email-segmentation

name	email-segmentation
description	Segment email lists for targeted, relevant messaging that improves engagement and conversion. Use when personalizing email campaigns.
origin	ECM

Email Segmentation

When to Activate

Use this skill when:

• Segmenting an email list for targeted campaigns
• Building personalization strategies beyond first-name merge tags
• Setting up behavioral segmentation based on user actions
• Implementing RFM (Recency, Frequency, Monetary) analysis
• Deciding between dynamic content and separate sends
• Collecting data for future segmentation
• Planning progressive profiling strategies

First Questions

Before segmenting, clarify:

1. What data is currently available? (Demographics, behavior, purchase history, engagement)
2. What email platform is being used? (Segmentation capabilities vary significantly)
3. What is the total list size? (Segments must be large enough to be actionable)
4. What is the business model? (B2B, B2C, e-commerce, SaaS, services)
5. What campaigns or sequences will use these segments?
6. Is there CRM or analytics data that can be connected to email data?
7. What level of personalization is the team equipped to maintain? (More segments = more content to create)

Core Rules

1. Relevance is the goal, segments are the means. The point is sending the right message to the right person, not creating complex segment logic for its own sake.
2. Start with 3-5 segments. Not 50. Over-segmentation creates maintenance burden and content gaps. Begin simple, refine over time.
3. Every segment needs its own message. If you segment but send the same email, you've wasted the effort. Different segment = different content or offer.
4. Behavioral data > Demographic data. What someone DOES predicts intent better than who they ARE.
5. Segments are living, not static. People move between segments. Set up rules for segment transitions.
6. Minimum viable segment size: 500-1,000. Segments smaller than this are hard to learn from and not statistically reliable for testing.
7. Data you collect but don't use is a privacy liability. Only collect data you will actually segment on.

Segmentation Types

1. Demographic Segmentation

Based on who the subscriber IS.

Data Point	Use Case	Example
Location	Localized offers, events, timezone-based sends	"Stores near you in [city]"
Age/generation	Tone, product recommendations, cultural references	Gen Z messaging vs Boomer messaging
Gender	Product lines, imagery, offers	Women's vs Men's product lines
Job title/role	B2B content relevance	CMO receives strategy content; manager receives tactical content
Company size	B2B solution positioning	Enterprise vs SMB messaging
Industry	Vertical-specific case studies and solutions	SaaS vs e-commerce examples

Limitation: Demographics tell you who someone is but not what they want right now. Always supplement with behavioral data.

2. Behavioral Segmentation

Based on what the subscriber DOES. This is the most powerful segmentation type.

Behavior	Segment Logic	Message Strategy
Pages visited	Visited pricing page	Sales sequence, demo offer
Content consumed	Downloaded [specific lead magnet]	Nurture on that topic
Email engagement	Opened 5+ emails in 30 days	VIP content, early access
Purchase history	Bought Product A	Cross-sell Product B
Cart abandonment	Added to cart but didn't purchase	Abandonment recovery sequence
Feature usage (SaaS)	Used Feature X but not Feature Y	Feature education email
Event attendance	Attended webinar on [topic]	Follow-up with related content
Inactivity	No open in 60+ days	Re-engagement sequence

3. Psychographic Segmentation

Based on what the subscriber VALUES. Harder to collect but highly effective.

Psychographic	How to Identify	Message Angle
Price-sensitive	Clicks on discounts, waits for sales	Lead with deals and value
Quality-focused	Buys premium, reads reviews	Lead with quality and craftsmanship
Convenience-driven	Fast checkout, repeat purchases	Lead with ease and speed
Status-seeking	Buys new releases, trending items	Lead with exclusivity and trends
Values-driven	Engages with mission content	Lead with impact and purpose

Collection methods: Surveys, quizzes, preference centers, behavioral inference (clicking discount emails = price-sensitive).

4. Lifecycle Segmentation

Based on where the subscriber is in their journey.

Subscriber → Lead → MQL → SQL → Customer → Repeat Customer → Advocate → Churned

Stage	Segment Criteria	Email Strategy
New subscriber	Joined in last 14 days	Welcome sequence
Engaged lead	Active but hasn't purchased	Nurture + soft sell
Sales-qualified	Visited pricing, demo request	Sales sequence
New customer	First purchase in last 30 days	Onboarding + post-purchase
Repeat customer	2+ purchases	Loyalty, cross-sell, referral
Advocate	High NPS, referrals, reviews	Ambassador program, exclusives
At-risk	Activity declining	Re-engagement before churn
Churned	No activity in 180+ days	Winback or sunset

5. RFM Segmentation

Recency, Frequency, Monetary — primarily for e-commerce and transactional businesses.

Segment	R	F	M	Strategy
Champions	Recent	Frequent	High spend	Reward, exclusives, referral asks
Loyal	Recent	Frequent	Moderate	Upsell, loyalty program
Potential loyalists	Recent	Low freq	Moderate-High	Nurture to increase frequency
New customers	Very recent	Low	Low-Moderate	Onboard, second-purchase incentive
At risk	Not recent	Was frequent	Was high	Winback offer, "we miss you"
Hibernating	Not recent	Low	Low	Re-engage or sunset
Lost	Very old	Low	Low	Final winback or remove

Scoring: Rate each dimension 1-5 (1=low, 5=high). Combine for composite scores. A 5-5-5 is a champion. A 1-1-1 is lost.

Data Collection for Segmentation

On Signup

• Keep forms short (email + 1-2 fields maximum)
• Use dropdown fields for structured data (not free text)
• Critical fields: email, name, one business-relevant qualifier (role, company size, interest area)

Post-Signup (Progressive Profiling)

• Collect one new data point per interaction
• Use preference center emails ("Tell us what you're interested in")
• Use surveys (1-3 questions max, incentivized)
• Use quizzes (interactive, value-driven, data-collecting)
• Track implicit data (clicks, page visits, purchases) — this requires no asking

From Behavior (Zero-Ask Data)

• Email engagement patterns (open times, click topics)
• Website behavior (pages viewed, time on site, search queries)
• Purchase history (products, frequency, average order value)
• Content consumption (which lead magnets, which blog topics)
• Support interactions (tickets, chat conversations)

Segment-to-Message Mapping

Every segment must have a corresponding message strategy. Use this template:

Segment: [Name]
  Criteria: [How someone enters this segment]
  Size: [Approximate count]
  Goal: [What action do we want from this segment?]
  Message angle: [What value proposition resonates?]
  Offer type: [What incentive or CTA?]
  Tone: [How does communication style differ?]
  Frequency: [How often do we email this segment?]
  Exit criteria: [When does someone leave this segment?]

Example Mapping

Segment: High-Intent Trial Users
  Criteria: SaaS trial user who has used 3+ features in first 7 days
  Size: ~800/month
  Goal: Convert to paid plan
  Message angle: "You're already getting value — unlock the full experience"
  Offer type: 20% first-month discount or extended trial
  Tone: Encouraging, data-driven, specific to their usage
  Frequency: Every 2-3 days during trial period
  Exit criteria: Converts to paid OR trial expires

Dynamic Content vs Separate Sends

When to Use Dynamic Content (One Email, Personalized Blocks)

• Minor variations (different product images, location-based details)
• Personalized recommendations
• Conditional text blocks (show/hide based on segment)
• When you have 5+ segments and the overall structure is the same

When to Use Separate Sends

• Completely different messaging angles
• Different offers for different segments
• When you want to A/B test messaging per segment independently
• When the design/layout needs to change per segment
• 2-4 distinct segments with meaningfully different needs

Dynamic Content Best Practices

• Default fallback for every dynamic block (what if the data is missing?)
• Test each variant — not just the default
• Document what each segment sees
• Keep dynamic logic simple — complex nesting creates QA nightmares

Progressive Profiling

Progressive profiling collects data gradually, over multiple interactions, instead of asking for everything upfront.

Implementation Plan

Interaction 1 (Signup): Email + first name only.

Interaction 2 (Welcome email click): Track which link they click = interest area. No asking required.

Interaction 3 (Preference center email, Day 7): "What topics interest you most?" — 3-5 checkboxes.

Interaction 4 (Survey, Day 30): "Quick question: what's your biggest challenge with [topic]?" — single question.

Interaction 5 (Content quiz, Day 60): Interactive quiz that provides personalized results + captures data.

After Interaction 5: You have enough data for rich segmentation without ever asking for a long form.

Progressive Profiling Rules

• Never ask for information you can infer from behavior
• Every ask must provide value to the subscriber (quiz results, personalized recommendations)
• Make every form/survey optional — never gate essential content behind data collection
• Store and update data centrally (CRM or ESP)
• Use preference centers — let subscribers self-segment

Segment Size Considerations

Segment Size	Viability	Action
Under 100	Too small for email	Combine with adjacent segment
100-500	Marginal	Use for high-value segments only (enterprise leads)
500-2,000	Viable	Good for targeted campaigns
2,000-10,000	Ideal	Large enough for testing and reliable metrics
10,000+	Very strong	Can sub-segment further

If a segment is too small: Broaden the criteria or merge with a related segment. Maintaining a tiny segment that requires unique content is rarely worth the effort.

Personalization Beyond First Name

Level	What It Looks Like	Data Needed
Level 0	Same email to everyone	Email address
Level 1	"Hi [First Name]"	First name
Level 2	Segment-based messaging	Segment data (role, interest, lifecycle)
Level 3	Behavioral triggers	Website/app behavior data
Level 4	Dynamic content blocks	Multiple data points per subscriber
Level 5	Predictive/AI-driven	Full behavioral + demographic data + ML model

The sweet spot for most teams is Level 2-3. Levels 4-5 require significant data infrastructure and content production capacity.

Segmentation Strategy Template

Use this template to document your segmentation strategy:

SEGMENTATION STRATEGY — [Brand/Product Name]

1. Primary Segmentation Axis: [Lifecycle / Behavioral / Demographic]
2. Number of Active Segments: [3-7 recommended]

SEGMENT 1: [Name]
  Criteria: [Entry rules]
  Size: [Count]
  Email strategy: [Sequences and campaigns this segment receives]
  Personalization: [What is customized for this segment]
  KPIs: [How success is measured for this segment]

SEGMENT 2: [Name]
  [Same fields]

[Continue for each segment]

DATA COLLECTION PLAN:
  Signup fields: [What's collected at signup]
  Progressive profiling: [What's collected over time and how]
  Behavioral tracking: [What actions are tracked]

REVIEW CADENCE:
  [Monthly / Quarterly review of segment performance and criteria]

Quality Gate

Before launching segmented campaigns, verify:

• Does every segment have a distinct message strategy? (If not, don't segment)
• Are segments large enough to be actionable? (500+ for most, 100+ for high-value)
• Is the data powering segments accurate and up to date?
• Are dynamic content fallbacks set for missing data?
• Have all segment/content combinations been tested visually?
• Are segment transition rules defined? (How does someone move between segments)
• Is there a documented segmentation strategy accessible to the team?
• Are segments reviewed and refreshed at least quarterly?
• Does the personalization feel helpful, not creepy? (Don't surface data that feels invasive)
• Is there a clear connection between segment and business outcome?