ROI【🔏Classified File】No. X044 | What is Persona Analysis

Detective's Memo: The ROI Detective Agency has distilled the essential customer understanding methodology "Persona Analysis" from countless marketing failure cases. Many practitioners categorize customers by mere attributes like "women in their 20s" or "executives," mass-producing messages that resonate with no one. Why does "trying to reach everyone" result in reaching no one? Why does designing a fictional individual in precise detail—"Taro Tanaka (38, sales manager at mid-sized company, married with 2 children, annual income ¥5.8M, spends 3 hours daily on Excel, struggling with efficiency)"—generate messages that pierce through to thousands of real people? Amazon, Apple, and Airbnb practice this technique of constructing "human narratives" invisible to data alone. Rather than statistical "men in their 30s," depicting customers as "individuals with proper names"—this resolution of fiction decisively transforms marketing precision. Identify the true nature of this paradoxical success formula: escape the trap of universal appeal, speak to one person alone, and thereby reach everyone.

What is Persona Analysis - Case Overview

Persona Analysis, formally "a methodology for visualizing true customer needs by designing fictional archetypal customer profiles in detail, thereby unifying product development, marketing, and design decisions around customer-centricity," was systematized as a user understanding theory by Alan Cooper in 1999. Rather than mere attribute classification like "women in their 20s" or "executive layer," it is recognized among clients as a fictional customer profile set in detail like a real person—name, age, occupation, family structure, values, concerns, behavioral patterns, information-gathering methods—all specified. However, in actual practice, it is often dismissed as "imaginary characters created arbitrarily," with most companies failing to understand its true strategic value: scientific design through integration with quantitative data, prioritization among multiple personas, and the paradoxical marketing effect that "speaking to one person reaches everyone."

Investigation Memo: Personas are not "fantasies" but "crystallized abstractions of reality." Why does "Taro Tanaka (34, sales position, struggling with Excel tasks, wakes at 6am daily, reads business books during commute)" serve better as product development criteria than statistical data "men in their 30s, income ¥5M"? We must illuminate this foundational technology of customer-centric design that clarifies "for whom is minimum" in MVP and specifies the "friend image who would recommend" in NPS measurement.

Basic Structure of Persona Analysis - Evidence Analysis

Fundamental Evidence: The resolution of one human determines marketing precision

Persona vs Target vs Segment

Segment:

Definition: Statistical customer classification
Example: "Males 30-39, income $40K-60K, residing in Tokyo"

Characteristics:
- Mechanical classification by attributes
- Calculated from mass data
- Objective, quantitative
- No visible humanity

Problems:
- Diverse needs even within same segment
- Cannot serve as decision-making criteria
- Cannot empathize

Target:

Definition: Marketing audience narrowing
Example: "Sales department managers at mid-sized companies"

Characteristics:
- More specific than segment
- Specifies occupation and position
- Broad framework for marketing direction
- Still abstract

Problems:
- Diversity within "managers" remains invisible
- Specific concerns and behaviors unclear
- Weak as message creation criteria

Persona:

Definition: Fictional yet detailed "individual person"
Example: "Taro Tanaka (38, sales manager, married with 2 children, 
         income ¥5.8M, 3 hours daily Excel work, seeking efficiency tools,
         wakes at 6am, 30-minute commute reading business books,
         struggles with subordinate development, 
         plays soccer with kids on weekends)"

Characteristics:
- Has proper name
- Specific down to lifestyle
- Enables empathy
- Serves as decision-making criteria

Effects:
- Can judge by "How would Tanaka-san feel?"
- Entire team shares same customer image
- Messages become concrete

Essential Difference:

Segment: "What attributes" (What)
Target: "Who to aim for" (Who)
Persona: "How does this person live" (How & Why)

Components of Persona

Basic Attributes (Demographics):

Essential Items:
- Name (proper noun crucial)
- Age, gender
- Occupation, position
- Income range
- Residence
- Family structure
- Education

Why Important:
Even fictional, calling someone "Taro Tanaka-san"
enables entire team to envision same person

Psychological Attributes (Psychographics):

Deep-dive Items:
- Values, beliefs
- Personality traits
- Lifestyle
- Hobbies, interests
- Goals, ambitions
- Anxieties, fears

Example:
"Want to return home early through efficiency to increase family time"
"Want to become a manager subordinates rely on"
"Anxious about learning cost of new tools"

Behavioral Patterns (Behavioral):

Observation Items:
- Daily time usage
- Information-gathering methods
- Purchase decision process
- Frequently used media/tools
- Decision-making criteria

Example:
"6am wake → business book on commute train →
 9am arrival → 3hrs Excel work → 2hrs meetings →
 7pm departure → play with children after home"

Challenges & Concerns (Pains):

Specification Items:
- Daily troubles
- Unachieved goals
- Stress factors
- Dissatisfaction, frustration

Example:
"Slow Excel work prevents overtime reduction"
"Can't secure time for subordinate guidance"
"No time to learn new tools"
"Pressure from superiors"

Goals & Desires (Gains):

What to Achieve:
- Short-term goals
- Long-term ambitions
- Ideal state
- Issues to resolve

Example:
"Leave on time to increase family time"
"Become manager subordinates trust"
"Improve department results through efficiency"
"Get promoted to increase income"

Information Sources & Touchpoints:

Where Information is Encountered:
- Frequently visited websites
- SNS usage status
- Trusted information sources
- References for purchase decisions

Example:
"Morning commute browsing NewsPicks"
"Business books referenced via Amazon reviews"
"Industry information gathering on LinkedIn"
"Colleague word-of-mouth most trusted"

Quotes & Voice:

Persona's Voice:
"If I spend 3 hours daily on Excel work,
 I can't do actual sales activities"

"New tools? More things to learn..."

"I want subordinates to work efficiently,
 but I need to set the example first"

Source of Realism:
Extracted from actual customer interviews

Evidence Analysis: The power of personas lies not in "detail" but in "realism." A resolution level where the entire team can judge "what would Tanaka-san think" enables customer-centric decision-making.

Persona Creation Process - Investigation Methods

Investigation Discovery 1: Airbnb's Persona Design Process

Case Evidence (Unicorn born from early-stage customer understanding):

Phase 1: Hypothesis Persona Failure (2008)

Initial Assumption:

"Young backpackers"
- Early 20s
- Budget travelers
- Using as hostel alternative
- Price-focused

Problem Discovery: - Actual users mostly 30s and older - Not "because it's cheap" but "seeking experience" - Value interaction with hosts - Assumed persona diverged from reality

Lesson: "Assumption-based personas are dangerous" → Scientific design based on data essential

Phase 2: Data-Driven Persona Construction (2009-2010)

Research Conducted:

Quantitative Research:
- 5,000 user behavior data analysis
- Booking patterns, stay duration, price range
- Repeat rate, review ratings
- Attribute data (age, occupation, residence)

Qualitative Research:
- 50 in-depth interviews
- Why chose Airbnb
- What experience sought
- Anxieties and delights

Ethnography:
- Accompanied actual stays
- Observed host conversations
- Recorded usage scenes, emotional changes

True Persona Discovered:

"Sarah Johnson"
- 34, graphic designer
- San Francisco resident, single
- Income $70K
- 3-4 international trips annually

Values:
"Want to interact with locals and experience local culture"
"Prefer local spots over tourist destinations"
"Tired of standardized hotel service"

Behavior:
- Thoroughly reads host profiles before travel
- Values reviews (especially detailed ones)
- Shares travel photos on Instagram
- Brags to friends "stayed at local's home"

Challenges:
"Safety concerns"
"Anxiety about host communication"
"First-use barrier"

This "Sarah" image was shared company-wide,
becoming the criterion for product development and marketing

Phase 3: Persona-Driven Decision Making (2010-2012)

Persona-Based Initiatives:

Product Improvements:
"Sarah values safety"
→ Strengthened identity verification for both hosts/guests
→ Enhanced review system
→ Introduced host guarantee program

Marketing:
"Sarah seeks experiences"
→ "Belong Anywhere"
→ Appealed experience value, not price comparison with hotels
→ Featured host interaction photos prominently

Communication:
"Sarah seeks detailed information"
→ Supported listing photo quality improvement
→ Host introduction text enrichment guide
→ Detailed surrounding area information

Results: - Surpassed 10 million nights by 2012 - 30s professionals like "Sarah" became main customers - Persona accuracy: 67% of actual main customers matched assumed persona

Phase 4: Multiple Persona Management (2013 onwards)

Diversification with Expansion:

Primary Persona (Most Important):
"Sarah" (experience-focused urban professional)
→ Company-wide top priority

Secondary Personas:
"Mike" (family trip father, 40s, cost-conscious)
→ Family-oriented features

"Lisa" (business traveler, 30s, convenience-focused)
→ Airbnb for Work expansion

Decision Criteria:
First ask "Would Sarah be happy?"
Address other personas within non-conflicting range

Investigation Discovery 2: Amazon's Persona "Empty Chair"

Case Evidence (Ultimate customer-centricity):

Jeff Bezos's Ritual:

An "empty chair" prepared in every important meeting

Whose chair is it?
→ "The Customer"

Constantly asking during meetings:
"What would the customer sitting in this empty chair
 think about this decision?"

Implicit Persona:

"The Busy Customer"
- Doesn't want to waste time
- Fatigued by too many choices
- Seeks simple, fast purchase experience
- Doesn't want to fail (values reviews)

This customer image becomes all decision-making criteria

Investigation Discovery 3: Spotify's Age-Based Persona Design

Case Evidence (Multiple persona prioritization management):

Main Personas Created:

"Emma" (16, high school student)
- Music = identity
- Wants to share with friends
- Enjoys discovering new artists
- Free plan user

"Tom" (28, marketer)
- BGM for commute and work
- Playlists matching mood
- Seeks ad-free
- Premium member

"Linda" (45, housewife)
- Listens during housework
- Likes nostalgic songs
- Uncomfortable with complex operations
- Family plan user

Prioritization:

Phase 1 (2008-2012): Emma focus
→ Youth acquisition, viral expansion

Phase 2 (2013-2017): Tom focus
→ Paid conversion, monetization

Phase 3 (2018-present): All persona support
→ Age-specific playlists, UI optimization

Power of Persona Analysis - Evidence Effectiveness

Power 1: Decision-Making Clarity

Without Persona:
"Should we add this feature?"
→ Discussion diverges, subjective pushing

With Persona:
"Would Tanaka-san use this feature?"
→ Clear judgment criteria, discussion converges

Power 2: Team Recognition Unity

Without Persona:
Member A: "For young people, right?"
Member B: "No, for businessmen"
→ Fragmented directions

With Persona:
Everyone: "Feature for Tanaka-san (38, sales manager)"
→ Unified customer image

Power 3: Deep Understanding Through Empathy

Statistics: "Men in 30s, income $50K"
→ Cold data, impossible to empathize

Persona: "Taro Tanaka-san, wakes 6am daily,
         struggles with Excel, wants to increase family time"
→ Empathizable, "want to help Tanaka-san"

Power 4: Sharp Marketing Messages

Universal Appeal:
"Boost productivity with efficiency tools!"
→ Resonates with no one

Persona-Directed:
"Turn 3 hours of daily Excel work into 30 minutes.
 Tanaka-san, return home on time to your family."
→ Pierces Tanaka-san
→ Reaches everyone like Tanaka-san

Limitations and Pitfalls of Persona Analysis - Investigation Traps

Pitfall 1: Imaginary Personas Created Without Research

❌ Bad Example:
"Some 30s salesman-ish person"
→ No data, no interviews
→ Creator's bias and assumptions

✅ Correct Method:
Quantitative data + qualitative interviews + observation
→ Abstraction and crystallization of real customers

Pitfall 2: Too Many Personas

Creating 10 personas
→ Ultimately unclear which to prioritize
→ Cannot serve as decision-making criteria

Recommended:
1-2 primary, 2-3 secondary maximum

Pitfall 3: Attributes Only, No Behavior or Psychology

❌ Shallow Persona:
"Taro Tanaka, 38, sales manager"
→ This alone cannot serve as judgment criteria

✅ Deep Persona:
"Taro Tanaka, 38, sales manager,
 struggles with 3hrs/day Excel work,
 wants to increase family time,
 anxious about new tool learning cost"
→ Usable for decision-making

Pitfall 4: No Regular Updates

Using 2020-created persona in 2025
→ Market and customers have changed
→ Outdated personas mislead judgment

Recommended:
Review and update every 6 months to 1 year

Pitfall 5: Ignoring Customers Outside Persona

Over-specializing for "Tanaka-san"
cutting off other important customer segments

Balance:
Prioritize primary persona
But consider secondary ones

Related Case Files - Investigation Line Intersections

X036_MVP (Minimum Viable Product) Persona clarifies "minimum for whom"

X040_NPS (Net Promoter Score) Persona concretizes the face of "the friend who would recommend"

X039_HEART (HEART Framework) Persona clarifies the "user" being measured

X022_AARRR (Pirate Metrics) Persona is the foundation for predicting customer behavior at each stage

X028_RCD (Record-Compare-Analyze Model) Persona defines "who" to record

Industry-Specific Persona Design Cases - Field Investigation Records

SaaS Industry: Slack

"Jessica the Designer"
- 28, startup employee
- Email fatigue, meeting fatigue
- Wants creative time
- Prefers simple, beautiful tools

Impact:
→ Design-focused UI
→ "Email killer" positioning
→ Rich emoji and customization features

E-commerce Industry: ZOZO

"Fashion-loving Aya"
- 24, apparel sales clerk
- Spends 30% of salary on clothes
- Information gathering via Instagram
- Try in store → purchase online

Impact:
→ ZOZOSUIT (size measurement) development
→ Instagram integration enhancement
→ Rich wearing image photos

Education Industry: Coursera

"Career-changing Mike"
- 35, sales position
- Wants IT skill acquisition
- Past self-study failure experience
- Learning while working

Impact:
→ Modular short-duration learning
→ Certificate issuance
→ Rich mobile support

Investigation Conclusion - ROI Detective's Verdict

The essence of Persona Analysis lies in "precision of fiction."

The abstraction "women in their 20s" moves no one, but the concrete "25-year-old Emi Tanaka (advertising agency employee, side blog operation, monthly ¥50K revenue goal, struggling with SEO, wakes 5am daily to write)" moves thousands like Emi.

Trying to reach everyone reaches no one. Speaking to one person reaches everyone with the same concerns.

This is the paradoxical power of Persona Analysis.

What matters is designing personas not as "imagination" but as "crystallized abstractions of reality" scientifically. Entities extracted from the accumulation of data and dialogue, representing the essence of someone real.

Airbnb's "Sarah," Amazon's "empty chair customer," Spotify's "Emma, Tom, Linda"—they are fictional, yet simultaneously represent the most real customer voices.

When discussion stalls in the meeting room, can you ask "What would Tanaka-san think?" Teams where this question arises naturally have achieved customer-centricity.

Fiction called persona most vividly reflects the reality called customer—this is marketing truth.

【ROI Detective Agency Classified File Series X044 Complete】

Case Closed