3 key points: How to disassemble information flow products?

News and information flow has now become the first interface between content-based Internet products and users.

Information flow product managers need to make choices between content positioning and business balance based on North Star indicators such as user dwell time, product conversion rate, content clicks, etc.

▲ Information flow products are composed of different feed types

If there are too many commercial advertisements, it will affect the reading experience and user stickiness of the consulting information flow product. If there are too few commercial advertisements, the product will not be profitable.

1. What is an information flow product?

Information flow products are essentially a "content distribution channel" that distributes contributed, created, and collected content through different sections of the platform to reach users.

The distribution method needs to solve three user difficulties, namely:

1. Content Creation

2. Content Distribution

3. How to solve the problem that if only content is generated based on user preferences, the information flow cannot retain content creators; if only user preferences are satisfied but there is no continuous content creation, the information flow readers cannot be retained.

First of all, we need to understand the two information flow structures of information flow products.

Information flow products can be layered using a tree stump structure, using the "two-part method" to help users quickly locate the content they want.

For example, use the "two-point method" to simulate the shortest path for user traffic information content.

▲ Tree structure diagram

For example, in a range of (1, 10), select a number, such as 3, and let user B guess the number.

How can B guess correctly? What is the easiest path?

Bigger than 2 but smaller than 5

Less than 4 but not more than 3

The final result is 3.

The above is the shortest path to 3 using the "two-point method".

The article details product design of Nuggets adopts a tree structure, which divides the titles into layers to help users locate the details of each chapter.

▲ Nuggets article directory navigation

There are also PMTalk topics and article details under the topics, which are still in a tree structure.

▲ The tree-like content layout of PMTalk Product Manager Community

▲ PMTalk topic list

Information acquisition from information flow does not require users to obtain a certain length of information content or image length, but rather the amount of information obtained.

For example, using 1,000 words + countless pictures and texts to express the origin of the Christmas Eve story and using 100 words + 1 picture to express the origin of the Christmas Eve story express the same amount of information.

Express information:

Users learned how Christmas Eve came about.

2. Topology Network Product Structure

The above mentioned is a stump structure that helps users quickly extract information. There is also a topological network structure that provides random paths to content viewers.

This ensures that every piece of content has a chance to be accessed.

The content will not be silenced because of timeline or dislike.

For example, for the related recommendations on Zhihu and a certain community, the order in which users visit is the topological network structure.

▲ Related recommendations from Zhihu

3. Information Flow Product Architecture

Information flow products can also build product architecture from the application layer to the bottom layer.

The following is the information flow from the front end (APP, web, third party), business carrier end, technical layer, and data storage.

▲ Information product architecture

1. Technical layer

It includes general technologies, algorithm models, including data collection, labels, user models, etc. for subsequent recommendation engines, all of which are at the technical level.

2. Business

The business is carried out to meet the needs of information product operators and managers. Includes paid users, whitelist users, editing, deleting, pinning, and recommending information content.

3. Data Storage

From the perspective of product managers, data storage stations need to pay attention to data storage time, cost, data stability, etc.

4. Why are information flow products so difficult?

Information flow products consist of two parts:

(1) Information flow product framework

(2) Content ecology of information flow

Many information flow products have no readers, short user stay time, and low growth. There are three problems:

(1) The content of the information flow is not good and cannot be liked by users

(2) The information flow product framework is poor, and users spend time reading the information flow and have a poor experience.

(3) The content is not well done and the product framework is not perfect.

1. Targeted content operations

User portrait, market value, and continuous content system engine

2. Design for products

The information content publishing experience of the process and the content migration of the product framework (web, H5, client) are all difficulties for product managers in building information products.

5. Data indicators of information flow products

The data indicators of information flow products can be divided into three aspects.

1. Content Production

The number of UGC contents produced, the number of user interaction behaviors (likes, comments, reposts, shares), and the calculation indicators are (hours/days/months/quarters/years), which can be used to check the content production trends. For example, the number of PMTalk columns, questions, application recommendations, and application Q&A...

2. Product functional indicators

Information flow product activity (different information products have different definitions of activity), user stay time, page visit depth, information flow product heat map, also calculate the month-on-month (daily, monthly, quarterly, annual), function usage number...

3. Business indicators

Number of paid ad clicks, conversion percentage, number of positive orders, number of reverse orders, month-on-month comparison (daily, quarterly, monthly, annual), number of registered users, number of paying users…

Build your own North Star indicator through the above three indicator dimensions. Only with the North Star Indicator can you understand the actual status of the information flow product.

Author: Kevin's Little Changes in the World

Source: Kevin's Little Changes in the World