ECOM 7121 Dynamic Digital Capabilities:

ABCD Business Models in US & China

Professor Amy Shuen

Dynamic Digital Capabilities in Sensing:

Shein & the AI-Driven Fashion Flywheel

SHEIN & ABCD, 6Ds and Breakout

Exponential Blue Ocean Session
Fast Fashion Strategy
What industry did Shein disrupt?

2
 From Pipeline to Platform…
idea-oriented, factory warehouse transportation store

warehouse-oriented,

little idea, a lot of customer feedback

get idea from social media

 Linear to Exponential

Andrew Ng’s AI-Data Virtuous Cycle or Flywheel

head of baidu: more data>>better product>>more users>>more data 4

Why Exponential Thinking Matters

Where’s the inflection point? 5

2020, the start of change
SHEIN Digital
Business
Model

[Link]
[Link]/our-
business/our-business-
model/

6
 SHEIN Dynamic
Digital
Capabilities
SENSE
SENSE: algorithmic trend
detection from TikTok &
Instagram; global design
crowdsourcing; 6k
bets/day
notice micro trend>>send to manufacturer>>influencers

SEIZE: rapid micro-batch

supplier network (via
Cloud)

RECONFIGURE: SEIZE
continuous assortment
refresh based on live
data RECONFIGURE

Which of these capabilities is hardest for competitors to copy? 7

SHEIN’s DDC
ABCD +
G(GenAI)

SENSE
A (AI): recommendation
algorithms

B (Blockchain): traceable
supplier payments (future
potential)

C (Cloud): elastic capacity,

distributed supply chain
dashboards

D (Data); millions of SKUS SEIZE

analyzed hourly

G (Gen AI): automated RECONFIGURE

product descriptions, image
generation, Kol content

8
SHEIN’s 6 Ds
mapping

SENSE
Digitalization-> Designs, data,
marketing digital. CAD-based
design files sourced globally

Deception-> Looked niche until

tipping point. Explosive micro-
trends in 6000 designs/day

Disruptionultra fast fashion

replaces linear fast-fashion model

Demonetization zero marketing

costs via social commerce,
influencers, gamification. Viral
marketing replaces paid ads. SEIZE

Dematerializationvirtual
storefronts and popups. No
physical stores RECONFIGURE

DemocratizationGlobal creators
and users co-create
9
Linear Fashion—6 month “lottery”

unused inventory

10
Reduce risk of excess stock or stockouts
& leverage fads, exponential demand…

Stockouts

less excessive inventory

Explosive
Demand

11
 Google’s Trends Finders

Predicting Image content in ads for click-

through rate
Quasi-
explosive Minimum order quantity MOQ for
experimentation

fashion KOLs Key Opinion Leaders and

trends Influencers

Spotting the “tipping point” faster

and more accurately real-time
 Huge Customer Database
• Own website not third party platform
• KOLs and celebrities (Key Opinion Leaders)
• Affiliate and Ambassador Programs
• Gamification—e-coupons, friends family
discounts, commissions
• Virtual events
• Pop up stores

13
 Zara vs. SHEIN
Zara SHEIN
Minimum batch size = 500 ?
Production cycle = 3-4 weeks ?
New SKUs per year = 12k ?

14
 The Economics of Agility
Zara SHEIN
Minimum batch size = 500 100
Design to market = 3-4 weeks 1 week
New SKUs per year = 12k 150k (in 2020)
New designs = ? 6k/day (est)

Who has the exponential edge?

15
 Forecasts updated hourly

Decision loops<24 hours

Exponentials Millions of micro-signals sensed per

at SHEIN day

Each viral post can trigger new SKUs

within days

Continuous refresh of assortments

real-time—6000 new “bets” per day
 Takeaways
• From Linear to Exponential Fashion
• In ultra-fast real-time fashion, exponential
feedback replaces forecasts
• ABCD + G technologies integrate sensing,
seizing, reconfiguring
• Data is the new fabric of fashion
• Ultra-fast-cycle learning via the AI-data
flywheel = dynamic competitive advantage

17
 ECOM 7121 Dynamic Digital Capabilities:
ABCD Business Models in US & China
Professor Amy Shuen

Dynamic Digital Capabilities in Sensing:

Shein & the AI-Driven Fashion Flywheel

SHEIN &
Blue Ocean Breakout
Exponential
Strategy Session
Fast Fashion
Blue Ocean Strategy Approach

19
 RETAIL STORE DESIGNS

How can you

reach new
large
underserved
markets?
WAREHOUSE VIRTUAL EVENTS
TIME (PRODUCT CYCLE) SOCIAL COMMERCE
BATCH SIZE INFLUENCERS

20
 Amazing Shein
“hauls” on Tiktok
 Can you find the “gamification”?
 What are the target customer
segments/profiles you see?
 How important is “influencer”or
KOL marketing to these
segments?
 What are the surprising cities or
locations mentioned?
 Why are smaller cities key?
 Why is SHEIN big is the US and
not China?
[Link]
pzDIGrLGs

21
SHEIN Blue Ocean Strategy: AI-sensing
unserved “explosive” micro-segments

Continuous AI data feedback

Physical Stores Number of new designs/day
Paid ads & marketing Global reach
Manual trend forecasting Number of “bets” on
explosive micro-trends

ELarge batch sizes

Time from design to cut Ultra-fast fashion ecosystem
Excess inventory with dynamic global AI
Ad campaigns making sensing and scaling of
celebrities or fashion icons micro-trends in days
the trend-setters
22
 Traditional Fast-Fashion Enabled By (ABCD + G &
Action Shein’s Blue Ocean Move
Practice DDC)
AI + Data (Sensing):
Physical stores and long Removed retail footprint
algorithmic trend
seasonal design cycles and replaced quarterly
Eliminate detection and demand
dependent on manual forecasting with
prediction; Cloud: always-
trend forecasting continuous digital sensing
on design pipeline
Reduced minimum batch Data + AI (Seizing): real-
Large batch sizes, excess
size from 500 to 100 units; time demand signals;
inventory, and expensive
Reduce minimized stockouts; GenAI: low-cost digital
ad campaigns reliant on
shifted from paid to viral creative content
celebrity marketing
social content generation

Raised design-to-launch
AI + Cloud (Sensing →
Speed of product iteration, speed from weeks to days;
Seizing): predictive
Raise personalization, and global increased personalization
modeling and agile micro-
data feedback integration through AI recommender
supplier networks
systems

Blockchain (Trust):
New digital ecosystem Created an ultra-fast, traceable supply chain;
linking designers, datafied fashion Cloud + GenAI
Create suppliers, influencers, and ecosystem that scales (Reconfiguring): virtual
customers in a continuous social trends globally showrooms, digital design
sensing loop within days studios, influencer
collaboration platforms
23
Apple AppStore Platform Ecosystem
Platform Ecosystem Players

1. Owner = Apple
2. Providers = iPhone/iPad
3. Producers = iOS app
developers
4. Consumers that use those
offerings

[Link]

24
SHEIN Ecosystem Platform

Platform Ecosystem Players and

Digital Sensing Nodes

Providers/Suppliers = small batch

1 week manufacturing suppliers
Producers = Designers and crowds
creating & combining micro-trends
Consumers=Users providing real-
time feedback and switching roles
to designers and influencers.
KOLs, Social media influencers =
social amplification signals

[Link]

25
 ECOM 7121 Dynamic Digital Capabilities:
ABCD Business Models in US & China
Professor Amy Shuen

Dynamic Digital Capabilities in Sensing:

Shein & the AI-Driven Fashion Flywheel

SHEIN &
Blue Ocean Breakout
Exponential
Strategy Session
Fast Fashion
TikTok Search #Shein

27
1. The Hashtag Page as Your “Trend Sensor”

Annotate the following in your screenshot:

 • Hashtag total = overall trend volume
 • ❤ Hearts = likes (engagement strength)
 • “Recent” tab = indicator of velocity
“The Shein sensing loop begins where new hashtags accelerate and
hearts multiply.”

28
2. Interpreting what you see
What You Observe What It Means How to Record It

Multiple new posts per hour Fast trend acceleration “10 new posts in 24 hours”

Many hearts on different High engagement and “Average 5K+ hearts per
creators’ clips diffusion post”

Repeated color palettes or Converging aesthetic signal “Pink metallic skirts + crop
product types jackets recurring”

Consistent audio reuse Shared cultural marker “Same sound trending across
clips”

29
3. Cross-check signal--Each platform
scales the sensing loop

• Instagram: confirm the visual identity of

the trend (#SheinHaul, #OOTD).
• Google Trends: verify rising search interest
in “Shein metallic skirt” or “Shein haul.”
• Reddit: read how communities describe or
critique the trend (“Is Shein still dominating
TikTok?”).

30
In your team, identify and analyze one emerging fashion or lifestyle micro-trend. Use
this worksheet to show how Shein (or your own startup) could sense, seize, and scale
this trend using ABCD + G technologies and the 6Ds of exponential growth. Draw your
sensing flywheel below.

1. Trend Name
What is your team’s chosen micro-trend?

2. Source Signal (Sensing)

Where and how did you first detect this trend? With what hashtags and search
terms? (TikTok)

3. Amplifying Signal (Sensing)

Could you identify amplifying platforms for this trend? Compare the timeframes or
diffusion patterns? (TikTok, Instagram, Google Trends, Reddit, etc.)

4. Draw Your Sensing Flywheel

Sketch or describe the loop: Data → AI → Micro-production → Feedback → More Data
7. Exponential Thinking Moment
Where is your trend’s inflection point? What metric signals exponential growth?
For example: when ……platform hearts/likes exceed ……….in ………..hours, order small
batch production.

Please submit your team worksheet to Moodle before the class wrapup. 31

•
Break Time…Please be
Back on Time for The
Breakout Sessions

Don’t forget to submit

your Team worksheets
to Moodle

32
 Thursday, October 16
Class 3

Crowdsourcing & Platform

Business Model Canvas
ABCD + G Threadless &
Technologies & Goldcorp Breakout
The Digital Duolingo & Session
Stone Soup Waze
Pre-reading reflection: Duolingo Case
1. How is crowdsourcing different than
crowdsensing at Duolingo?
2. How does Duolingo use ABCD + G
3. Trace Duolingo’s journey through the 6Ds
4. Do you see a Blue Ocean Strategy for
Duolingo? Edutainment Digitalization: online learning
Disruption: edutainment, gamify the process of language
education
Deception: exponentially growth during COVID-19
Demonetization: online payment, subscription
Dematerialization: language learning, remove the time and
place limitation; language ability test online, that is verified
by universities
Democractization: allow everyone to get access to
education of maths, languages, chess

34