MarTech · 27 พ.ค. 2026 · 7 min read

Data Stack สำหรับ E-commerce / D2C ในไทย ปี 2026

คู่มือ data stack สำหรับ e-commerce และ D2C ในไทย ปี 2026 — Shopify, Klaviyo, Segment, BigQuery, GA4 e-com, attribution, customer.io, retention stack

TL;DR: E-commerce / D2C data stack 2026 = Shopify/Woo + Klaviyo + GA4 + Segment + BigQuery + Reverse ETL. Differ จาก B2B stack: focus ที่ purchase events + LTV + cohort + retention มากกว่า lead score + ABM. คู่มือนี้ครอบคลุม: e-com platform choice, email/SMS automation (Klaviyo/Customer.io), CDP role for D2C, warehouse setup, attribution gaps with marketplace (Shopee/Lazada), LTV modeling, retention stack, subscription handling

E-com Data Stack ต่างจาก B2B ยังไง

มิติ	B2B Data Stack	E-com / D2C
Primary CRM	HubSpot/Salesforce	Shopify/WooCommerce (= “CRM” effectively)
Email Engine	Marketo/Pardot	Klaviyo / Omnisend / Customer.io
Critical Event	Demo request / SQL	Purchase + AOV + LTV
Funnel Length	30-180 วัน	<7 วัน (often <1 hour)
Identity	Email + Company	Email + Phone + LINE ID
Activation	Sales workflow	Email/SMS lifecycle automation
Attribution	Multi-touch over months	Last-click + DDA, short window
Subscription	Annual contract	Recurring purchase (subscription model)
3rd-party Data	Firmographic	Behavioral + Lookalike audiences

E-commerce Platform Choice

Shopify (Cloud SaaS)

Best for: D2C, fast launch, ≤$10M GMV

✅ Pros:
– Fastest setup (days, not months)
– Best-in-class app ecosystem
– Native integrations: Meta, Google, TikTok, Klaviyo
– Payment integrations (TH: PromptPay, Omise, 2C2P)
– Mobile-first checkout

❌ Cons:
– Monthly fee + transaction fee
– Less customization than self-hosted
– Plus tier ($2k/month) needed for scale

TH pricing 2026: Basic $39, Shopify $105, Advanced $399, Plus $2k+/month

WooCommerce (Self-hosted WordPress)

Best for: Already on WordPress, content+commerce mix, custom needs

✅ Pros:
– Free core
– Full customization
– Many plugins
– TH agencies many

❌ Cons:
– Self-host = manage server
– Performance tuning required
– Security responsibility
– Slower checkout (without optimization)

TH common stack: WooCommerce + Cloudways/Kinsta hosting + premium plugins

LINE MyShop

Best for: TH-native, leverage LINE OA followers

✅ Pros:
– In-LINE checkout
– High conversion rate (TH-trust)
– Free to start
– TH-native payment + delivery

❌ Cons:
– Limited customization
– LINE-only ecosystem
– Less reporting depth

Marketplace-only (Shopee + Lazada)

Best for: Test product, leverage marketplace traffic, ไม่ต้องสร้าง brand site

❌ Limitation: ไม่ได้ customer data — marketplace owns it

Recommendation by stage

Test product: Marketplace + LINE MyShop
Brand validation: Shopify + Marketplace
Scale: Shopify Plus + WooCommerce + Marketplace + LINE

GA4 E-commerce Setup

Required Events (Enhanced E-commerce)

view_item            — product detail page
view_item_list       — category page
select_item          — click product card
add_to_cart
remove_from_cart
view_cart
begin_checkout
add_shipping_info
add_payment_info
purchase             — primary conversion
refund               — if applicable

Implementation

Shopify: Native GA4 setup via Shopify Admin → Sales Channels → Google Channel
WooCommerce: Plugin “GA4 for WordPress” หรือ GTM
Custom: GTM Server-side container — recommend สำหรับ accuracy

Key Reports

Monetization → E-commerce purchases — revenue + items
Acquisition → Traffic acquisition — channel attribution
User → User lifetime — LTV cohorts
Custom Audiences for Ads — push to Meta/Google

Data Stack สำหรับ E-commerce / D2C ในไทย ปี 2026 — แผนภาพที่ 1

Klaviyo — Email + SMS Engine

Why Klaviyo dominates D2C

Built for e-com (vs HubSpot, Mailchimp generic)
Native Shopify integration (deep, real-time)
Pre-built flows (cart abandon, welcome, post-purchase)
Segmentation based on purchase behavior
AI-powered send time + subject line

Klaviyo Pricing

Free: ≤250 contacts + 500 email sends/month
Email: $30-$2000+/month (scale by contacts)
SMS: $5 base + per-message (TH: ~0.50-1.50 บาท/SMS)

Core Flows ต้องมี

Welcome Series — 3-5 emails after subscribe
Abandoned Cart — 1-3 emails, 1 hour / 24 hr / 72 hr
Abandoned Browse — viewed product, didn’t add
Post-Purchase — thank you + delivery + review request
Win-back — inactive 60-90 days
VIP — top 10% customers

TH-Specific

SMS still works in TH (used for OTP + delivery notifications)
LINE integration via 3rd-party plugin (Klaviyo → LINE OA broadcast)
Email open rates lower TH (~15-20% vs 25-30% US) — adjust expectations

Alternatives

Omnisend — cheaper, similar feature set
Customer.io — more flexible, dev-led
Postscript — SMS-first
Mailchimp — last resort (weak for e-com)

CDP for E-com — Different Profile

When E-com needs CDP

Multi-store (Shopify + Marketplace + LINE MyShop)
Subscription business (recurring billing)
App + Web both
500k MTU

Klaviyo as Light CDP

Klaviyo for many TH D2C = good enough as “CDP-lite”:
– Profile data
– Behavioral events
– Segmentation
– Activation to Meta/Google audience

Proper CDP

Segment — for multi-source
RudderStack — open-source friendly
Tealium — enterprise

When to graduate

Need data ใน warehouse for BI/MMM
Multiple email/SMS tools needed
Custom destinations (Snowflake → custom internal app)

Subscription / Recurring Stack

Subscription Tools

Recharge (Shopify) — most popular
Bold Subscriptions
Loop Subscriptions
Native: Shopify Subscriptions — basic but free

Subscription Data Challenges

LTV calculation (cohort-based)
Churn modeling
Trial-to-paid conversion
Subscription pause/cancel reasons

Data Stack Add-ons for Subscription

Push subscription events → CDP/warehouse
Build cohort retention dashboard
Trigger flow on churn risk (Klaviyo)

Marketplace Data Integration

Problem: Marketplace = Closed Data

Shopee/Lazada sell ให้ลูกค้าคุณ — แต่คุณไม่เห็น:
– Customer email (Shopee masks)
– Customer phone (masked)
– Browse behavior before purchase
– Cross-purchase data

What you CAN get

Order data (export weekly)
Customer name + shipping address
Order value + items
Repeat purchase rate (in-marketplace)

How to Integrate

Manual CSV export → warehouse (weekly)
3rd-party tool: SellerCenter sync, Shippop
Build custom scraper/API

Activate Marketplace Data

Calculate true cross-channel LTV
Identify cross-platform customers (same address/phone)
Build lookalike on Meta from marketplace buyers (use email hash if available)

LTV Modeling

Why LTV matters more than ROAS

ROAS = single transaction
LTV = total revenue per customer over lifetime
D2C with subscription/repeat = LTV often 3-10× first purchase
Optimize CAC vs LTV, not vs first-order revenue

Simple LTV Formula

LTV = AOV × Purchase Frequency × Gross Margin × Customer Lifespan

Example:
– AOV: 1,500 บาท
– Frequency: 4× per year
– Gross margin: 40%
– Lifespan: 3 years
– LTV = 1,500 × 4 × 0.40 × 3 = 7,200 บาท

Cohort-Based LTV

แบ่งลูกค้าตามเดือนที่ acquired → track revenue per cohort over time

Predictive LTV

Use machine learning to predict LTV early (เช่นจากการซื้อแรก)
– Tools: GA4 Predictive Audiences, custom in BigQuery ML
– Use case: bid higher for high-pLTV customers ใน Meta/Google

Retention Stack

Tools (beyond email)

Loyalty programs: Smile.io, Yotpo, LoyaltyLion
Reviews: Yotpo, Stamped, Judge.me
UGC: Stamped UGC, Loox
Referrals: ReferralCandy, Stamped, Friendbuy
SMS: Postscript, Attentive (US-focused), local TH SMS

TH-Specific Retention Channel

LINE OA — broadcast, 1:1 chat, rich menu
LINE Stamp — engagement
LINE MyShop — repurchase friction-less

Retention KPIs

Repeat purchase rate
Customer Retention Cohorts (% returning by month)
90-day repeat rate
LTV/CAC ratio (>3 healthy)

Reverse ETL Use Cases for E-com

Custom Audience push — top customers, churn risk
Suppression list — exclude purchasers from prospecting ads
Lookalike seeds — high-LTV customers → Meta/Google LAL
Email segmentation — based on warehouse-calculated traits
CRM enrichment — push behavior back to CRM

Tools

Hightouch (most popular)
Census
RudderStack (open-source)

Privacy + PDPA for E-com

Consent Capture

Cookie banner at site visit
Email opt-in at checkout (separate from terms)
SMS opt-in explicit (TH PDPA + telco rules)
Marketing communications opt-in

Data Retention

Customer order data — keep for tax (7 years TH)
Behavioral data — 1-2 years standard
Inactive customer — delete after 2 years

Right to Delete

Process within 30 days TH
Document delete: warehouse, CRM, Klaviyo, ad platforms

Stack by Stage

Stage 1: First $100k GMV (1-12 months)

Shopify Basic + Theme
Klaviyo Free / Starter
GA4 + Meta Pixel + CAPI (via Shopify channel)
Google Sheets for marketplace data
Cost: $50-300/month

Stage 2: $100k-$1M GMV (12-24 months)

Shopify Shopify tier
Klaviyo Email + SMS
GA4 + BigQuery export
Recharge if subscription
Reviews (Yotpo)
Cost: $500-2,500/month

Stage 3: $1M-$10M GMV (24+ months)

Shopify Advanced
Klaviyo Pro tier
Segment + BigQuery + dbt
Hightouch for activation
Predictive LTV in BigQuery ML
Cost: $3,000-15,000/month

Stage 4: $10M+ GMV (Enterprise)

Shopify Plus / Headless
Segment Business + Klaviyo Custom
Snowflake / BigQuery production
Custom data team (2-5 people)
MMM + Incrementality testing
Cost: $30,000+/month

Common Pitfalls

Trust Shopify reports — Shopify last-click = overstate Meta
No CAPI → Meta/TikTok signal loss 30-40%
Marketplace orders not in warehouse — incomplete picture
Single attribution model — ROAS lies (see attribution guide)
Email broadcast instead of behavior-triggered — low engagement
No subscription analytics — churn surprise
Customer ID fragmented — guest checkout = anonymous
Stack without ownership — tools rot
Skip predictive LTV — bid same on everyone
Focus on AOV, ignore LTV — short-term ROAS, long-term loss

Quick Start Checklist

Week 1-2 — Foundation

[ ] Shopify + Klaviyo + GA4 setup
[ ] Meta Pixel + CAPI via Shopify channel
[ ] Google Channel app for Merchant Center
[ ] TikTok Pixel + EAPI

Month 1 — Core Flows

[ ] Welcome Series
[ ] Abandoned Cart (3-email)
[ ] Post-Purchase + Review request
[ ] Win-back
[ ] BigQuery export from GA4

Month 2-3 — Data Layer

[ ] Marketplace data weekly import
[ ] Basic LTV dashboard in BigQuery
[ ] Custom Audience automation (Klaviyo → Meta)
[ ] Subscription analytics (if applicable)

Month 4+ — Optimization

[ ] Predictive LTV
[ ] MMM model (quarterly)
[ ] Reverse ETL for ad platform activation
[ ] Cohort retention dashboard

บทความที่เกี่ยวข้อง

Data Stack สำหรับ B2B — companion for B2B
MarTech Analytics & Tracking Guide — GA4 + GTM Server
Attribution & Modeling Guide — MMM + Incrementality
Meta Ads สำหรับ B2C — Catalog + DPA setup
LINE & Marketplace Ads Guide — TH commerce ecosystem

TL;DR ย้ำ: E-com data stack 2026 = Shopify + Klaviyo + GA4 + BigQuery + Reverse ETL. Focus ที่ LTV cohort + retention, ไม่ใช่แค่ ROAS ครั้งแรก. Marketplace data ต้อง manual import. Predictive LTV + custom audience push คือ unlock การ scale CAC ที่ profitable