PM Customer Segmentation
Guide (2026 Edition)
5 types of segmentation from demographic to JTBD, 6 rules for using segments well, and 6 common segmentation mistakes.
Build User Segmentation Skills Daily — Free →5 Types of Segmentation
1. Demographic (weakest)
Age, gender, location, income.
✅ Useful for
Basic targeting, marketing campaigns.
⚠️ Limitation
Doesn't explain behaviour — two 28-year-old Bangalore women can have completely different product needs.
2. Behavioural
Segmentation by in-product actions — feature usage, frequency, depth.
✅ Useful for
Retention work, feature prioritisation, lifecycle marketing.
⚠️ Limitation
Describes WHAT users do but not WHY. Still useful for action.
3. Needs-based
Segmentation by user goals — what they're trying to achieve.
✅ Useful for
Product positioning, onboarding customisation, feature roadmap.
⚠️ Limitation
Requires qualitative research to map correctly. Worth the investment.
4. Jobs To Be Done (JTBD)
Segmentation by the job users hire the product to do.
✅ Useful for
Strategic product direction, new market entry, understanding switching behaviour.
⚠️ Limitation
Hardest to identify but most strategically powerful. Best-in-class PM segmentation.
5. Value-based (B2B)
Segmentation by ARR tier, growth trajectory, or strategic importance.
✅ Useful for
Customer success prioritisation, feature access, roadmap decisions.
⚠️ Limitation
Mostly a B2B tool. Less relevant for consumer products.
6 Rules for Segmenting Well
Pick segments that would lead to different product decisions — if not, you're overfitting
Validate segments with data — don't assume the segments in your head are real
Size each segment — a beautiful segment with 0.5% of users isn't worth targeting
Design for one segment at a time — trying to serve all loses everyone
Revisit segmentation annually — user base shifts, segments evolve
Share segmentation widely — design, marketing, CS all benefit from shared mental model
6 Common Mistakes
Creating 10+ personas — no team remembers more than 3
Demographic-only segmentation — 'millennials' isn't a product strategy
Inventing segments without validating them in data
Over-indexing on vocal minority — loudest users ≠ most important segment
Segmentation frozen in time — user base shifts in 12 months
Not linking segments to decisions — segments that don't change what you build are theatre
FAQ
What's the difference between a segment and a persona?
A segment is a measurable group in your user base. A persona is a narrative representation of a segment — a name, a story, demographics. Segments are for analysis; personas are for alignment. PMs often confuse them. Use both: segments for the spreadsheet, personas for the PRD.
How many customer segments should a PM product team track?
3–5 primary segments. Fewer and you're grouping dissimilar users; more and no one remembers the segments (so they get ignored). Start with 3, add as you genuinely need to differentiate product decisions. If two segments always get the same feature decisions, they're one segment.
Build Segmentation Intuition Daily
Scenarios that force you to define users specifically — not vaguely.
Start Free Trial →