Needs-Based Segmentation

First PublishedMay 24, 2026Last UpdatedMay 25, 2026ByAtif Alam

The decision this page enables: how to slice your market by the outcome customers want, so your segments stay stable even as demographics and technology shift.

What needs-based segmentation is

Needs-based segmentation (also called outcome-based or JTBD-driven segmentation) groups customers by the job they’re hiring the product to do — not by who they are demographically or what they currently do behaviorally.

The premise, from Clayton Christensen and Theodore Levitt before him: people don’t buy products; they “hire” products to make progress on a job they care about. Find the job, and you’ve found the real segment. Two 35-year-old urban professionals with identical income who use your app three times a week might be hiring it for entirely different jobs: one to feel less anxious after work, the other to prepare for a wedding photo. Those are different segments — even though every other dimension says they’re the same.

Confusion to clear up: This is easily mixed up with Behavioral segmentation.

Behavioral = what they DO (the observable actions: 3 sessions/week, used feature X).

Needs-based = what they’re TRYING TO ACCOMPLISH (the underlying job: get back in shape before a milestone). Two users in the same behavioral cohort can be in different needs-based segments. Two users in the same needs-based segment can be in different behavioral cohorts.

Why it matters

Needs-based segmentation has two structural advantages over the other dimensions:

Jobs are more stable than identities or behaviors. A customer’s age, company size, and tech stack change. A customer’s job-to-be-done (“show my team I’m shipping faster”) stays roughly the same for years. Segments built on jobs survive longer.
The same person hires different products for different jobs. This shifts the competitive question from “who competes with us?” to “what are we being hired against?” — which is usually a much more useful framing.

It also reframes a lot of product and marketing conversations:

Roadmap — “which jobs are we under-serving?” beats “which features are we missing?”
Positioning — “the job we’re best at vs. our competitors” replaces “the feature we have that they don’t.”
Onboarding — different jobs mean different first-five-minutes flows for the same product.
Pricing — willingness-to-pay is anchored to the value of the job done, not the cost of the product.
Expansion — adjacent jobs (not adjacent features) are the highest-leverage expansion paths.

Core concepts

The job statement

A canonical JTBD job statement has a specific structure:

When [context / situation], I want to [outcome I’m trying to achieve], so I can [deeper motivation / progress I want to make].

Each clause is doing work:

“When” sets the trigger — the specific moment of need. This is what makes a job real instead of abstract. (“When the team grows past 10 people” is real; “to be productive” is not.)
“I want to” captures the functional outcome — the thing the user actively wants accomplished.
“So I can” ladders up to the emotional and social progress — why the outcome matters. This is what differentiates the job from a feature request.

A few examples:

When the team grows past 10 people and Slack starts losing context, I want to put our work into a shared workspace, so I can stop being the person who answers the same question three times a week.

When my partner and I have an anniversary trip coming up in 8 weeks, I want a structured workout plan I can fit in 30 minutes, so I can feel confident in photos without wrecking our normal routine.

If you can’t write a job statement this specifically about your customer, you haven’t done enough interviews yet.

The Four Forces of Progress

Why does a customer switch to your product? Why do they not? The JTBD answer: at the moment of decision, four forces compete:

flowchart LR
    A[Push of the current<br/>situation] --> D{Switch?}
    B[Pull of the<br/>new solution] --> D
    C[Anxiety about<br/>the new solution] --> D
    E[Habit / inertia of<br/>the current solution] --> D
    D -->|Yes| F[Hired your product]
    D -->|No| G[Stays with current solution]

The first two forces FOR change:

Push — how much does today’s solution hurt? (Slack is losing context faster than I can keep up.)
Pull — how attractive is the new solution? (A unified workspace would solve this in one move.)

The two forces AGAINST change:

Anxiety — what could go wrong? (What if my team rejects the new tool? What if data gets lost in migration?)
Habit — what’s the inertia of the current solution? (We have 2 years of conversation history in Slack we don’t want to leave behind.)

A customer “hires” your product only when Push + Pull > Anxiety + Habit. Most marketing and product teams over-invest in Pull (features) and under-invest in reducing Anxiety (proof, migration ease, switching guarantees) and Habit (data-import tooling, hybrid usage).

Outcome-Driven Innovation (ODI) — the rigorous version

Tony Ulwick’s ODI methodology operationalizes JTBD with outcome statements — measurable phrasings of “what ‘done’ looks like.” The structure:

[direction of improvement] + [unit of measure] + [object of control] + [contextual clarifier]

Example: “Minimize the time it takes to onboard a new team member to a shared knowledge base, regardless of their prior tooling.”

Each customer rates each outcome on importance and satisfaction. The gap (importance minus satisfaction) is the opportunity score. High-importance, low-satisfaction outcomes are your highest-leverage segments to serve.

This is more rigorous than the four-forces model and worth the investment if you have ≥30 interview transcripts to mine. For early-stage teams, the simpler job-statement + four-forces lens is usually sufficient.

Data sources

Source	What you get	Notes
Customer-discovery interviews	The job statements and the four-forces narrative, in customers’ own words	The gold-standard source — see Customer Discovery Interviews
Win/loss interviews	Why customers hired or didn’t hire your product — the forces in revealed form	Cross-link with Win/Loss Analysis
Sales call recordings	The job language prospects use during evaluation	Use Gong, Chorus, Fathom, etc.; tag for job-statements
Support and churn tickets	The jobs you failed to do well; the un-served jobs	Free; underrated; see Voice of Customer
Switch / hire interviews	Specifically why a customer switched to you, and from what	The most direct evidence of Push and Habit forces
Job-statement surveys (e.g. PSST templates)	Quantification once jobs are identified qualitatively	Use after qualitative work, never before

You cannot run needs-based segmentation from quantitative data alone. It requires qualitative depth — typically 12–25 customer interviews per major job hypothesis — before any quantification is meaningful.

How to build a needs-based segmentation — step by step

Conduct 12–25 customer-discovery interviews (existing customers, ideally split across heavy and light users). Use open prompts that surface the moment of hire (“Tell me about the day you decided to start using this”) rather than feature questions. See Customer Discovery Interviews for the script.
Transcribe and tag. Code each transcript for: trigger event, outcome the customer wanted, deeper motivation, current alternative, the four forces, language verbatim.
Cluster by job, not by customer. Two customers can share a job; one customer can have multiple jobs. The unit of analysis is the job-customer pair, not the customer alone.
Write job statements in the canonical “When X, I want to Y, so I can Z” form. Each cluster should yield one job statement everyone on the team can recite.
Test stability across customers. A good job statement should describe the situation of 3+ different customers without needing to be customized. If you have to change the words for every customer, the job is too narrow.
Score each job for opportunity — importance to the customer × dissatisfaction with current solution × addressable market size. High on all three = priority segment.
Name each needs-based segment by the job, not by the customer. “Stop-losing-context-as-the-team-grows” segment is meaningful; “Founders & ICs” is just a persona.
Cross-validate against firmographic / behavioral data. A real needs-based segment usually shows up as a concentration of one firmographic or behavioral profile, but not perfect overlap. Pure overlap means you’ve rediscovered the firmographic dimension.

Templates

Needs-based segment canvas

For each needs-based segment, capture:

Segment name (job-based):
  "[short, recognizable label of the job — e.g. 'Stop losing context']"

Job statement:
  When [trigger / situation],
  I want to [outcome they're trying to achieve],
  so I can [deeper motivation / progress they want].

Customer language (verbatim phrases from interviews):
  - "..."
  - "..."
  - "..."

Current alternatives they hire instead:
  - [tool / process / workaround they're currently using]
  - [what they consider switching from / to]

Four Forces of Progress for this job:
  Push (current pain):           [what hurts today]
  Pull (your value):             [what makes your product attractive]
  Anxiety (what could go wrong): [the fears they need addressed]
  Habit (inertia):               [what they'd lose by switching]

Opportunity score:
  Importance (1-10):           [how much this job matters to them]
  Satisfaction today (1-10):    [how well current solution serves it]
  Opportunity (importance - satisfaction): [_]

Demographic / firmographic overlap:
  [the typical profile of customers in this segment — but note: this is
   description, not definition]

Measurable success criteria (when the job is "done"):
  - [outcome 1]
  - [outcome 2]

Validation:
  Customers interviewed in this segment:     [N]
  Distinct customers using this language:     [N]
  Closed deals citing this job as primary:    [N]

Four-forces worksheet for one switch decision

When you’re trying to understand why one customer switched (or didn’t), fill this:

Customer:                [name / company]
Decision date:           [YYYY-MM-DD]
What they hired:         [your product / a competitor / no change]
What they fired:         [previous solution / status quo]

PUSH of the current situation
  - [pain point 1, in their words]
  - [pain point 2, in their words]
  - Strength of push (1-10): __

PULL of the new solution
  - [feature / benefit that pulled them]
  - [demo moment that landed]
  - Strength of pull (1-10): __

ANXIETY about the new solution
  - [fear / concern raised in evaluation]
  - [what would have to be true for them not to worry]
  - Strength of anxiety (1-10): __

HABIT / inertia of the current solution
  - [data, workflows, relationships keeping them in the current solution]
  - Strength of habit (1-10): __

Net switch force:  (Push + Pull) - (Anxiety + Habit) = __
Decision actually made: [why this matched / didn't match the score]

Metrics to track

Activation rate by needs segment — % of new users who reach the job-specific activation milestone in the activation window. The activation milestone is different per segment: for “stop losing context” it’s the first multi-person doc; for “look organized to investors” it’s the first dashboard or summary export.
Retention by needs segment — 90-day retention. Best vs. worst needs segment should differ by 15–25 percentage points at day-90 if the segmentation is doing real work.
Expansion rate by segment — % of customers in segment buying an adjacent product / tier within 12 months. JTBD-aligned expansion paths usually outperform feature-aligned expansion paths by 2–3x.
NPS by needs segment — needs-segment-aligned customers score higher than mismatched ones. Healthy delta between aligned and mismatched: +20 NPS points.
Job-language match in inbound — % of inbound demo requests or support tickets that use language tagged for a specific need segment. Floor for “we’ve made this job legible”: 40%.
Switch interview coverage — % of new paying customers you’ve completed a four-forces interview with within 30 days. Floor: 20%. Below that, your understanding of why customers actually hire you will rot.

Worked examples

SaaS workspace (B2B)

The team runs 18 discovery interviews on existing paid workspaces. They find that the same firmographic segment (10–49 person seed-to-Series-A tech teams) actually hires the product for three distinct jobs, and each job needs different first-week activation, different sales narrative, and different expansion path:

Segment N1: "Stop losing context"
  Job statement:  "When the team grows past 10 people and important
                   decisions are getting lost in Slack threads, I want
                   to consolidate decisions and context into one place,
                   so I can stop being the person who answers the same
                   question three times."
  Customer says:  "I can't find what we agreed last sprint."
                  "Context dies in DMs."
  Currently hires: Notion + Slack + (scattered Google Docs)
  Four forces:    Push high (team growth is real and recent),
                  Pull high (clear value),
                  Anxiety moderate (will the team adopt?),
                  Habit moderate (years of Slack history).
  Onboarding focus: Multi-user invite + first shared doc in week 1
  Expansion path: Docs → Tasks → Wiki

Segment N2: "Look organized to investors"
  Job statement:  "When we're approaching a board meeting or investor
                   update and I want to demonstrate operational maturity,
                   I want a place where our metrics, decisions, and
                   roadmap are visibly organized, so I can spend the
                   meeting on strategy instead of on context-setting."
  Customer says:  "My board can see we're not just winging it."
                  "I need to look like we have our act together."
  Currently hires: Google Docs + spreadsheets + screenshots in slides
  Four forces:    Push moderate (cyclical: board meetings),
                  Pull high if framed right (status / credibility),
                  Anxiety low (low-stakes adoption),
                  Habit low (no real incumbent for this job).
  Onboarding focus: Dashboards + investor-update template in week 1
  Expansion path: Dashboards → Reporting → Read-only investor seats

Segment N3: "Ship faster as the team scales"
  Job statement:  "When we're trying to ship more from the same team
                   without adding meetings, I want a workspace that
                   tracks who's doing what without making me chase
                   anyone, so I can keep velocity flat as headcount grows."
  Customer says:  "I want to know who's blocked without asking."
                  "Standups are wasting an hour a week."
  Currently hires: Linear or Jira + Slack + standups
  Four forces:    Push high (velocity matters acutely),
                  Pull high if integrated with code/PRs,
                  Anxiety high (PM/eng workflows already in place),
                  Habit very high (engineering tooling is sticky).
  Onboarding focus: Tasks + integrations + sprint view in week 1
  Expansion path: Tasks → Roadmap → Time-tracking

Same product, same firmographic segment — three needs-based segments. Each gets a different homepage variant, different trial-onboarding flow, different sales narrative, and different expansion playbook. This is the kind of decision tree that demographic or firmographic segmentation alone cannot produce.

Consumer fitness app (B2C)

For the consumer fitness app, the same demographic profile (28–44 urban professional) reveals three distinct jobs:

Segment N-A: "Feel less anxious after work"
  Job:  When I get home stressed from work and my partner has already
        had a hard day, I want a short workout that resets my mood,
        so I can be present for the evening instead of grinding through it.
  Hires: A 10-15 minute mood-focused session; meditation+movement combo.
  Currently hires: Long walks, occasional yoga apps, sometimes just wine.
  Activation:    First "mood-reset" session within 3 days of signup.
  Expansion:     Mindfulness add-on, sleep content.

Segment N-B: "Lose weight by a specific date"
  Job:  When I have a milestone in 8-16 weeks (wedding, reunion, trip),
        I want a structured plan with measurable progress, so I can show up
        confident on that date.
  Hires: Calorie tracker + structured 8-week plan + before/after photos.
  Currently hires: Crash diets, gym memberships, dedicated trainers.
  Activation:    Plan started + goal photo uploaded in week 1.
  Expansion:     Coach add-on, accountability community.

Segment N-C: "Stay in shape while traveling"
  Job:  When I'm traveling for 3-10 days, I want a fitness option that
        works in a hotel room with no equipment, so I can come back without
        feeling I lost ground.
  Hires: Bodyweight programs; offline-downloadable content; short sessions.
  Currently hires: Hotel gyms (when adequate), nothing (when not).
  Activation:    Saved workout for offline use; completed first travel session.
  Expansion:     Mobility, recovery content.

Note that segments N-A and N-B may apply to the same person at different points in their year. JTBD allows for that — “the same human can be in different needs segments simultaneously.” That’s a feature, not a bug; it implies your product can be re-hired multiple times by the same customer for different jobs, which is the highest-LTV pattern in B2C.

Common pitfalls

Confusing features sought with jobs. “Wants a darkmode toggle” is a feature. “Wants the product to feel like part of their environment” is a job. The job is the level that survives the feature being shipped or removed.
Jobs too narrow. “Press the upload button” is not a job; it’s a task within a job. If you can complete the job in 2 minutes with a single click, you’ve identified a feature, not a job.
Jobs too broad. “Be productive” is not a job; it’s a wish. A real job has a specific trigger and a measurable outcome. If everyone in your TAM has the job, the job is too broad.
Inventing jobs without interviews. Job statements written in a strategy offsite without customer transcripts behind them are wishlists, not segments. The discipline of needs-based segmentation is grounding every job statement in 3+ customer interviews using that language.
One person = one job assumption. The same customer can hire your product for multiple jobs over time. Forcing one customer into one segment loses the most valuable insight: which customers are cross-job buyers (highest LTV).
Skipping the four forces. Naming jobs without understanding why customers don’t switch leaves half the strategy on the table. Anxiety and Habit are at least as important as Push and Pull.
Confusing the segment name with the segment definition. “Stop losing context” is a label; the definition is the job statement plus the customer language plus the four forces. If a teammate can’t recite all three, they don’t understand the segment.