---
title: "2022 H1 Douyin Marketing Growth Review: H2 Content-Driven Growth Methodology"
description: "This article reviews the shift in e-commerce during 618, noting Tmall's decline and Douyin's rise, and introduces the FASD closed-loop growth model for Douyin. It details the 'Super Content System' with CAFE, SEVA, and AIMT methodologies for effective seeding and harvesting, using Dermalogica as a case study."
author: "飞扬"
publisher: "New Distribution"
email: "zhaobo258@gmail.com"
telephone: "+8615854817671"
published: "2022-07-02"
language: "en"
canonical: "https://xinjignxiao.com/en/articles/2022-h1-douyin-marketing-growth-review-h2-content-driven-growth-methodol-6aaf6b4d/"
markdown: "https://xinjignxiao.com/en/articles/2022-h1-douyin-marketing-growth-review-h2-content-driven-growth-methodol-6aaf6b4d.md"
original_source: "https://mp.weixin.qq.com/s/U3GJbRARHFd18vxLSd8Uqg"
translation: "https://xinjignxiao.com/zh/articles/2022%E4%B8%8A%E5%8D%8A%E5%B9%B4%E6%8A%96%E9%9F%B3%E8%90%A5%E9%94%80%E5%A2%9E%E9%95%BF%E5%A4%8D%E7%9B%98-2022%E4%B8%8B%E5%8D%8A%E5%B9%B4%E5%86%85%E5%AE%B9%E9%A9%B1%E5%8A%A8%E5%A2%9E%E9%95%BF%E6%96%B9%E6%B3%95%E8%AE%BA-6aaf6b4d.md"
attribution: "New Distribution — https://xinjignxiao.com/en/articles/2022-h1-douyin-marketing-growth-review-h2-content-driven-growth-methodol-6aaf6b4d/"
usage_policy: "https://xinjignxiao.com/ai-policy.txt"
---

# 2022 H1 Douyin Marketing Growth Review: H2 Content-Driven Growth Methodology

> This article reviews the shift in e-commerce during 618, noting Tmall's decline and Douyin's rise, and introduces the FASD closed-loop growth model for Douyin. It details the 'Super Content System' with CAFE, SEVA, and AIMT methodologies for effective seeding and harvesting, using Dermalogica as a case study.

********Click to read the original article for details********
This year's 618 saw fewer Tmall performance reports on WeChat Moments, while Douyin's reports increased significantly. Based on industry and client insights, as I predicted last year, 2021's Double 11 marked a turning point in the e-commerce era. The trend at 2022's 618 is almost irreversible: many brands are experiencing declining or even cliff-like drops in Tmall sales, while brands deeply operating within Douyin's closed loop performed well during 618.
Many mid-to-high-end international brands or traditional local brands entered Douyin relatively late, but this hasn't hindered their operations over the past few months. New entrants have achieved monthly sales exceeding ten million yuan on Douyin. How did they achieve late-mover advantage? Today, I'll share a review and some experience.
**1**
**FASD Douyin Closed-Loop Growth Model**
Today, consumers encounter new brands and products primarily through short-video content and KOLs. High-quality content first achieves the **'seeding effect'**—consumers become interested, and the accumulation of content continuously reinforces category and brand awareness. The accumulation of **'content building'** is the path to future **'brand building'**—a process of **'saving money'**. When savings are sufficient, this accumulation forms **'brand momentum'**.
In Douyin's closed-loop system, consumers seeded by products may purchase directly via linked carts or, through active **'search behavior'** or passive **'traffic reach'** , often complete purchases in brand live rooms, which are currently the most efficient harvesting tool. Ultimately, brands complete sales harvesting through seeding—a **'withdrawing money'** process—forming a closed loop from content seeding to sales conversion.
**In Douyin's closed loop, brand seeding is 'saving money,' and sales harvesting is 'withdrawing money.'**
Through balanced allocation of **'traffic supply'** , ad spending has clear goals: either amplify the content's **'seeding effect'** to build effective target audience reservoirs, or amplify the **'harvesting effect'** of sales to maintain both **'brand momentum'** and **'sales GMV'** growth in the brand's Douyin ecosystem.
Finally, through 'Cloud Map' scientific measurement of '5A audience assets,' data assets feed back into marketing strategy, iterating Douyin operational strategies, ultimately achieving effective growth in the brand's Douyin closed loop.
**The core logic of 'FASD based on Douyin closed-loop growth model' is the balance between 'seeding' and 'harvesting'** , achieving brand growth with both brand and performance effects. This is our Douyin closed-loop growth methodology.
**2**
**What is the Purpose and Significance of Seeding?**
In Douyin's closed-loop operational logic, the ultimate goal is to harvest sales and achieve brand sales growth on Douyin. If marketing investment cannot drive sales, such investment may be ineffective.
So, we need to explore: What is the purpose and significance of seeding? Let me first sort it out using reverse logic.
Sales in Douyin's closed loop consist of three parts: linked carts, store live rooms, and达人 live rooms. Linked carts are KOL seeding; due to high Star Map KOL quotes, sales output and ROI are relatively limited. The main sales come from store live rooms and达人 live rooms, and the KPI for达人 live rooms is often set based on store live room sales, varying by category, roughly in the range of 1:1 to 1:3—i.e., if store live room sales are 1 million,达人 live room sales target is 1-3 million. Thus, store live rooms become the cornerstone for measuring Douyin closed-loop sales.
From the ultimate goal and operational logic, brand store live rooms are the Douyin flagship store. To achieve stable, sustained, and scalable growth in store live rooms, we can clearly deduce the purpose of Douyin seeding: we need a large number of A1-A3 'effective target audiences' to improve harvesting efficiency and GMV scale.
Countless cases have verified that reaching non-target audiences rarely yields good harvesting results; the effectiveness of target audiences directly impacts store live room results. In Douyin's 5A audience assets, KOL seeding is given high weight, but there's a bug: Douyin can identify the number of people reached by KOL seeding but struggles to measure audience quality. Often, front-end videos are lively, but back-end conversions are zero.
My solution is: in the KOL seeding phase, attach shopping carts and look at the 'content model' ROI—the ratio of sales generated by that content to traffic costs through content ad spending. Based on ROI, we can know seeding effectiveness. Higher ROI indicates stronger effectiveness of the target audience behind the seeding video. This is the principle for scaling 'content ad spending,' thereby achieving scalable A1-A3 effective audience assets.
What's the logic behind the ROI of a KOL seeding video's 'content model'? There are two core points: First, front-end, look at the KOL's 'seeding ability,' which can be measured in advance through metrics like comment-like rate, cart click rate, GPM, etc.
Second, back-end, look at the KOL's 'fan purchasing power.' If the fan profile matches the brand's target audience well, metrics like female fan ratio, 24-30 index, iPhone index, and fan difference rate can indicate purchasing power. Good ROI means both 'seeding mindshare' and 'audience assets' have been effectively achieved.
Now, the hardest part: achieving seeding effects that break through consumer mindshare and drive good purchase conversion—everyone wants that. How to achieve it? And how to define 'seeding mindshare'? Honestly, it's quite difficult, with many details interconnected. Only by doing every step well can we achieve effective seeding. Our solution is called the 'Super Content System,' with three content methodologies to define effective seeding and create content that achieves seeding mindshare.
Finally, let me sort it out with forward logic: **Through KOL effective content seeding, achieve 'seeding mindshare'; based on 'content model' ROI efficiency, validate effectiveness; implement 'content amplification' ad strategy; thereby achieve scalable effective A1-A3 'audience assets.'** The ultimate goal is to complete effective audience reservoir for Douyin store live rooms, achieving efficient ROI conversion and scalable GMV harvesting.
**3**
**Super Content System: CAFE Scientific Account Selection**
This year, we served Dermalogica's Douyin closed-loop growth full case—a mid-to-high-end international beauty brand with an average order value of around 500 yuan. Based on this brand's operational experience, I'll share the application of the 'Super Content System' methodology and some insights from the process.
**Effective KOL seeding requires completing three steps with high quality simultaneously: scientific account selection, content co-creation, and ad amplification, to achieve effective creation and dissemination of 'seeding mindshare' content.**
The CAFE scientific account selection methodology is based on extensive Douyin KOL selection experience, implemented through technology as a Douyin KOL 'algorithm engine.' It measures KOL seeding value from four dimensions: communication power, commercial power, fan power, and growth power, resulting in a seeding-ability-oriented 'KOL RANK.' This forms an effective scientific account selection system, avoiding fake accounts, improving selection efficiency, and enhancing ad effectiveness.
For mid-to-high-end brands, the target audience is much smaller than for mass brands, so KOL requirements are much higher. High-AOV products necessarily require strong seeding mindshare in content and high match between fan profiles and brand target audiences to achieve effective content seeding.
**In front-end seeding ability, we focused on 'commercial power' metrics:**
Comment-like rate (comments + shares / likes): represents seeding ability and fan stickiness.
Cart click rate: represents potential purchase intent.
GPM + sales range: represents direct带货 ability (these two must be viewed together, as some KOLs may inflate play counts, causing GPM to be artificially high).
For high-AOV products, we need to check the product prices in带货 data—whether there are similar competitors or same-level AOV products. Good low-AOV带货 doesn't prove high-AOV will work. We also reference category preferences from带货 data, similar to AOV.
**In back-end fan purchasing power, we focused on 'fan power' metrics:**
From the start, after reviewing the brand's target audience profile, we knew Douyin's audience would be much smaller. High-AOV products require higher fan profile purchasing power. I won't cover basic metrics; let me discuss a few explicit indicators related to purchasing power (high AOV).
iPhone index: iPhone share over 40% indicates a majority of wealthy, attractive women.
24-30, 31-40 index: These two age groups must have high proportions; they are the brand's main audience. Under 24 is not.
First-tier index: The combined proportion of first-tier and new first-tier cities should be high; first-tier city purchasing preferences differ greatly from lower-tier cities.
Additionally, through reverse inference from Cloud Map audience assets, Douyin's new eight major audience profiles are important references for mid-to-high-end brands. Don't be misled by audience names—for example, urban blue-collar workers are the real consumption mainstay; they are the urban workers around you.
From Douyin's new eight major audience flow relationships, we can see that high-spending beauty consumers are mainly concentrated in new sharp white-collar, refined mothers, urban blue-collar, and established middle class, which basically aligns with Douyin industry report analysis on beauty penetration.
But after reviewing many KOL fan profiles, we found very few new sharp white-collar—possibly due to dynamic role changes—and relatively few established middle class, which is understandable. The remaining two explicit high-spending groups are refined mothers and urban blue-collar.
**Finally, the ideal fan profile combination is: high proportion of ages 24-30 & 31-40 + high iPhone share + high first-tier & new first-tier city share + high proportion of refined mothers & urban blue-collar in the new eight major groups.**
In our final review, the most effective seeding content had KOL fan profiles matching this definition, with refined mothers and urban blue-collar as the main groups, and even a relatively high proportion of established middle class.
**4**
**Super Content System: SEVA Content Creation**
**SEVA is the second content methodology of the 'Super Content System,' a content creation methodology centered on Douyin's strong seeding logic.** It attracts target audiences through scene association, evokes resonance through emotional engagement, achieves effective seeding through value delivery, and completes the final step of conversion through action directives, helping brands achieve **'brand and performance unity'** in content creation and dissemination.
For high-AOV products, content quality requirements are higher. Based on seeding mindshare logic, product efficacy and communication scenarios must align more precisely; otherwise, it's hard to hit pain points. Previously, I've shared vertical talking-head KOL seeding content. Today, I'll use SEVA to deconstruct a VLOG-style seeding content.
**0-10s Scene Association:** The opening 'day and night reversed'切入熬夜场景 (night owl scenario), and the product is a night-repair emergency mask, with precise communication scenario matching. Another interesting point: the opening question 'What method can make your husband more pleasing to the eye?' helps break through traffic pool recommendations and precisely targets refined mothers.
**11-30s Emotional Engagement:** The husband is exhausted, and the wife cares for him, with a warm, touching scene. Emotional engagement comes from emotional connection—either evoking memories or aspirations.
**31-140s Value Delivery:** The product's core selling points are night-repair emergency and de-yellowing/brightening. For the anti-aging review point, they used 'The future is long, and I don't want to age so fast' to transform it. In the effect demonstration, there's a big highlight: first the wife applies the mask, then the husband—double demonstration, clear effect comparison, but not violating rules.
**141-160s Action Directive:** An implied action directive: 'I can't remember how many nights you've stayed up for our family,' tying back to the product seeding. 'You accompany me, I cooperate with you; perhaps this is the best form of marriage.' The ending may not only seed but also move—buying it is a longing for beautiful love.
So, whether it's vertical talking-head KOLs or pan-vertical lifestyle KOLs, as long as content creation targets **'seeding mindshare'** , the SEVA content creation methodology and framework are highly applicable.
**5**
**Super Content System: AIMT Performance Amplification**
**AIMT is the third methodology of the 'Super Content System'—an ad strategy for content amplification.** Today, mid-tier KOLs on Douyin get only a few hundred thousand natural plays, but their Star Map quotes may exceed six figures. If we only look at the ROI of natural traffic, we've calculated that normally ROI is less than 0.3. So why do so many brands still invest in KOLs one after another?
We generally calculate ROI in two ways: one is based on the relationship between KOL带货 output and Star Map cost, calculating the KOL's Star Map ROI. The other is based on the relationship between KOL带货 output and traffic cost, calculating the 'content model' ROI, aimed at ad amplification—usually called the ad ROI model.
For example, a KOL's Star Map quote is 100k, natural plays 250k,带货 20k. Star Map ROI is 1:0.2—a bit miserable. Using the content model, 250k plays cost about 10k,带货 20k, so the ad ROI model is about 1:2—very good. Theoretically, with such a good content model, if I keep buying traffic, sales should rise accordingly. That's the general logic, but short videos have a clear decay cycle; ROI declines over time. The core of content ad spending is **'racing against time'** .
Based on the AIMT performance amplification methodology, for high-quality KOL seeding content with an ideal ad ROI model, we need to leverage traffic leverage well, often using Dou+/随心推, content services, and达人竞价 (AD/千川) to amplify KOL native videos 50-100 times, up to the 50 million play ceiling.
In Dermalogica's 618 campaign, the content amplification ad strategy was clearly implemented from account selection. Because high AOV makes it harder to find KOLs with good cart conversions than typical projects, once the best 'content model' appears, most content ad ammunition should be fired within 2 weeks to quickly end the battle, grabbing enough volume before content decay. If there are more good content models, the ROI and scale of content ad spending can go even higher.
Facts have proven that even in high-AOV mid-to-high-end beauty, the AIMT methodology is more applicable, but operational difficulty is higher because high-quality KOLs and high-spending audiences are limited. The best you can do is maximize traffic leverage; under-investing in content amplification is the biggest loss.
Take the seeding video I just deconstructed with SEVA: its plays have now reached nearly 30 million—10 times the KOL's natural traffic—and still increasing. The premise for ad amplification is that content has the best ROI model, so we used the 'content amplification' ad strategy, adding at least 10 times traffic leverage. That's the value of content with 'seeding mindshare.'
Because of such high-quality KOL seeding content, Dermalogica gained 50 million+ effective target audiences during 618, with high utilization of content ad leverage, significantly reducing brand communication costs (low CPM for content ad spending) and improving efficiency (scaling based on best content ROI model), achieving brand and performance unity and effectively supporting 618 promotion conversion and harvesting.
**6**
**Verifying the Relationship Between Seeding and Harvesting**
In the FASD Douyin closed-loop growth model, I highly value the **'seeding' and 'harvesting' balance** , which is key to achieving brand growth with both brand and performance effects.
In the past, few could clearly quantify the relationship between seeding and harvesting. Over the past six months, we've been validating this, and many projects have confirmed it—especially for brand store live rooms with monthly sales above 5 million, subsequent ROI stability and sales growth depend on KOL seeding effects. During store live room harvesting, we highly value Cloud Map A1-A3 audience assets; of course, the more A2-A3 audiences, the higher the harvesting efficiency.
Taking Dermalogica's 618 KOL campaign as an example, the overall seeding burst occurred after May 20, especially after some high-quality videos were released, with substantial content ad support, enabling the brand to achieve 50 million+ A1 audiences and tens of millions of A2-A3 audience growth during the promotion.
Because the content ad strategy is based on content model ROI, high-conversion content is backed by the most precise target audiences. Even though most KOLs use short videos with shopping carts, a large amount of traffic still spills over to the brand's Douyin live room and Tmall flagship store. During this period, third-party data shows that Douyin search index trends and Taobao keyword searches both rose year-over-year with the KOL content seeding effect.
So, what about the harvesting efficiency in Douyin's closed loop (store live rooms)? With KOL 'seeding mindshare' content plus the amplification effect of 'content ad spending,' audience assets exceeded 50 million+, store live room daily sales jumped from 50k to 500k+, and the conversion rate and ROI for the multi-dimensional mask steadily improved. Dermalogica's Douyin closed-loop sales broke through the 10 million monthly sales mark.
Due to space limitations, I won't elaborate on store live room tactics and strategies. This article focuses on the relationship between 'seeding' and 'harvesting.' If there's another opportunity, I'll share the operational logic and experience of brand store live rooms.
**7**
**Conclusion**
Let's review again. The reverse logic is:
Based on the goal of 'store live room harvesting,' we need scalable effective A1-A3 'audience reservoirs.' Effective audience acquisition is through 'content amplification' ad strategy, which selects 'content models' with good ROI, and the best content models often come from KOL 'seeding mindshare' content creation.
The forward logic is:
Through KOL effective content seeding, achieve 'seeding mindshare'; based on 'content model' ROI efficiency, validate effectiveness; implement 'content amplification' ad strategy; thereby achieve scalable effective A1-A3 'audience assets.' The ultimate goal is to complete effective audience reservoir for Douyin store live rooms, achieving efficient ROI conversion and scalable GMV harvesting.
Whether forward or reverse, creating KOL 'seeding mindshare' content is not easy; other steps are about validating this. We've defined the 'Super Content System,' which includes three methodologies: CAFE scientific account selection, SEVA content creation, and AIMT performance amplification. These are interconnected; when all are done well, the brand achieves effective seeding and completes Douyin closed-loop harvesting.
In the past, when I talked about 'content-driven growth,' it was more of a prediction. Today, it's been truly validated: **Douyin's closed-loop ecosystem has high growth efficiency, the linkage between 'seeding' and 'harvesting' is very strong, and audience asset value is gradually being scientifically measured, becoming a decision basis for marketing growth strategies.**
The end of this year's 618 marks the end of an e-commerce era. The new era of content-driven growth, whether you perceive it or not, has arrived and is unstoppable.
Source: Growth on the Road (ID: gh_d3e055ed2c1b) Author: Feiyang
**Feiyang, CEO of Yiling Growth** , has 15 years of internet entrepreneurship, 10 years of FMCG marketing and e-commerce experience, and is a senior social effect marketing expert.
He founded China's largest beauty mobile community, Beauty Tips APP, with over 80 million mobile users. He is the founder & CEO of Beauty Tips and former VP of a well-known 4A advertising group.
In the past 3 years, in the social effect placement field, he has served many new, local, and international brands including Florasis, Perfect Diary, Winona, Yourshine, Chando, Osman, Vaseline, Dove, Clear, Banila Co., L'Occitane, Nestlé, Mondelez, Corona, Dyson, Philips, Gerber, and Neiwai.
As a marketer, he created Douyin placement systems like 'Douyin Three Axes,' 'CAFE Algorithm Account Selection,' and 'STOM Full-Link Placement.' As a product manager, he developed the WiseONE content marketing middle platform. He is also the author of the well-known marketing public account 'Growth on the Road.'
 _-END-_


---

## Copyright and AI use

This article is sourced from New Distribution. Search, quotation, summarization, and model training are permitted, but every use must credit New Distribution and retain the canonical source URL.

Contact: zhaobo258@gmail.com · +86 158 5481 7671