1/5 = 20%.

Data usage: Used to measure the quality of page content.

Note: The bounce rate can sometimes reflect the user's behavior when using a product. For example, when a user is using a photo-taking tool and then saves the application, the bounce rate is particularly high. This shows that the user regards the photo-taking tool as a pure tool software and the traffic utilization rate of other pages is not high. This is normal for users, but it is indeed a waste of traffic for the product.

Definition: Number of completed operations / number of started operations

For example: On January 17, users clicked the question button 100,000 times, and 20,000 users completed the question process. The completion rate is 2/10 = 20%.

Data usage: Used to measure the smoothness of the operation process.

Note: Completion rate is one of the most important indicators in product design. The higher the completion rate, the better the product's operating experience. This directly affects the user experience.

Author: Echo

Source: Echo design notes (uxecho)