Community Discussion · Policy

New Media Operations Practical Manual for Three Platforms - Part A Collection & Part B Acceptance

stevenAug 212026/08/21 479 views

WorkBuddy AI Software Learning & Traffic Generation Plan: Douyin "A-Attract, B-Capture" · Xiaohongshu · Bilibili Three-Platform Practical Manual

Plan Positioning: Initially focus on promoting WorkBuddy and similar AI software to help ordinary people learn AI and improve themselves. Use a three-tier funnel of "Personal Account → Official Blue-V Account → Physix Frontier Website" to ultimately direct traffic to the website (physixfrontier.com/cn). Website traffic is the core KPI.

Version: v4 · 2026-08-20

Applicable Platforms: Douyin / Xiaohongshu / Bilibili

──────────────────────────────────────────────────

I. Core Strategy Overview

1.1 One-Sentence Positioning

Using the "Douyin A-Attract, B-Capture" model as the core traffic generation framework, combined with content matrices on Xiaohongshu and Bilibili, create content around "AI software like WorkBuddy helping ordinary people learn AI." Guide anxious professionals and college students from content platforms to the official Physix Frontier website (physixfrontier.com/cn), completing the conversion from "seeing content" to "actively visiting the website for in-depth use."

1.2 Core Model: A-Attract, B-Capture (AI Learning Edition)

A (Personal account, posting AI software learning content, capturing organic traffic)

→ Middle (Soft guidance redirecting to Blue-V homepage)

→ B (Blue-V official account, hosting website link for capture)

→ Physix Frontier Website (Endpoint · Core KPI)

Why use A-Attract, B-Capture:

Personal accounts have strong organic traffic but cannot engage in commercial activities or risk throttling; they also cannot compliantly host external links.
Blue-V enterprise accounts can host official website links but have poor organic traffic.
The two complement each other: Account A handles traffic acquisition and direction, while Account B handles compliant capture and endorsement.
The Blue-V account is the only node in the entire chain that can compliantly host external website links.

1.3 Target Audience and Pain Points

Audience

Profile

Core Pain Point

Content Angle

Office Professionals

25-35 years old, administrative/HR/finance/operations roles

Fear of being replaced by AI, fear of layoffs, fear of being left behind by the times

"Use AI to keep your job," "Transitioning office roles into AI tool users"

College Students / Fresh Graduates

18-24 years old, seeking to enhance competitiveness

Job hunting anxiety, unsure what AI skills to learn, fear of unemployment upon graduation

"5 must-learn AI tools for college students," "Using AI to write resumes/papers"

Career Changers

28-40 years old, traditional industries looking to switch to AI-related fields

Don't know where to start, perceive high barriers to entry for AI, fear inability to learn

"First lesson in learning AI from zero," "Is it too late to switch to AI at age 30?"

SME Owners / Self-Employed

30-45 years old, seeking cost reduction and efficiency improvement

Heard AI is powerful but don't know how to use it, fear falling behind peers

"One AI tool equals 3 employees," "Must-learn AI tools for small business owners"

1.4 Value Proposition

Enable ordinary people to use AI tools like WorkBuddy to learn AI from scratch, improve themselves, and avoid being left behind by the times.

Not teaching you to code, but teaching you to use AI software.
No technical knowledge required; just follow the steps to learn.
10 minutes a day doubles your workplace competitiveness.
Complete learning path available on the official website (guiding landing).

1.5 Three-Platform Division of Labor and Priority

Platform

Monthly Active Users (2025 Data)

Core Role

Traffic Path

Time to Results

Douyin

1.022 billion

Main traffic engine (A-Attract, B-Capture model)

Account A short videos → Soft redirect → Account B Blue-V → Official website link

1-4 weeks

Xiaohongshu

242 million

Deep seeding + Search SEO

Image-text notes → DM for materials → Fan groups → Official website

2-6 weeks

Bilibili

223 million

Trust building + Long-tail search

In-depth tutorials/reviews → Bio links → Official website

1-3 months

Data source: QuestMobile May 2025 Monthly Active User data.

Priority Ranking: Douyin > Xiaohongshu > Bilibili

Douyin has the largest traffic entry point; the A-Attract, B-Capture model is proven viable, so prioritize efforts here.
Xiaohongshu supplements long-tail search traffic, with low costs and long content lifecycles.
Bilibili builds trust through in-depth tutorials; content can be reused as master material for Douyin and Xiaohongshu.

1.6 Traffic Acquisition Logic: Four-Step Funnel

Step 1 · Pain Point Attraction: Don't sell products; sell the antidote to anxiety. When users search for "Will AI replace me?", "What should office workers do?", or "Learn AI from scratch," the content appears in search results.

Step 2 · Content Seeding: Use content ranging from 15 seconds to 15 minutes to demonstrate the real effects and operational steps of AI tools like WorkBuddy. Users feel after watching, "I can learn this too."

Step 3 · Authoritative Trust: Create systematic tutorials, cross-tool reviews, and learning path planning to establish professional trust with users in the decision-making phase who "want to learn but don't know where to start."

Step 4 · Website Conversion: All content guides users to the Blue-V official website link via compliant paths, leading them to the Physix Frontier website to access complete tutorials and tools.

1.7 Content Ratio Allocation (Douyin Account A)

Content Type

Ratio

Purpose

Corresponding Website Module

AI Tool Tutorials

40%

Attract search traffic + Capture organic recommendations

WorkBuddy Feature Tutorials

Pain Point Resonance

25%

Hit anxieties, trigger shares

Survival Guide for the AI Era

Review & Comparison

20%

Build professional trust + Seed WorkBuddy

AI Tool Cross-Reviews

Learning Path

15%

Systematic feel + Guide deep learning on official site

Learning Paths/Courses

──────────────────────────────────────────────────

II. List of Information Pending Confirmation

The following items are marked 🔴 as questionable. The current plan proceeds based on default suggestions and can be adjusted upon user confirmation.

No.

Item Pending Confirmation

Default Suggestion

Reason

Scope of Impact

Whether the ICP filing entity matches the Blue-V certification entity

Assume they match; plan assuming official website links can be hosted

The Blue-V official website link component requires the domain's ICP filing entity to match the business license entity. This is a prerequisite for the A-Attract, B-Capture model to work.

Determines whether the entire Account B capture chain is viable

Ad budget scale

Start with a small budget (DOU+ 100-300 yuan/post, monthly budget 2000-5000 yuan)

First validate content direction with organic traffic, then amplify viral posts with paid ads; risk is controllable

Impacts the intensity of paid promotion in Phase 2 and the speed of matrix expansion

Whether Account A uses a real persona or virtual persona

Recommend using a real persona (AI Learning Coach)

Real personas have higher trust, better comment interaction, and faster account growth; virtual personas have no exposure risk but incur higher trust-building costs

Impacts Account A's content format and growth speed

Launch single Blue-V first or start A+B simultaneously

Recommend starting A+B simultaneously

Account A acquiring traffic + Account B capturing is a complete closed loop; starting together validates the entire chain faster. Single Blue-V has poor organic traffic, making standalone operation less meaningful.

Impacts the launch pace of Phase 1

──────────────────────────────────────────────────

III. Douyin Content Operation Strategy (A-Attract, B-Capture Model · AI Learning Edition)

3.1 Role and Division of Labor Definition

Role

Douyin Account Type

Core Task

Can Host External Website Link?

Traffic Characteristics

Quantity

A (Traffic Account)

Personal Account

Post AI software learning content, capture organic recommendation traffic, guide to Blue-V

❌ Cannot

✅ Strong organic flow

1-2 initially, matrix later

Middle (Redirect Component)

——

Personal account bio @Blue-V + Verbal guidance + Pinned comments

——

Soft redirect, avoids violations

——

B (Capture Account)

Blue-V Enterprise Account

Corporate endorsement, host official website link on homepage, unified capture

✅ Can (Only compliant entry point)

⚠️ Poor organic flow

Endpoint

Physix Frontier Website

Consolidate complete AI learning tutorials, tool resources, learning paths

——

Owned Asset

——

3.2 Account A Positioning and Persona

A₁ AI Software Learning Mentor Account (Recommended Priority):

Persona: AI Software Learning Mentor / AI Learning Coach. Adopt the tone of "Teaching you one AI tool every day," acting like a real AI learner sharing practical tips. Persona keywords: Down-to-earth, no jargon, understandable for beginners, progress little by little every day.
Name Suggestions: Include keywords like "AI," "Learning," "Tools," "Tutorials," e.g., "AI Learning Coach Little X," "Learn an AI Tool Daily," "AI Beginner Evolution Diary," "WorkBuddy Usage Guide."
Bio Template: Teaching you one AI tool daily | Learnable from scratch | Full tutorials and tools on @PhysixFrontierOfficial homepage website.
Content Format: Screen recording operations + voiceover explanation, or live-action demonstrations (live action recommended for real personas). Focus on "Follow along and you'll learn."

A₂ Vertical Tool Account (Optional Second Account):

Persona: Vertical AI tool review account, e.g., "AI Tool Review Room," "WorkBuddy Practical Guide."
Name Suggestions: Directly state the tool or niche, e.g., "Daily AI Tool Recommendations," "WorkBuddy Tips Station."
Bio Template: Sharing one practical AI tool daily | Full usage tutorials on @PhysixFrontierOfficial.
Content Format: Tool demos + effect comparisons, no need for live-action appearances.

Unified Command Rule: All Account A scripts consistently point to the same Account B (Blue-V) → The same official website link.

3.3 Account B (Blue-V) Positioning and Capture

Must Complete:

Corporate Certification (600 yuan/year) — Requires business license + matching ICP filing entity.
Decorate homepage to host the "Official Website Link" component → Pointing to physixfrontier.com/cn/.
Clearly state in homepage bio: Physix Frontier · AI Learning Platform for Ordinary People · WorkBuddy Official Tutorials.
Unify brand visuals for avatar and cover image.
Pin 3 videos: WorkBuddy Introduction, AI Learning Path, Official Website Usage Guide.

Account B Content Strategy:

Posting Frequency: 2-3 posts per week, focusing on corporate endorsement and tutorial-style content.
Content Types: WorkBuddy feature updates, AI learning path releases, tool collections, livestream replays.
Purpose: Not to acquire traffic, but to serve as a trust endorsement landing page—users click into the Blue-V homepage, see it's a legitimate corporate account with an official website link and proper tutorials, only then dare to click the external link.

3.4 Compliant Soft Redirect Writing (Middle)

✅ Compliant Practices:

Location

Writing Example

Personal Account Bio

"See full AI tool tutorials on @PhysixFrontierOfficial homepage website link"

Video voiceover

"For more complete step-by-step instructions and tool links, check the pinned official account on my homepage—visit the website there."

Pinned comment section

"See @WujieFrontierOfficial's homepage for detailed tutorials and tool download links."

End-of-video subtitles

"More tutorials → Homepage @WujieFrontierOfficial"

❌ Prohibited Practices (2025 New Regulations: Violations Result in Traffic Restriction/Account Ban):

Directly posting URLs in videos or comments
Sending links or WeChat IDs via private messages
Guiding users to add WeChat/QQ or other third-party platforms
Relying on Little Windmill/Lead Generation Components (Construction stopped 2025.9.19; fully disabled 10.8)

3.5 Content Topic Directions and Specific Topic List

Content Production Mechanism (Website → Douyin Material Mapping):

Website Module

Douyin Content Format

Role

Call-to-Action Landing Point

WorkBuddy Feature Tutorials

60-second AI Tool Teaching Short Videos

Account A drives traffic

"Full tutorial on the official website"

AI Tool Collections

"3 Must-Learn AI Tools" Roundup

Account A drives traffic

"Tool collection on the official website"

AI Learning Paths

"Zero-Basis AI Learning Roadmap"

Account A drives traffic

"Complete learning path on the official website"

Tool Comparison Reviews

WorkBuddy vs Other AI Tools

Account A drives traffic

"Detailed comparison on the official website"

Workplace/Student Case Studies

"Boost Efficiency 10x with AI" Real Cases

Account A drives traffic

"More cases on the official website"

Category A · AI Tool Teaching Type (40%, Attracts Search + Recommendation Traffic)

No.

Topic

Search Keyword Coverage

Hook Design

Use WorkBuddy to finish a PPT in 10 minutes; boss praises directly

WorkBuddy, AI make PPT

Efficiency comparison + contrast

This AI tool helps you write weekly reports; no more overtime needed

AI write weekly report, AI office tools

Pain point + solution

AI writing tool usable even by beginners; tested and it's great

AI writing tool, AI writing

Low barrier + effect demonstration

Hidden features of WorkBuddy that 90% of people don't know about

WorkBuddy tips, AI tool tips

Suspense + numbers

Make a professional poster in 1 minute with AI; designers are shocked

AI image making, AI design

Effect demonstration + contrast

5 essential AI tools for workers; double your efficiency

AI tool recommendations, office AI

Numbers + list

How to use WorkBuddy? Most comprehensive beginner tutorial

WorkBuddy tutorial, AI entry-level

Search terms + hand-holding guide

Organize meeting minutes with AI; handle 2 hours of recording in 5 seconds

AI meeting minutes, AI audio transcription

Efficiency comparison + scenario

Category B · Pain Point Resonance Type (25%, Hits Anxiety, Encourages Shares)

No.

Topic

Pain Point Direction

Duration

35-year-old admin staff; if you don't learn AI now, you'll really get laid off

Workplace anxiety, replacement

60s

College students who can't use AI face real difficulties finding jobs after graduation

Job search anxiety, competitiveness

60s

The difference between earning 5,000/month and 50,000/month is whether you can use AI

Income gap, self-improvement

60s

Stop scrolling short videos; isn't spending 10 minutes daily learning AI better?

Self-discipline, growth

60s

The new post-00s employee at the company finished my week's work in one day using AI

Workplace competition, efficiency

60s

The fastest way for ordinary people to learn AI is not learning programming

Learning methods, zero basis

60s

It's not humans being replaced by AI, but those who don't know how to use AI

Sense of crisis, call to action

60s

Category C · Review & Comparison Type (20%, Establishes Professional Trust)

No.

Topic

Content Direction

WorkBuddy vs Other AI Assistants: Which is Better for Ordinary People?

Horizontal comparison + scenario recommendation

Tested 10 AI writing tools; only kept these 2 in the end

Review + filtering + recommendation

Are free AI tools actually usable? I tested 5 to tell you the answer

Free vs Paid + real testing

Top 5 AI Tools Worth Learning in 2026; Don't Learn the Wrong Ones

Annual roundup + learning advice

Why Do I Recommend Ordinary People Start with WorkBuddy?

Product recommendation + reasoning

Category D · Learning Path Type (15%, Systematic Feel + Guides to Official Website)

No.

Topic

Content

Complete Roadmap for Learning AI from Zero Basis (Recommended to Save)

Learning path + stage goals

For Ordinary People Learning AI, These 3 Tools Are Enough to Start

Entry-level recommendations + priority

30-Day AI Learning Plan: 10 Minutes Daily, Even Beginners Can Learn It

30-day plan + actionable

AI Learning Roadmaps for Different Roles; Just Match Your Position

Role matching + personalization

3.6 Viral Structure Formula

First 3 Seconds Completion Rate > 45% Required for Recommendation Weight Boost (Ocean Engine data). Structure uses "Pain Point/Hook + Effect Demonstration".

Title Formula: Search Keywords + Numbers/Emotion + Suspense

Template No.

Formula

Example

Search Term + Number + Result

WorkBuddy Tutorial | Learn to Make PPT with AI in 5 Minutes

Pain Point Scenario + Solution

Don't Want to Be Replaced by AI? Learn These 3 Tools First

Comparison + Suspense

WorkBuddy vs Doubao: The Results Were Unexpected

"90% of People Don't Know" + Info

Hidden WorkBuddy Features That 90% of People Don't Know About

Identity + Experience + Conclusion

After 3 Months of Learning AI, Here Are My 5 Honest Takeaways

Golden First 3 Seconds (Determines Completion Rate):

Type

Opening Writing Style

Example

Pain Point Type

Directly hits user anxiety

"35-year-old admin staff; if you don't learn AI now, you'll really get laid off"

Effect Type

Show results first, then teach method

"With this AI tool, I finished a day's work in 10 minutes"

Counter-Intuitive Type

Breaks fixed perceptions

"You don't need to learn programming to master AI; here's the fastest way for ordinary people"

Number Type

Uses numbers to create impact

"5 AI tools boosted my efficiency by 10x"

Question Type

Directly asks search intent

"Where should ordinary people start when learning AI?"

Body Structure (60-Second Version):

0-3s: Hook (Pain point/Effect/Counter-intuitive/Numbers/Question)

3-10s: Introduce tool/topic + One sentence clarifying value

10-45s: Core content (Operation demo + Key steps + Effect comparison, with subtitles)

45-55s: Summary/Perspective elevation

55-60s: CTA Guidance ("Full tutorials and more tools are in the pinned official account on my homepage; visit the website there")

3.7 Cover Copy

Douyin Cover = Video First Frame Image + Overlaid Large Text Title. Large text must include search keywords.

Cover Type

Copy Elements

Example

Tool Teaching Type

Tool Name + Effect

"WorkBuddy Makes PPT"

Pain Point Type

Pain Point + Number

"Age 35: Too Late If You Don't Learn Now"

Comparison Type

Left VS Right

"Before Using AI vs After Using AI"

Number List Type

Number + Keyword

"5 Must-Learn AI Tools"

3.8 Tag Strategy

Add 5 to 8 tags per video, divided into three layers:

Big Tags (1-2): #AI #ArtificialIntelligence #AITools
Medium Tags (2-3): #WorkBuddy #AILearning #OfficeEfficiency #AITutorial
Long-Tail Tags (2-3): #LearnAIFromZero #AIToolRecommendations #EssentialForWorkers #EssentialForStudents

3.9 Posting Time Slots and Frequency

Time Slot

Recommended Time

Reason

Morning Rush Hour

7:30-8:30

Commuters watching videos; suitable for quick tool news

Lunch Break

11:30-13:00

Fragmented lunch time; suitable for tool tutorials

Evening Rush Hour

17:30-19:00

Post-work commute; suitable for review/comparison types

Late Night

21:00-22:30

Pre-sleep browsing; high completion rate; suitable for deep tutorials and pain point resonance

Frequency: During cold start phase, 3-5 posts per account per week; run for 4 weeks to observe data
First 3 Days: Only post tool tutorials without hard ads; nurture account + test algorithm tags
Content Ratio: First 3 days: Tool Tutorials 60% + Pain Points 20% + Reviews 20%

3.10 Comment Section Operations

Action

Specific Practice

Purpose

Pinned Comment

Pin on each video: "Detailed tutorials and tool downloads are in @WujieFrontierOfficial's homepage website link"

Guide users to Blue V to find links

Reply Speed

Reply to all comments within 1 hour of posting

High interaction rate boosts recommendation weight

Interaction Guidance

Voiceover at end of video: "Which AI tool do you want to learn next? Tell me in the comments"

Increase comment rate

Regular User Comments

Arrange 2-3 sub-accounts to comment "Where to learn full tutorial?" "How to find the official website?"

Naturally guides users to the pinned comment

Topic Extension

Supplement operational tips not covered in the video in the comments

Increases comment depth

3.11 Complete Conversion Path

Account A Short Videos (Drive organic traffic)

→ Pinned comment/Voiceover guidance "Go to official account to see website"

→ Users click Account A homepage → Bio mentions @WujieFrontierOfficial

→ Click to enter Account B Blue V homepage

→ Blue V homepage 'Website Link' component

→ WujieFrontier Website (Endpoint)

3.12 Data Review Metrics (Weekly Review · Website Traffic as Core KPI)

Metric

Passing Line

Excellent Line

Meaning

Completion Rate

>30%

>50%

Is content pacing good?

Interaction Rate

>3%

>8%

Do users like it?

Homepage Click-Through Rate

>2%

>5%

Is guiding to homepage effective?

Blue V Jump Rate

>1%

>3%

Efficiency of A→B soft jump

Follower Growth Rate

>0.5%

>2%

Does content have retention value?

Search Exposure Share

>15%

>30%

Is SEO working?

Website Link Click-Through Rate

>5%

>10%

Account B conversion efficiency

Website UV (Core KPI)

Daily Increase > 20

Daily Increase > 100

Ultimate Goal: Website Traffic

Review Method: Each week, select the top 3 and bottom 3 videos by view count. Compare topics/titles/openings/posting times to find patterns and replicate viral hits. Focus on 'which type of content brings the most clicks to the official website'.

──────────────────────────────────────────────────

IV. Xiaohongshu (RED) Content Operations Strategy (AI Learning Edition)

4.1 Account Positioning

Persona: AI Learning Coach / AI Tool Curator. Positioned as "helping ordinary people find the most practical AI tools and learning methods," with highly altruistic content.
Name Suggestions: Include keywords like "AI," "Learning," "Tools," "Tutorial," e.g., "AI Study Notes," "AI Tool Lab," "Beginner's Guide to AI."
Bio Template: Focused on AI learning for ordinary people / Sharing practical AI tools daily / Learnable from zero basics / Full tutorials pinned on homepage.
Homepage Setup: Do not directly place website links (this triggers traffic throttling). Use "DM to receive materials" for compliant traffic diversion to private domains, then guide users to the website.

4.2 Content Formats

Core Format: Primarily image-text notes (cover image + multi-image steps + text description), occasionally short videos.
Cover Image: Tool screenshots/effect comparisons + large title text + numbers. Must ensure "the topic of this note is instantly clear at a glance."
Number of Images: 6 to 9 images. The first is the cover; the last one is a "Call to Action: Follow/DM to receive materials."
Body Text: Write in bullet points (using 1. 2. 3. numbering), one sentence per point, paired with corresponding images.

4.3 Topic Directions & Specific Topic List

Category A · Tool Lists (40%, Core Search SEO)

No.

Topic

Targeted Search Terms

Type

10 Must-Learn AI Tools in 2026 | Beginner-Friendly

AI tool recommendations, AI tool rankings

Long-tail

5 Essential AI Tools for Office Workers | Double Efficiency

Office AI tools, productivity tools

Long-tail

Must-Read for College Students Learning AI | 5 Tools to Boost Competitiveness

College student AI, AI learning

Trending

Complete WorkBuddy Usage Guide | From Beginner to Expert

WorkBuddy, AI assistant tutorial

Trending

Free & Useful AI Tools | These 8 Are Truly Great

Free AI tools, AI tool recommendations

Trending

Category B · Step-by-Step Tutorials (30%, Demonstrating Tool Value)

No.

Topic

Tutorial Content

Image Structure

WorkBuddy PPT Tutorial | Create Professional Presentations in 5 Steps

Operation steps + effect showcase

Step screenshots + final result

Write Weekly Reports with AI | Get Boss-Satisfying Results in 10 Minutes

Prompts + effect comparison

Prompt templates + finished product

Learn AI Writing from Zero Basics | Comprehensive Step-by-Step Guide

Full process from registration to output

6-step flowchart + case studies

AI Poster Design Tutorial | Create Professional Graphics Without Photoshop

Tool operation + template recommendations

Operation screenshots + finished designs

Auto-Generate Meeting Minutes with AI | Done in 5 Seconds

Tool comparison + operation steps

Recording → Minutes comparison chart

Category C · Pain Points & Growth (20%, Resonance + Saves)

No.

Topic

Analysis Dimensions

Guide for Professionals Over 30 Learning AI | It's Really Late If You Don't Start Now

Anxiety + solutions + roadmap

The Correct Order for Ordinary People to Learn AI | Don't Start with Coding, Try This First

Learning path + priorities

5 Ways to Make Money with AI Side Hustles | Personally Tested & Effective

Side hustle + monetization + practical ops

Why Can't You Use AI Tools Well Even After Learning Many?

Misconceptions + methodology

Category D · Collections & Resources (10%, High Save Rate)

No.

Topic

Content

[Save This] Massive Collection of AI Learning Resources | Tools + Tutorials + Websites

Resource collection + categorized organization

AI Tool Comparison Table by Job Role | Just Match Your Position

Role matching + tool recommendations

4.4 Title Templates

Xiaohongshu Title Formula: First 8 characters contain core keyword + number + pain point/result

Template

Formula

Example

Keyword + Number + Result

AI Tool Recommendations | 10 Must-Learn Tools in 2026

Pain Point + Solution

Don't Want to Be Replaced by AI? Learn These 3 Tools First

Identity + Recommendation + Number

Curated by AI Learning Coach | 5 Must-Learn AI Tools

"Comprehensive Tutorial" + Topic

Comprehensive Tutorial | WorkBuddy from Beginner to Expert

Comparison + Suspense

Before vs. After Using AI | The Gap Is Huge

4.5 Cover Copy

Cover Type

Copy Elements

Design Key Points

List Type

"TOP10" + Tool Names

Place numbers + tool logos arranged on cover

Tutorial Type

"Comprehensive Tutorial" + Tool Name

Operation screenshot + step labels

Pain Point Type

Pain Point Word + "Savior"

Large text + high-contrast colors

Collection Type

"Save This" + Topic

Sense of value + category tags

4.6 Tagging Strategy

Add 10 to 15 tags per note, divided into three layers:

Broad Tags (2-3): #AI #ArtificialIntelligence #AITools
Mid-Tier Tags (3-5): #WorkBuddy #AILearning #OfficeEfficiency #AITutorial #SelfImprovement
Long-Tail Tags (5-7): #LearnAIFromZero #AIToolRecommendations #EssentialForWorkers #EssentialForStudents #AIWriting #AIDesign #ProductivityTools

4.7 Posting Times & Frequency

Time Slot

Recommended Time

Suitable Content

Morning Rush

7:00-9:00

Tool lists/news updates

Lunch Break

12:00-13:30

Step-by-step tutorials

Evening

18:00-20:00

Pain points & growth topics

Late Night

21:00-23:00

Collections & resources

Frequency: 3 notes per week (one each on Monday, Wednesday, Friday), maintaining consistency for 30 days without breaks.
First 7 Days: Only post tool lists + tutorial content, without mentioning specific product names. Purely for account nurturing + keyword testing.
From Day 8 Onwards: Begin integrating WorkBuddy, approaching it from the angle of "AI learning tool recommendations."

4.8 Comment Section Operations & Compliant Traffic Diversion

In 2026, Xiaohongshu platform rules are extremely strict. Directly leaving WeChat IDs or website links in the body text = traffic throttling + account ban.

Diversion Stage

Specific Actions

Compliance Key Points

Within Note

End with: "I've compiled a complete AI toolkit. DM 'Toolkit' if you need it."

Do not include WeChat ID/links

Within DMs

After user DMs, provide value first (tool list/tutorial PDF), then guide them to join the fan group.

Fan groups are an official feature, compliant.

Within Fan Group

Post website links and exclusive benefits inside the group.

Links can be posted within groups.

Enterprise WeChat Reception

Guide users to add Enterprise WeChat for long-term operations.

Shift from "traffic harvesting" to "user value."

4.9 Conversion Path

Xiaohongshu Note → User sees via search/recommendation → Clicks to read

→ End of note "DM to receive toolkit" → User sends DM

→ Reply with materials + guide to fan group

→ Post website link in fan group

→ Wujie Frontier Website

4.10 Data Review Metrics

Metric

Pass Line

Excellent Line

Meaning

Click-Through Rate (CTR)

>5%

>12%

Attractiveness of cover and title

Search Exposure Share

>20%

>40%

Is SEO working?

Engagement Rate

>3%

>8%

Content quality

DM Rate

>1%

>3%

Effectiveness of material hooks

Follower Growth Rate

>0.3%

>1%

Account weight

Note Indexing Rate

>80%

>95%

Is content indexed by search?

Official Site UV (Core KPI)

Daily Increase > 5

Daily Increase > 30

Ultimate Goal: Website traffic

──────────────────────────────────────────────────

V. Bilibili Content Operations Strategy (AI Learning Edition)

5.1 Account Positioning

Persona: AI Learning Coach / Tech UP-master. Bilibili users prefer authentic, deep, and valuable content; they dislike overly promotional content.
Name Suggestions: Can use the brand name "Wujie Frontier" or a personal IP name. E.g., "Wujie Frontier," "AI Learning Lab."
Bio Template: Focused on AI learning for ordinary people / Making tutorials, testing tools, explaining methods / Weekly deep-dive AI tutorials / Complete toolkit available on official website.
Homepage Setup: Personal space bio can include website links (allowed by Bilibili); video descriptions can also include website links.

5.2 Content Formats

Core Format: Landscape long-form videos, 5 to 15 minutes.
Filming Method: Screen recording operations + real voiceover (or on-camera presence). Enhance watchability with step demonstrations and effect comparisons.
Visual Requirements: 1080P landscape 16:9, high information density, key operation points annotated with subtitles.
Supplementary Content: Simultaneously publish column articles (Bilibili Columns) to capture both Bilibili + Baidu dual search entry points.

5.3 Topic Directions & Specific Topic List

Category A · Systematic Tutorials (40%, Core Bilibili Content)

No.

Topic

Duration

Core Content

Complete WorkBuddy Tutorial: From Beginner to Expert (2026 Latest Version)

15 mins

Full feature breakdown + practical cases + advanced tips

Learn AI Tools from Zero Basics: 5 Tools to Double Your Efficiency

12 mins

Tool introduction + operation demo + scenario application

Complete Guide to AI Writing: Prompts + Tools + Practice

12 mins

Prompt methodology + tool comparison + cases

AI Learning Roadmap for Ordinary People (2026 Edition)

10 mins

Learning path + stage goals + resource recommendations

AI Office Practice: Master PPT/Excel/Word All-in-One

15 mins

Three major office scenarios + tools + operation steps

Category B · Tool Reviews (30%, Long-Tail Search Traffic)

No.

Topic

Duration

Core Content

Testing 10 AI Assistants: Where Does WorkBuddy Rank?

10 mins

Multi-dimensional review + ranking + scenario recommendations

Horizontal Review of Free AI Tools: These 5 Are Truly Competitive

12 mins

Real-world testing of free tools + pros/cons + recommendations

In-Depth Review of WorkBuddy: Is It Worth Using?

8 mins

Features + pricing + target audience + pros/cons

Major Horizontal Review of AI Image Generation Tools: Midjourney vs. Kling vs. Others

12 min

Multi-tool comparison + demo + recommendations

2026 Annual AI Tools Review: The 10 Most Worth Learning

10 min

Annual review + categorized recommendations + learning advice

Category C · Growth Methodology (20%, Trust + Brand)

No.

Topic

Duration

Core Content

After 6 Months of Learning AI, Here Are My 10 Honest Takeaways

10 min

Learning insights + pitfall avoidance guide + real experiences

Why I Recommend Every Regular Person Learn AI?

8 min

Trend analysis + personal growth + action advice

Is It Too Late to Switch Careers into AI at Age 30?

8 min

Career planning + path suggestions + real cases

How I Use AI Tools to Gain an Extra 2 Hours Every Day

10 min

Efficiency methods + tool combinations + daily routine

Category D · Hot Topic Interpretation (10%, Virality)

No.

Topic

Duration

Core Content

5 Major AI Events in 2026: What Do They Mean for Ordinary People?

8 min

Hot topic interpretation + impact on ordinary people + action advice

5.4 Title Templates

Bilibili Title Formula: Keyword + Depth/Number + Conclusion/Suspense

Template

Formula

Example

"Complete Tutorial" + Tool + Conclusion

Complete Tutorial | WorkBuddy from Beginner to Master 2026

"In-depth Review" + Tool

In-depth Review | Is WorkBuddy Worth Using?

Comparison + Suspense

WorkBuddy vs Doubao | How Big is the Gap?

Number + Review

10 Must-Learn AI Tools | Explained Clearly in One Go

"Side-by-Side Review" + Number + Conclusion

Side-by-Side Review of 5 AI Tools | This is My Top Recommendation

5.5 Body Structure (5 to 15 Minute Version)

0-15s: Opening Hook (Conclusion/Pain Point/Demo/Comparison)

15-60s: Background Introduction + One Sentence Clarifying the Episode's Theme

60s-Main Body:

Tutorial Type: Tool Intro → Basic Operations → Advanced Tips → Real-world Cases → Summary

Review Type: Review Criteria → Individual Tests → Comparative Ranking → Scenario Recommendations

Methodology Type: Core Viewpoint → Supporting Arguments → Case Evidence → Action Advice

Last 30s: Summary + CTA

"The complete toolkit and learning materials are in the official website link in the description."

"If you also want to learn AI, check out Wujie Frontier."

5.6 Conversion Path

Bilibili Video → User sees via search/recommendation → Clicks to watch

→ Website link in video description → Direct jump to website

→ Pinned comment guides "Complete toolkit in description" → Description link

→ Wujie Frontier website

Bilibili Column → User sees via search → Reads article

→ Website link at end of text → Jump to website

Bilibili has the shortest traffic funnel among the three platforms: both video descriptions and columns allow direct external links.

5.7 Data Review Metrics

Metric

Passing Line

Excellent Line

Meaning

Completion Rate

>25%

>40%

Content Quality (Benchmark for Bilibili long videos)

Engagement Rate

>5%

>12%

Ability to Spark Discussion

Triple Combo Rate (Like/Coin/Favorite)

>3%

>8%

Content Approval Level

Follow Conversion Rate

>1%

>3%

Appeal of Series Content

Description Click-through Rate

>2%

>5%

Effectiveness of CTA Guidance

Search Traffic Share

>20%

>40%

Long-tail Search Traffic

Website UV (Core KPI)

Daily Increase > 5

Daily Increase > 20

Ultimate Goal: Website Traffic

──────────────────────────────────────────────────

VI. Cross-Platform Content Matrix and Repurposing Strategy

6.1 One Fish, Many Meals: Create Once, Distribute Across Three Platforms

Using a 10-minute in-depth Bilibili tutorial as the "Mother Content," split it into multiple sub-contents:

Bilibili 10-min In-depth Tutorial (Mother Content)

├→ Douyin Account A: Split into 3-4 short videos of 60 seconds each (one per tip)

├→ Xiaohongshu: Split into 2-3 image-text notes (tool list + step-by-step tutorial)

├→ Bilibili Column: Convert to text-based article

└→ Comment Section Material: Extract user questions → Topics for next episode

6.2 Douyin Matrix Expansion Strategy

Architecture Upgrade Path: A₁ → A₁+A₂ → A₁…Aₙ (Matrix) + B (Blue V) ×1

Matrix Expansion Directions (by traffic source):

AI Learning Coach Account (Trust traffic, live person on camera)
Vertical Tool Account (WorkBuddy tips/AI writing/AI image generation, precise traffic)
Employee/Founder Account (Personalized branding)
Optional: Workplace AI Account, Student AI Account, Side Hustle AI Account

Unified Command Rules (Key to Matrix):

All A accounts use consistent scripts pointing to the same B (Blue V) → Same official website link
B account only handles reception and endorsement, does not distribute external links separately
Avoid posting identical content across multiple accounts to prevent being flagged as reposting/marketing (stagger topics, vary angles)

6.3 Content Scheduling Matrix

Using "One Theme Per Week" as the unit, unify themes for distribution across three platforms:

Week

Theme

Douyin Account A

Xiaohongshu

Bilibili

Tool Basics

3 short videos on tool basics

2 tool list notes

0 (Do not post during account nurturing period)

Deep Dive into WorkBuddy

4 WorkBuddy teaching videos

2 WorkBuddy tutorial notes

1 full tutorial video (15 mins)

Pain Point Resonance

3 pain point short videos

1 growth methodology note

1 learning roadmap video (10 mins)

Tool Comparison

3 review short videos

1 comparison note

1 tool comparison video (12 mins)

Real-world Cases

2 case study short videos

1 case study note

1 practical tutorial video (10 mins)

Learning Path

2 pathway short videos

1 pathway note

1 systematic tutorial video (12 mins)

Douyin Account A posts 3-5 times weekly, Xiaohongshu posts 3 times weekly, Bilibili posts once weekly. Weekly content themes are unified to facilitate cross-platform traffic diversion.

6.4 Unified Keyword Layout

Layout around the same set of core keywords across all three platforms to achieve comprehensive search coverage:

Keyword Tier

Keyword Examples

Platforms

Brand Terms

Wujie Frontier, WorkBuddy

All platforms (Brand terms must dominate search results)

Industry Terms

AI tools, AI learning, Artificial Intelligence

All platforms

Tool Terms

WorkBuddy, AI writing, AI image generation

All platforms

Scenario Terms

Office AI, College student AI, Learn AI from scratch

All platforms

Long-tail Terms

How to use WorkBuddy, AI tool recommendations, AI writing tools

Xiaohongshu + Douyin

Comparison Terms

WorkBuddy vs XX, Best AI tools

All platforms

Tutorial Terms

AI tool tutorials, WorkBuddy usage guide

Bilibili + Xiaohongshu

──────────────────────────────────────────────────

VII. Phased Implementation SOP

Phase One (Weeks 1-4): Early Stage · Solo/Two-Account Start

Goal: Build the minimal closed loop of "A attracts, B captures," validate content direction and jump efficiency

Week

Douyin

Xiaohongshu

Bilibili

Register Blue V + submit certification; Launch 1 Account A (AI Learning Coach), define persona; Post 3 tool teaching videos to test data

Blue V certification approved + attach website link; Account A posts 4 videos (2 teaching + 2 pain points); Test jump scripts

Post 2 tutorial notes; Optimize covers

Prepare script for 1st video

Account A posts 5 videos (2 teaching + 2 reviews + 1 pain point); Review which content types perform best; Optimize comment section operations

Post 2 growth notes; Observe search traffic

Record + edit 1st video

Monthly review: Select optimal content direction; Evaluate A→B jump rate and website click-through rate; Decide whether to launch A₂ account

Monthly review: Note types with highest search traffic

Publish 1st video; Observe data

Key Actions for Phase One:

1. Register Blue V (Business license + ICP filing must match), configure official website link component

2. Launch 1 personal account (A₁ AI Learning Coach), define persona and content direction

3. Edit 5 WorkBuddy teaching short videos, test completion rate of first 3 seconds

4. Unify scripts to guide users to the Blue V official website link

Acceptance Criteria for Phase One:

Account A has published 12+ videos cumulatively
Highest views for a single video > 5000
Official website link on Blue V homepage receives clicks
Identify 1-2 high-potential content directions
Daily average website UV > 10

Phase Two (Weeks 5-8): Validation + Paid Promotion Scaling

Goal: Use paid promotion to validate viral models, optimize conversion funnel

Week

Douyin

Xiaohongshu

Bilibili

Invest small budget in DOU+ for winning content (100-300 RMB/video); Launch A₂ account (Vertical Tool Account)

3 posts weekly, focus on optimizing keywords

Publish 2nd video

Check Blue V backend "Website Link Clicks" data to reverse-engineer content direction; Account A updates daily

Establish DM conversion flow; Optimize DM scripts

Prepare script for 3rd video + record

Host 1 livestream teaching AI tools, verbally guide to website link; Unify matrix account scripts

Increase search traffic share to 30%+

Publish 3rd video (Systematic tutorial type)

Paid promotion review: ROI, cost per click, content direction; Determine matrix expansion pace

Monthly review

Key Actions for Phase Two:

1. Invest small budget in DOU+ for winning content

2. Check Blue V backend "Website Link Clicks" data to reverse-engineer content direction

3. Host 1 livestream teaching AI tools, verbally guide to website link

4. Launch A₂ account (Vertical Tool Account)

Acceptance Criteria for Phase Two:

Positive ROI on paid promotion (website click cost is acceptable)
Stable operation of 2 A accounts + 1 B account
Combined daily average website visits across three platforms > 50 UV

Phase Three (Week 9+): Matrix Scaling

Goal: Scale replication, establish stable traffic pipeline

Key Actions:

1. Once data stabilizes, expand the matrix with A-tier accounts (e.g., Workplace AI, Student AI) according to the matrix strategy.

2. Run paid traffic on Blue V enterprise accounts (via Ocean Engine Qianchuan), optimizing for clicks to the official website to scale up.

3. Establish a content production SOP to achieve batch output.

4. Implement cross-platform synergy, forming a traffic matrix across Douyin, Xiaohongshu (RED), and Bilibili.

Matrix Expansion Cadence:

Weeks 9-12: Expand A-tier accounts to 3-4 (add 1-2 vertical tool-focused accounts).
Weeks 13-16: Expand A-tier accounts to 5-6; begin launching employee accounts and vertical accounts for workplace/student segments.
Week 17+: Continuously optimize based on data to stabilize matrix scale.

──────────────────────────────────────────────────

VIII. Compliance and Feasibility Boundaries (Verified · Non-Negotiable)

⚠️ The following are actual platform rules; plan execution must strictly adhere to these:

Item

Status

Description

Blue V Homepage "Official Website Link" Component

✅ The only stable compliant external link entry point

Requires domain ICP filing and entity matching business license; review takes 7-10 days

Personal Account Organic Traffic

✅ Strong

Core value of A-tier accounts

Little Windmill / Lead Generation Components

❌ Creation stopped on Sep 19, 2025; fully invalid by Oct 8

Previous methods are obsolete; do not use

Posting URLs or WeChat IDs in DMs/Comments

❌ Third-party inducement controls upgraded on Nov 14, 2025; automatic interception

Only use the homepage official website link

Paid Traffic to Supplement Organic Flow

✅ Feasible and necessary

DOU+ / Ocean Engine Qianchuan Enterprise Accounts

Compliance Exemption Benefit: Official policy allows "sending the official website for product quality inspection/certification inquiries"—Wujie Frontier is positioned as an AI learning platform for ordinary people, providing AI tool tutorials and learning resources, which falls under compliant scenarios.

──────────────────────────────────────────────────

IX. Tools, Resources, and Risk Mitigation

9.1 Recommended Tools and Resources

Purpose

Tool

Description

Keyword Research

5118, Baidu Index, Ocean Engine Data

Identify question-based keywords frequently searched by users

Graphic Design

Canva, Chuangketie

Create cover templates for Xiaohongshu and Douyin

Video Editing

CapCut (Mobile + Desktop versions)

Edit Douyin short videos; dynamic text for keywords

Screen Recording

OBS Studio, System Built-in Recorder

Record AI tool operation demos

Data Analysis

Platform Creator Dashboards, Xinbang

Monitor views/search/engagement data

Live Streaming Tool

Douyin Live Companion

For Blue V live streaming

Paid Traffic Tool

Ocean Engine Qianchuan, DOU+

Scale Douyin traffic via paid ads

Prompt Engineering Tool

WorkBuddy Built-in

Demonstrate directly within the product

9.2 Key Risks and Responses

Risk

Impact

Response

ICP Filing Entity Mismatch with Blue V

Official website link cannot be attached; "A-acquire, B-capture" model fails

Verify entity consistency in advance; Alternative: Guide DMs from B-account, send materials + website via DM

Account Flagged as Marketing Account

Traffic throttling, account ban

Nurture account with pure tool tutorial content for first 7 days; integrate products naturally, avoid hard selling

Xiaohongshu Traffic Diversion Ban

Traffic throttling, account ban

Do not place any links/WeChat IDs in post body; use compliant path: "DM → Fan Group"

Low Douyin Completion Rate

Content fails to enter recommendation pool

First 3 seconds must grab attention; keep videos under 60 seconds

Low A→B Conversion Rate

Wasted traffic

Optimize guidance scripts; multi-touchpoint guidance (bio + voiceover + pinned comment)

Matrix Accounts Flagged as Marketing Matrix

Batch traffic throttling

Differentiate content across accounts (topic angles, personas, formats); never post identical content

Low Bilibili Search Ranking

Search traffic stagnates

Include search terms like "tutorial," "review," "guide," "beginner" in titles; publish columns simultaneously

Content Homogenization

User disinterest

Each piece of content must feature unique operational tips/comparisons/angles; no reposting

Insufficient Manpower

Update gaps

Prioritize updates for Douyin A-account (lowest cost); one Bilibili update per week is sufficient

Paid Traffic ROI Below Target

Budget waste

Test with small budget first; scale only after validating positive results; cut losses quickly on poor content directions

──────────────────────────────────────────────────

This plan is formulated based on public operational data from the three major platforms as of August 2026, official Douyin Enterprise Account rules, and the "A-acquire, B-capture" operational model. Once WorkBuddy's specific features and data resources are confirmed, all topics, titles, and copy can be replaced with real-world scenarios for a second round of refinement.

──────────────────────────────────────────────────

X. In-depth Analysis of Platform Recommendation Mechanisms and Account Launch Rules

💡 Disclaimer: All recommendation mechanisms, weight metrics, traffic pool data, and threshold parameters in this chapter are compiled from publicly available industry sources during 2024-2025 (including public analysis articles from SMZDM, Zhuobo Tech, Qinggua Media, Everyone is a Product Manager, Jizhile Data, etc.). Platform algorithms iterate continuously; specific weights and thresholds must follow the latest official platform rules. This chapter does not claim access to any non-public internal platform data. All cited sources are labeled for operational reference, not as absolute standards.

10.1 Douyin Recommendation Mechanism and Account Launch Rules

10.1.1 Traffic Pool Tiers and Progressive Unlocking Mechanism

Douyin employs a six-tier progressive racing distribution model. New videos start at the lowest tier, advancing based on performance metrics:

Traffic Pool Tier

Exposure Volume

Unlock Condition

Demotion Mechanism

L1 Cold Start Pool

100-300 precise exposures

Entered immediately upon posting; system matches initial audience based on account tags

If metrics fail, stays in this tier; no further amplification

L2 Basic Recommendation Pool

1,000-5,000

Core metrics met in L1 stage (completion rate + engagement rate qualified)

Can demote back to L1

L3 Pending Recommendation Pool

5,000-30,000

L2 stage data consistently outperforms competitors in same pool

Can demote back to L2 if data drops

L4 Pending Viral Pool

30k-500k

Completion rate + engagement rate + share rate all met in L3 stage

Can demote back to L3

L5 Hot Pool

500k-5M

Comprehensive data enters top tier in L4 stage

Can demote back to L4

L6 Global Viral Pool

5M+

Data continues to lead in L5 stage, triggering global distribution

Rarely demotes

*💡 Key Change: In the July 2026 algorithm update, the cold start pool exposure was reduced from the original 500-1000 to 100-300 precise exposures, emphasizing "small but refined" initial audience matching. The evaluation mechanism shifted from "24-hour short-term racing" to "7-day long-term value assessment"—even if data rebounds 48 hours after posting, the video can be reactivated into higher traffic pools.

Source: Zhuobo Tech "2026 Douyin Algorithm: 6-Tier Traffic Pool Progressive Unlocking Mechanism"*

10.1.2 Core Weight Metric Ranking

In 2026, Douyin's weighting metrics underwent significant changes—completion rate is no longer the primary weight; save rate has taken its place. Current ranking (highest to lowest):

No.

Metric

Description

Save Rate

Highest weight—reflects long-term content value; users deem it worth revisiting repeatedly

Revisit Rate

Proportion of users reopening the same video; reflects content attractiveness

Iron Fan Engagement Rate

Interaction behavior of followed fans; weighted higher than general user interactions

5-Second Completion Rate

Retention capability in the first 5 seconds; determines whether more exposure is granted

Overall Completion Rate

Proportion of full views

Comment Quality

Comment length, interaction depth (not just comment count)

Like Rate

Basic interaction metric

Share Rate

Lowest weight

*💡 Key Insight: Content design should shift from "pursuing completion" to "pursuing saves and revisits"—meaning content must have enough value density that users want to "save it for later." AI tool tutorials, learning paths, and tool lists are inherently suited to this mechanism as "save-worthy content."

Source: SMZDM "2026 Douyin Algorithm 5 Major Changes: Completion Rate Dethroned, Save Rate Reigns Supreme"*

10.1.3 Distribution of Four Major Traffic Entrances

In 2026, Douyin traffic has fragmented from a single recommendation feed into four major entrances:

Traffic Entrance

Characteristics

Recommendation Feed

~45%

Algorithm-driven push; primarily competes on content data

~25%

Keyword-driven; long-tail traffic; rising SEO value

Mall Product Cards

~15%

E-commerce oriented; product cards drive direct conversions

Followings/Private Domain

~15%

Fan reach; high weight for iron fan engagement

*💡 Key Insight: With search traffic accounting for 25%, keyword placement in titles and descriptions directly impacts search discoverability. Relying solely on the recommendation feed for virality is becoming harder; simultaneous search SEO layout is required. AI tool content has strong search intent, making it suitable for capturing search traffic.

Source: Account Data "2026 Douyin Algorithm 5 Major Changes"*

10.1.4 Account Rating System (SABCDE)

Douyin implements an SABCDE five-level rating system for accounts, which directly affects initial traffic pool size and recommendation caps:

Rating

Characteristics

Traffic Bias

S-Level

High effective follower count, high interaction quality, high verticality

Maximized initial pool size; priority entry into higher-tier traffic pools

A-Level

Stable output, good interaction

Larger initial pool; normal progressive unlocking

B-Level

Basic compliance

Standard initial pool

C-Level

Low interaction, insufficient content verticality

Smaller initial pool; requires better data to upgrade

D/E-Level

Violations, low quality, reposting

Throttled or not recommended

*💡 Key Metric: Ratings hinge on "effective follower count"—i.e., genuinely active fans, not zombie followers. Buying followers will directly lower the rating.

Source: Account Data "2026 Douyin Algorithm 5 Major Changes"*

10.1.5 Cold Start During Launch Period and 7-Day Racing Mechanism

Cold Start Three Stages (7-Day Cycle):

Stage

Timeframe

Core Task

Data Goal

Stage 1: Cold Start

Days 1-2 post-launch

System matches initial precise audience; tests content quality

Completion rate qualified; 5-second retention >50%

Stage 2: Secondary Activation

Days 3-4

If data rebounds, system re-amplifies traffic

Improved save rate and comment quality

Stage 3: Long-Tail Explosion

Days 5-7

Accumulated data enters long-term assessment

Revisit rate and iron fan engagement qualified

*💡 Key Change: The 7-day long-term racing means you don't need to aim for "instant viral success upon posting." If a video's data rebounds on Day 3 due to a comment sparking discussion, the system can still reactivate it and push it into higher traffic pools. This requires content to possess "long-tail discussion value" rather than just chasing momentary spikes. AI tool tutorial content naturally has long-tail search value.

Source: SMZDM "Douyin 7-Day Assessment Cycle: Three Stages"*

10.1.6 Weighting Logic for the First 5/10/20 Videos of a New Account

The content publishing rhythm during the cold-start phase of a new account directly impacts the establishment of account tags and initial weighting:

First 5 videos: The system focuses on collecting account positioning tags, which determine subsequent traffic distribution directions. It is essential to maintain high verticality (same field, same style). It is recommended to select high-heat topics within that field, but do not chase hot trends (new accounts lack a weight base; chasing trends makes it easy to be suppressed by large accounts in the same field). The goal is not view count, but "letting the system accurately identify your account tags."

First 10 videos: Account tags are preliminarily established, and the system begins to stably match precise audiences. You can try slight differentiation (e.g., different content formats within the same field). Observe data for 24-48 hours after each video is published; adjust directions promptly if the completion rate is below 30%. Interaction targets: Like rate > 3%, Comment rate > 0.5%.

First 20 videos: Account tags are solidified, and the system begins to provide a more stable initial traffic pool. You can start chasing hot trends (with an existing weight base, trending content becomes competitive). At this stage, if a specific video performs exceptionally well, it can trigger an overall increase in account weighting. It is recommended to conduct a comprehensive data review after 20 videos to identify 1-2 optimal content directions for concentrated effort.

💡 Key Principle: Do not change fields within the first 20 videos. Once you switch tracks midway, the accumulated tag weights are reset to zero, equivalent to restarting the cold-start process. For AI learning accounts, all of the first 20 videos must revolve around "AI Tools/AI Learning."

10.1.7 AI Content Labeling Management

In 2026, Douyin implements mandatory labeling management for AI-generated content:

Videos/images/copy generated using AI must have the "AI Generated" label checked upon publication
If unlabeled content is identified as AI-generated by the system, traffic will be restricted by 50%-80%
Labeling AI does not restrict traffic, but the system will display an "AI Generated" tag when recommending the content
"Pseudo-original" content consisting of pure reposts + AI rewriting will also be identified and downweighted

*💡 Implications for WorkBuddy Content: If AI is used to assist in generating scripts or visuals, ensure proactive labeling. AI labeling does not affect normal traffic distribution, but concealing labels will result in severe traffic restrictions. Screen recording operational content (real operation demonstrations) does not qualify as AI-generated content and requires no labeling.

Source: Account Data "5 Major Changes in Douyin Algorithm 2026"*

10.1.8 Practical Implications for WorkBuddy Traffic Acquisition

Based on the above mechanisms, WorkBuddy's operations on the Douyin platform should follow these principles:

1. Design content with "Save Value" at the core: Each video must include practical information worth saving (tool tutorials, step-by-step guides, tool lists, learning paths), rather than purely emotional content

2. Lock in the "AI Tools/AI Learning" vertical tag with the first 5 videos: Topics, titles, and tags should all revolve around the same tag matrix; do not cross boundaries

3. Layout search keywords in titles: Utilize the 25% search traffic entry point by embedding high-frequency search terms such as "WorkBuddy," "AI tools," "AI learning," and "Learn AI from scratch" into titles

4. Maintain active comment section management during the 7-day competition cycle: Actively reply to comments and guide discussions within 7 days of publication to improve comment quality and interaction with core fans

5. Proactively label AI content: If AI is used to generate visuals or scripts, check the label box upon publication

6. Prioritize core fan operations over general traffic: The interaction weight of the 15% follower/private domain traffic is higher than that of ordinary users; guide follows and engage deeply in the comment section

10.2 Xiaohongshu (RED) Traffic Distribution Mechanism and Account Launch Rules

10.2.1 CES Scoring Formula and Weights

Xiaohongshu uses the CES (Community Engagement Score) model to measure note quality, determining whether to push notes into higher traffic pools:

*💡 CES Scoring Formula:

CES = Likes × 1 + Saves × 1 + Comments × 4 + Shares × 4 + Follows × 8*

Interaction Behavior

Weight Coefficient

Interpretation

Follows

× 8

Highest weight—if a note generates new followers, it indicates extremely high content value

Comments

× 4

High weight—comments represent deep engagement, offering greater value than passive likes

× 4

High weight—shares indicate the content has propagation value

Saves

× 1

Base weight—saves reflect long-term value

Likes

× 1

Base weight—the lowest barrier to interaction

*💡 Key Insight: The weight of comments and shares is 4 times that of likes; guiding comments yields a much higher ROI than guiding likes. The weight of follows is 8 times that of likes; notes that generate follows receive the highest traffic boost. Content design should include "comment hooks"—questions at the end, voting interactions, or controversial viewpoints to stimulate discussion. AI tool tutorial content naturally has high save rates.

Source: Qinggua Media "Analysis of Xiaohongshu CES Scoring Formula and Traffic Pool Mechanism"*

10.2.2 Distribution of Four Major Traffic Entry Points

Traffic Entry Point

Proportion

Characteristics

Discovery Page (Recommendations)

~60%

Algorithm-driven active recommendation, powered by CES scores

Search Page

30%-70%

Long-tail traffic, keyword-driven; high-quality notes can sustain traffic for months

Following Page/Private Domain

Increases with follower growth

Reaches followers, high interaction rate

Local/City Page

Increases with location tags

Suitable for local merchants and regional content

*💡 Key Insight: Search traffic accounts for up to 30%-70% (fluctuating with note quality) and represents long-tail traffic—a note with good search ranking can gain stable exposure for consecutive months. Therefore, keyword layout is the core lever of Xiaohongshu operations, being more sustainable than chasing instantaneous trends. AI tool content has strong search intent, making it a golden track on Xiaohongshu.

Source: Qinggua Media "Distribution of Xiaohongshu's 4 Major Traffic Entry Points"*

10.2.3 Tiered Traffic Pools and Real-Time Competition

After publication, Xiaohongshu notes enter tiered traffic pools, undergoing real-time competition screening to advance through levels:

Tiered Traffic Pools:

Pool Level

Exposure Volume

Entry Condition

Initial Pool

100-500

Entered immediately upon publication

Level 2 Pool

500-2,000

Initial pool CES score meets standards

Level 3 Pool

2,000-10,000

Consistently outperforms similar notes in Level 2 Pool

Viral Pool

10,000+

Data enters top tier in Level 3 Pool, triggering site-wide recommendations

Three Stages of Real-Time Competition:

Stage

Time

Mechanism

Initial Screening

5 minutes post-publication

System pushes to a small range of precise users to test initial interaction rates

Same-Pool Elimination

30 minutes post-publication

Competes against similar notes published in the same time slot; those with higher CES advance

Tier Determination

60 minutes post-publication

Final traffic tier is determined; subsequently enters long-tail recommendation

*💡 Key Insight: The first hour after publication is the golden window. If the initial interaction rate during the 5-minute screening falls below the average of the same pool, the note may get stuck in the initial pool without further expansion. Therefore: Publish during peak activity hours for target users (lunch break 12:00-13:00, evening 20:00-22:00); actively interact in the comment section within 30 minutes of publication to boost comment metrics; covers and titles must drive initial clicks and interactions within the first 5 minutes.

Source: Qinggua Media "Xiaohongshu Real-Time Competition and Tiered Traffic Pool Mechanism"*

10.2.4 New Account Observation Period Mechanism

Xiaohongshu sets an observation period (approx. 7-14 days) for newly registered accounts:

Notes published during the observation period receive lower initial exposure compared to older accounts (approx. 50-200 exposures)
The system primarily audits content quality and account authenticity (determining if it is a marketing/repost account)
It is recommended to publish 3-5 high-quality original notes during the observation period to establish content tags
After the observation period ends, if content quality meets standards, the account enters the normal traffic pool
Do not frequently publish low-quality content during the observation period, as this may lead to classification as a marketing account and long-term traffic restriction

Account Launch Recommendations:

1. Spend 3-5 days nurturing the account after registration (browse, like, and save content in the same field to build interest tags)

2. Select a topic with high search volume for the first note, but ensure the content is deeply original

3. Keep all of the first 5 notes focused on the same field; do not cross tracks

4. Stay online for 1 hour after publishing each note and reply to comments promptly

10.2.5 Title/Cover/Keywords and Traffic Weighting

Element

Traffic Weight

Best Practices

Title

★★★★★

Include core search terms + emotional hooks, under 20 characters, with keywords in the first 10 characters

Cover

★★★★★

3:4 vertical image, large and clear text, theme instantly recognizable, strong color contrast

Keywords

★★★★☆

Layout keywords in the title, first paragraph of body, last paragraph of body, and tag sections

Note Type

★★★☆☆

Image-text notes have higher search weight than video notes; compilation/practical tips notes have higher save rates

Body Length

★★★☆☆

800-1500 words is the optimal range; too short lacks information density, too long reduces read-through rate

Interaction Guidance

★★★☆☆

Set questions/votes/comment section hooks at the end to boost comment weight

Keyword Layout Rules:

Core word: 1 (e.g., "AI tools"), appearing in the title, first paragraph, last paragraph, and tags
Long-tail words: 3-5 (e.g., "WorkBuddy tutorial," "Learn AI from scratch"), naturally integrated into the body
Scenario words: 2-3 (e.g., "Office efficiency," "Essential for college students"), matching user search intent

💡 Source: Qinggua Media "Three Major Upgrades in Xiaohongshu Algorithm: Keyword Quality Scoring"

10.2.6 Three Major Algorithm Upgrades

There are three significant upgrades to the Xiaohongshu algorithm in 2025-2026:

1. Keyword Quality Scoring: The system no longer just checks if keywords appear but evaluates the relevance of keywords to content and the naturalness of keyword density. Stuffing keywords will result in downweighting.

2. Enhanced Interaction Modeling: The system distinguishes between "genuine interactions" and "fake/inflated interactions"—concentrated interactions from the same IP address in a short time are identified and removed. Comment quality is assessed by word count and depth; short comments (e.g., "Good," "Like") have extremely low weight.

3. AI Content Detection: The system can detect AI-generated body text and images. Purely AI-generated notes without labeling will be downweighted. It is recommended to perform deep manual rewriting after AI-assisted generation.

💡 Source: Qinggua Media "Three Major Upgrades in Xiaohongshu Algorithm"

10.2.7 Practical Implications for WorkBuddy Traffic Acquisition

1. CES-Oriented Content Design: Each note includes a "comment section engagement hook" (e.g., "Which AI tool do you most want to learn? Tell me in the comments"), leveraging the ×4 weight for comments

2. Keyword Matrix Layout: Each note centers on 1 core keyword + 3-5 long-tail keywords + 2-3 scenario-based keywords, naturally integrated into the body text

3. Image-Text Prioritized Over Video: Image-text notes have higher search weighting and lower production costs, making them suitable for AI tool tutorial content

4. Golden Window: First Hour After Posting: Choose lunch breaks or evenings for posting, and immediately reply to every comment after publishing

5. Account Nurturing Strategy During Observation Period: The first 5 posts should all focus vertically on "AI Tools/AI Learning," avoiding cross-domain trending topics

6. Deep Rewriting of AI Content: If using AI to generate drafts, they must be manually rewritten to achieve a natural tone to avoid system demotion

10.3 Bilibili Traffic Distribution Mechanism and Account Launch Rules

10.3.1 AI-Driven Content Distribution (2026 New Regulations)

*💡 On April 1, 2026, Bilibili officially retired the "Guess You Like" recommendation module and fully launched an AI-driven content distribution system. This is the most significant underlying overhaul of Bilibili's recommendation mechanism in recent years:

Old Mechanism: "Guess You Like" relied on collaborative filtering, recommending similar content based on user historical behavior
New Mechanism: AI-driven distribution, based on content understanding + full-chain user behavior prediction, achieving more precise content matching
Impact: The weight of content quality (rather than account follower count) in recommendations has further increased. New accounts with high-quality content now face a fairer competitive environment.*

Source: Smzdm.com "Bilibili Retires 'Guess You Like' and Launches AI-Driven Content Distribution on April 1, 2026"

10.3.2 Three-Tier Traffic Pool Mechanism

After publication, Bilibili videos go through three traffic pools, each with different assessment dimensions:

Traffic Pool Tier

Time Window

Assessment Metrics

Passing Thresholds

Verification Phase (Initial Pool)

3-7 hours after posting

Completion rate primarily

Short videos >45% completion rate; Long videos >35% completion rate

Diffusion Phase (Recommendation Pool)

7-24 hours

Triple-action rate + Danmaku rate + Comment depth

Triple-action rate >8% enters the premium recommendation pool

Long-tail Phase (Search Pool)

7-60 days

Search ranking + sustained interaction

Keyword match + Completion rate >60% boosts search ranking

*💡 Freshness Mechanism: Newly published videos receive a "popularity boost" within 24 hours—under equal data metrics, new videos receive more recommendations than older ones. Therefore, the first 24 hours after posting are the critical window determining the video's fate.

Source: Smzdm.com "Detailed Explanation of Bilibili's AI Recommendation System Launch"*

10.3.3 Triple-Action Rate and Weight Ranking

The Triple-Action Rate (combined ratio of Coining + Collecting + Liking) is Bilibili's most core traffic distribution metric:

Triple-Action Rate Range

System Judgment

Traffic Distribution Result

15%

Extremely High-Quality Content

Triggers site-wide recommendations, enters the Hot List pool

8%-15%

High-Quality Content

Enters the recommendation pool, receives stable traffic distribution

5%-8%

Qualified Content

Remains in the initial pool, limited recommendations

< 5%

Low-Quality Content

Stuck in the initial pool, no further amplification

Triple-Action Weight Ranking (from highest to lowest):

1. Coining—Highest weight, only 1 coin per person per day, representing the user's highest endorsement

2. Collecting—Reflects long-term value, indicating users find it worth re-watching

3. Liking—Lowest weight, the interaction action with the lowest barrier

Other Interaction Metrics Affecting Traffic Distribution:

Metric

Weight

Description

Danmaku Rate

High

Bilibili's signature interaction; danmaku density reflects content discussion heat

Comment Depth

High

Comment length, depth of threaded discussions, not just simple comment quantity

Share Rate

Medium

Sharing off-platform brings new users; the system grants weighted credit

Completion Rate

High

Core metric for the verification phase; >45% (short videos) is the passing line

💡 Source: Smzdm.com, Jizhile Data "Complete Logic of Bilibili's Recommendation Mechanism"

10.3.4 Search Ranking Mechanism

Bilibili search traffic can account for over 60% during the long-tail period. The search ranking mechanism is as follows:

Search Ranking Formula (Weight Distribution):

Dimension

Weight

Description

Keyword Match Degree

40%

Degree of match between keywords in title, tags, and description and the search query

Interaction Data

35%

Historical interaction data such as triple-action rate, danmaku rate, and comment volume

Account Weight

25%

Follower count, level, historical content quality, account activity

💡 Relationship Between Completion Rate and Search Ranking: Videos with a completion rate exceeding 60% can see their search ranking improve by 2-3 positions. Thus, high completion rates not only affect the recommendation pool but also have a long-term impact on search traffic.

Keyword Layout Suggestions:

Include core keywords in the first 15 characters of the title (Bilibili titles truncate at approximately 15-20 characters)
Place 5-10 relevant tags in the tag section, including core words + long-tail words
Include keywords in the first sentence of the description, supplemented by a content summary

💡 Source: Smzdm.com "Bilibili Search Ranking Mechanism"

10.3.5 New Creator Support Program

Bilibili offers multiple support mechanisms for new UP hosts:

Support Program

Content

Eligibility Criteria

SUPERUP Program

10 billion traffic support for 100 fashion & lifestyle creators

Premium new creators in the fashion & lifestyle domain

Creation Incentive Program

Cash incentives based on views/interaction data

Followers >1000 or video views meeting thresholds

Traffic Ignition Program

Additional exposure weighting for new videos

First 10 videos from new accounts

Cold Start Support

Newly registered accounts can apply for a 50k-200k exposure new creator traffic pool after publishing 3 vertical content pieces

Specific categories (e.g., Tech Zone)

Account Launch Suggestions:

1. The first 3 videos from a new account must be in the same category and same vertical direction to trigger the new creator tag

2. Choose categories with support programs (e.g., Tech Zone/Knowledge Zone) to apply for additional traffic pools

3. Utilize the "Traffic Ignition Program" extra weighting for the first 10 videos; select topics with high search volume keywords

4. Apply for the "Creation Incentive Program" once followers reach 1000 to obtain cash incentives + traffic tilt

💡 Source: Smzdm.com "Bilibili New Regulations for H1 2025"

10.3.6 Invalid Traffic and Fraud Detection

Bilibili's rules for detecting invalid traffic and fraudulent behavior:

Behavior

Judgment Result

Viewing time below 30%

Not counted as valid completion; completion rate data excludes such views

Multiple accounts coining/liking/commenting under the same WiFi

Judged as "Network Environment Anomaly"; related interaction data cleared

Large influx of interactions concentrated in a short time

Triggers "Content Stability Check" anomaly; may result in throttling after attribution analysis

Concentrated negative reviews (mass dislikes + bad ratings)

"Positive Interaction Rate" too low, triggers throttling

Reposted/Duplicate Content

Identified by AI distribution system, resulting in demotion or non-recommendation

*💡 Key Insight: Mutual like/coin groups are not only ineffective on Bilibili but harmful—interactions from the same IP will be cleared and flagged as anomalies. Focus on genuine interaction quality.

Source: Jizhile Data, ZOL Q&A "Bilibili Recommendation Mechanism and Three-Tier Architecture"*

10.3.7 Practical Implications for WorkBuddy User Acquisition

1. Design Content Around "Coin-Driving": Coining has the highest weight; content must make users feel it is "worth spending today's only coin"—system tutorials, tool comparisons, and complete learning path content are optimal

2. Category Selection: Tech Zone/Knowledge Zone best fits the positioning of AI tool learning and have stable search traffic

3. Layout Keywords in the First 15 Characters of the Title: E.g., "WorkBuddy Complete Tutorial from Beginner to Master," covering core search terms in the first 15 characters

4. Danmaku Guidance: Set up "danmaku interaction points" in videos (e.g., "Type 1 if you learned this") to increase danmaku density

5. Lock Vertical Direction for First 3 Videos: Trigger the new UP host tag and apply for the cold start traffic pool

6. Do Not Participate in Mutual Like Groups: Same-IP mutual likes will be cleared and flagged; focus on real content quality

7. Long-Tail SEO Layout: Arrange search keywords in the title/tags/description of each video to leverage search traffic during the 7-60 day long-tail period

10.4 Overview of Traffic Distribution Mechanisms Across Three Platforms

Dimension

Douyin

Xiaohongshu (RED)

Bilibili

Traffic Pool Tiers

6 levels (Cold Start → Site-Wide Viral)

4 levels (Initial → Viral)

3 tiers (Verification → Diffusion → Long-tail)

Core Distribution Metric

Collection Rate (Highest Weight)

CES Score (Follow ×8 Highest)

Triple-Action Rate (Coining Highest Weight)

Weight Ranking

Collections > Re-visits > Die-hard Fan Interactions > 5-second Completion > Overall Completion > Comment Quality > Likes > Shares

Follows > Comments = Shares > Collections = Likes

Coining > Collecting > Liking; Danmaku Rate + Comment Depth

Competition Cycle

7-day long-term competition

5-min initial screening → 30-min same-pool elimination → 60-min tier determination

3-7 hour verification → 24-hour diffusion → 7-60 day long-tail

Search Traffic Share

Approx. 25%

30%-70% (Long-term)

60%+ (Long-tail period)

Key to New Account Launch

First 5 posts lock tags; No track change in first 20 posts

Observation period nurturing; First 5 posts vertical original

First 3 posts lock category; Apply for new creator traffic pool

AI Content Policy

Mandatory labeling; Unlabeled results in 50%-80% throttling

AI detection; Unlabeled results in demotion

Identified by AI distribution system; Reposting results in demotion

Fraud Detection

Fake followers lower SABCDE rating

Concentrated interactions from same IP cleared

Same-WiFi mutual likes cleared + flagged as anomaly

Best Content Type

Short videos (15-60 seconds), collection-value oriented

Image-text notes (800-1500 words), keyword driven

Medium-long videos (5-15 minutes), deep substantive content

Golden Posting Times

12:00-13:00 / 20:00-22:00

18:00-20:00 (Evening Peak)

💡 All weights, thresholds, and percentage data in the comparison table above are compiled from publicly available industry standards for 2024-2025. Please refer to the latest official platform rules.

10.5 30-Day Roadmap for Launching Accounts on Three Platforms (Based on Traffic Distribution Mechanisms)

Douyin Account Launch Roadmap (30 Days)

Phase

Time

Core Actions

Mechanism Basis

Nurturing Period

Days 1-3

Browse videos in the same field, like and collect, establish account interest tags

System collects account tags, determining subsequent traffic direction

Cold Start

Days 4-8

Publish first 5 vertical AI tool tutorial videos; Select topics with high search volume keywords

First 5 posts lock account tags; 25% search traffic entry point

Tag Stabilization

Days 9-15

Publish posts 6-10; Small-scale testing of content format differences (Pain Points/Reviews/Paths)

Tags preliminarily established in first 10 posts; System matches stably

Weight Enhancement

Days 16–25

Publish posts #11–20, chase AI trends + review data

After the first 20 posts, tags stabilize; you can compete on trending topics

Focus your efforts

Days 26–30

Review the best-performing direction and concentrate on publishing 1–2 viral topic ideas

7-day long-tail competition allows high-quality content to overtake early posts

Xiaohongshu Account Launch Roadmap (30 Days)

Phase

Timeframe

Core Actions

Mechanism Basis

Account Nurturing

Days 1–3

Browse and save AI tool notes in the same niche to establish interest tags

Observation period mechanism; system assesses account authenticity

Observation Period Publishing

Days 4–10

Publish the first 3–5 deep-dive original AI tool image-text notes

Initial exposure is low during the observation period; focus on quality

Normal Traffic Push

Days 11–20

Publish posts #6–10, deploy keyword matrix strategy

End of observation period; entry into normal traffic push pool

Search Long-Tail

Days 21–30

Continue publishing + optimize keywords for historical notes

Search traffic accounts for 30%–70% as long-term traffic

Bilibili Account Launch Roadmap (30 Days)

Phase

Timeframe

Core Actions

Mechanism Basis

Category Lock-in

Days 1–5

Determine category (Tech/Knowledge), publish first 3 vertical AI tutorials

First 3 posts trigger new UP creator tags

Cold Start Application

Days 6–10

Publish posts #4–6, apply for new user traffic pool

Cold start support: After 3 vertical content pieces, can apply for 50k–200k impressions

Validation Sprint

Days 11–20

Publish posts #7–10, optimize completion rate and triple-action rate (like/coin/favorite)

Traffic boost plan provides extra weighting; triple-action rate >8% enters recommendation pool

Search Long-Tail

Days 21–30

Continue publishing + optimize title/tag keywords

Long-tail phase search traffic >60%; completion rate >60% improves search ranking

All traffic push mechanism data in this chapter is compiled from publicly available industry sources for 2024–2025, including analysis articles from Smzdm.com, Zhuobo Tech, Qingua Media, Everyone Is a Product Manager, and Jizhile Data. Platform algorithms are continuously iterated; specific weights and thresholds are subject to the platform's latest official rules.

──────────────────────────────────────────────────

This v4 plan is formulated based on public operational data from the three major platforms as of August 2026, Douyin Enterprise Account official rules, and the "A-Content/B-Monetization" operational model. The core direction has been adjusted to: primarily promoting AI software like WorkBuddy in the early stages to help ordinary people learn AI, with the ultimate goal of driving traffic to the Physix Frontier website (physixfrontier.com/cn).

Policy