LINE TECHPULSE 2022 - AI Solutions of MarTech in LINE

by LINE Developers Taiwan

Slide 1

Slide 1 text

Johnson Wu / Charlie Tang / Nina Cheng / TW Data Dev AI Solutions of MarTech in LINE

Slide 2

Slide 2 text

Tonight, I’ll be eating Martech(?) ….

Slide 3

Slide 3 text

Most Common Marketing Challenges

Slide 4

Slide 4 text

Most Common Marketing Challenges How can I get more users? How can I boost the revenue? How can I save the cost? How can I keep my customers alive? (SBQI &NCFEEJOH PO6TFS "DRVJTJUJPO 6QMJGUPO"E QFSGPSNBODF #PPTUJOH "MHPSJUINPO $VTUPNFS -JGFUJNFWBMVF

Slide 5

Slide 5 text

Marketing Align the User Journey Content & Experience Social & Relationship Commerce & Sales Management Data Advertising & Promotion • Talent Mgt. • Finance Mgt. • Quality Mgt. • CRM • Channels & Partners • Sales Automation & Intelligence • CDP/DMP • Business Intelligence • Data Governance • Customer Exp. • KOL • Community & Review • Content • Experience Mgt. • SEO • Marketing • Advertising • PR

Slide 6

Slide 6 text

Apply graph embedding solutions to invite new service users. User Acquisition via Advertising:

Slide 7

Slide 7 text

Across Multiple services Correlate users with behaviors of other services .VTJD 4IPQQJOH $PNJD ?

Slide 8

Slide 8 text

Graph embedding can help Service knowledge graph provides abundant information 1 2 https://arxiv.org/pdf/1710.10903.pdf https://arxiv.org/pdf/1706.02216.pdf 1 2 Algorithm choice depends on data traits › Semantic matching based: ComplEX   › GraphSage algorithm   › Graph Attention network with deepwalk

Slide 9

Slide 9 text

New acquired users Case study on LINE POINTS wall : how to acquire new users? -*/&10*/54XBMMVTFST OPO-*/&10*/54XBMMVTFST 0 ffi DJBM "DDPVOU GPMMPXJOH TUBUVT ,OPXMFEHFHSBQI NPEFMJOH $MVTUFSJOH ,OPXMFEHF HSBQINPEFMJOH   $MVTUFSJOH OPO-*/&10*/54 XBMMVTFST NPTUMJLFMZUP KPJO-*/&SFXBSE Compose triples Compose triples Top k similar users

Slide 10

Slide 10 text

- Flexible and complex decision making - Embeddings show clustering effect w.r.t campaign New acquired users Case study on LINE POINTS wall : effectiveness and fully data orientation User followed OA category target size in 202106 target size in 202107 KG target size (tunable) Type A 2.2M 3.1M 500k Type B 10.4M 10.2M Type C 3.3M 4.1M

Slide 11

Slide 11 text

New acquired users Case study on LINE Reward : Steady growth of new users Broadcasted user count New user count

Slide 12

Slide 12 text

Takeaway User Acquisition on LINE Reward 1. Manual Decision Making of Audience Filtering Is Exhausting, but Martech Is More Flexible 2. Graph-Based Model Could Handle a Large Scale of Data With Reasonable Logics.

Slide 13

Slide 13 text

Apply uplift model to target the persuadable Advertising: How to Maximize Campaign Profit?

Slide 14

Slide 14 text

Boost the Profits with Incentives $PVOUPGDBNQBJHO .POUIPG:FBS 3FDFJWF 0"NFTTBHF 1VSDIBTF JODBNQBJHOQFSJPE &BSO -*/&$0*/-*/&10*/54

Slide 15

Slide 15 text

User Segments by Uplift High Low Low High Buy if treated Buy if not treated Sure Thing Sleeping Dog Lost Cause The Persuadable

Slide 16

Slide 16 text

User Segments by Uplift EFMJWFSDPTU JODFOUJWFDPTU High Low Low High Buy if treated Buy if not treated Sure Thing Sleeping Dog Lost Cause The Persuadable

Slide 17

Slide 17 text

User Segments by Uplift EFMJWFSZDPTU SFWFOVF High Low Low High Buy if treated Buy if not treated Sure Thing Sleeping Dog Lost Cause The Persuadable

Slide 18

Slide 18 text

User Segments by Uplift EFMJWFSZDPTU High Low High Low Buy if treated Buy if not treated Sure Thing Sleeping Dog Lost Cause The Persuadable

Slide 19

Slide 19 text

User Segments by Uplift EFMJWFSZDPTU *ODFOUJWFDPTU SFWFOVF High Low Low High Buy if treated Buy if not treated Sure Thing Sleeping Dog Lost Cause The Persuadable

Slide 20

Slide 20 text

User Segments by Uplift QPTJUJWFVQMJGU OFHBUJWFVQMJGU OPVQMJGU OPVQMJGU High Low Low High Buy if treated Buy if not treated Sure Thing Sleeping Dog Lost Cause The Persuadable Uplift M odel

Slide 21

Slide 21 text

User Segments by Uplift High Low Low High Buy if treated Buy if not treated Sure Thing Sleeping Dog Lost Cause The Persuadable 3FTQPOTJWF.PEFM Uplift M odel

Slide 22

Slide 22 text

A/B Testing to Collect Lookalike User Groups Treatment Group Control Group

Slide 23

Slide 23 text

User 360 Degree Features Campaign Period Observation Period, 30 days Purchase or not Purchase or not Browsing behavior Demographics Purchase history OA engagement Ad event Sticker collections Treatment Group Control Group

Slide 24

Slide 24 text

Slide 25

Slide 25 text

Meta-learners, S-Learner Campaign Period Observation Period, 30 days Browsing behavior Demographics Purchase history OA engagement Ad event Sticker collections Purchase probabilities Treatment Group Control Group $MBTTJ fi FS treatment = 1 treatment = 0 Purchase or not ≈ *OUVJUJWFBOEFBTZUPUSBJO -PXWBSJBODFBOE fl FYJCMF

Slide 26

Slide 26 text

Estimate Individual Treatment Effects Campaign Period Observation Period, 30 days Browsing behavior Demographics Purchase history OA engagement Ad event Sticker collections $MBTTJ fi FS Purchase probability Purchase probability treatment = 1 treatment = 0 candidate

Slide 27

Slide 27 text

Estimate Individual Treatment Effects Campaign Period Observation Period, 30 days Browsing behavior Demographics Purchase history OA engagement Ad event Sticker collections $MBTTJ fi FS treatment = 1 treatment = 0 Uplift Purchase probability candidate

Slide 28

Slide 28 text

Uplifts by Deciles 6QMJGU The Persuadable Sure Thing/Lost Cause Sleeping Dog

Slide 29

Slide 29 text

Facilitate LINE STICKER to Optimize Campaign Performances ROI OAP Profit -60% +10%

Slide 30

Slide 30 text

Lesson Learnt Consider the user segmentation with uplift A well-designed experiment is the must

Slide 31

Slide 31 text

Actionable and Explainable User Segmentation from RFM to CLV Customer Lifetime Value: How to identify valuable customers?

Slide 32

Slide 32 text

Know Your Customers ) 5 0 Historical Unknown Time Now From Buying Profile

Slide 33

Slide 33 text

Measure Your Customer Lifetime Value From RFM to CLV Past Now Future Measure Former Value Predict Future Value RFM CLV

Slide 34

Slide 34 text

Measure Your Customer Lifetime Value From RFM to CLV Past Now Future RFM CLV Measure Former Value Predict

Slide 35

Slide 35 text

Measure Former Value - RFM Modeling RFM Scores Historical data R F M Segments

Slide 36

Slide 36 text

Measure Former Value - RFM Modeling R F M RFM Segments Historical data Clustering

Slide 37

Slide 37 text

RFM Modeling - Clustering R F

Slide 38

Slide 38 text

RFM Modeling - Clustering R F

Slide 39

Slide 39 text

Measure Former Value - RFM Modeling R F M Historical data Clustering

Slide 40

Slide 40 text

Measure Your Customer Lifetime Value From RFM to CLV Past Now Future RFM CLV Measure Former Value Predict Future Value

Slide 41

Slide 41 text

Predict Future Value - CLV Modeling Demographic Browsing history Buying pro fi le OA interaction Feature Engineering Modeling Tags CLV High ↑ CLV Median CLV Low ↓ { > 50% High valued customers <= 50% Medium valued customers 0, Churned customers 450 days 180 days Now Cascade model

Slide 42

Slide 42 text

Predict Future Value - CLV Modeling Feature Engineering Modeling Tags CLV High ↑ CLV Median CLV Low ↓ { > 50% High valued customers <= 50% Medium valued customers 0, Churned customers 450 days 180 days Now User Embedding Demographic Browsing history Buying pro f i

Slide 43

Slide 43 text

Predict Future Value - CLV Modeling User Embedding Time of View Content Views from customers AD 1 AD n Product 1 Article 1 Article n Product n … … … Context window

Slide 44

Slide 44 text

Actionable and Explainable User Segmentation R↑F↑M↑ CLV Low Retain !

Slide 45

Slide 45 text

Actionable and Explainable User Segmentation R↑F↑M↓ CLV High Upselling !

Slide 46

Slide 46 text

Facilitate LINE TRAVEL TW

Slide 47

Slide 47 text

Acquire New User Maximize Campaign Profit Manage Customer Relationship Graph Embed. Uplift CLV MarTech Solution Pipelines

Slide 48

Slide 48 text

Scalability of MarTech How to build General MarTech solution for each service?

Slide 49

Slide 49 text

Pain of Reusability & Scalability 4PMVUJPO1JQFMJOF" 4PMVUJPO1JQFMJOF$ 4PMVUJPO1JQFMJOF% 4PMVUJPO1JQFMJOF# Sources Pipelines Service A Service C Service B Service D

Slide 50

Slide 50 text

Centralization and Automation Data Source %BUB 8BSFIPVTF Data Integration 'FBUVSF4UPSF Application "DRVJSFOFXVTFS "%5BSHFUJOH #PPTUSFWFOVF *NQSPWF3FUFOUJPO … ,(4PMVUJPO1JQFMJOF $-74PMVUJPO1JQFMJOF 6QMJGU   4PMVUJPO1JQFMJOF AI Solution 6TFSTFHNFOUBUJPO   4PMVUJPO1JQFMJOF … LINE Family Services Demographic Log

Slide 51

Slide 51 text

Applied MarTech Services LINE Family Services Benefit Content & Experience Social & Relationship Commerce & Sales Management Data Advertising & Promotion