DOMAIN:CREATIVE — AUDIENCE_RESEARCH¶
OWNER: valentijn (Brand Strategist) ALSO_USED_BY: aimee (client scoping), joshua (market research), tjarda (brand management), margot (client comms) UPDATED: 2026-03-28 SCOPE: persona development, jobs-to-be-done, empathy mapping, behavioral segmentation, audience insight extraction, survey design
CORE_PRINCIPLE¶
You cannot position a brand without understanding the audience it serves. Audience research is not optional — it is the FOUNDATION of all brand strategy.
RULE: never assume audience needs — validate with research RULE: research the AUDIENCE, not just the market. Markets are abstract; audiences are people with specific frustrations. RULE: behavioral data beats demographic data. What people DO reveals more than what they ARE. RULE: audience research feeds directly into the Positioning Canvas (see domains/creative/brand-strategy.md)
PERSONA_DEVELOPMENT¶
WHAT_A_PERSONA_IS¶
A persona is a research-based archetype representing a segment of the target audience. It synthesizes behavioral patterns, goals, frustrations, and decision-making processes into a usable reference for strategy and creative work.
WHAT_A_PERSONA_IS_NOT: - Not a demographic profile ("35-year-old male in Eindhoven") - Not a fictional character you invented in a workshop - Not a single real person - Not a wish list of who you want your customers to be
PERSONA_TEMPLATE¶
PERSONA: [Name — use a realistic name, not "Tech-Savvy Tom"]
SEGMENT: [behavioral segment this persona represents]
ROLE / CONTEXT
Job title: [or life situation]
Company size: [if B2B]
Industry: [if relevant]
Decision authority: [sole decision-maker, influencer, or part of committee]
BEHAVIORAL PROFILE
Information sources: [where they learn about solutions]
Decision timeline: [impulse, weeks, months]
Risk tolerance: [conservative, moderate, adventurous]
Tech comfort: [builds own tools, uses tools, avoids tools]
Buying triggers: [what event causes them to actively search for a solution]
GOALS
Primary: [the main thing they want to achieve]
Secondary: [supporting goals]
Aspirational: [what success looks like to them long-term]
FRUSTRATIONS
Primary: [the biggest pain point related to your category]
Secondary: [other relevant frustrations]
Current workarounds: [how they solve the problem today without you]
JOBS TO BE DONE (see JTBD section below)
Functional job: [what they need to accomplish]
Emotional job: [how they want to feel]
Social job: [how they want to be perceived]
DECISION CRITERIA
Must-have: [non-negotiable requirements]
Nice-to-have: [preferences that influence but don't decide]
Deal-breaker: [what causes immediate rejection]
MESSAGING IMPLICATIONS
Lead with: [what resonates most]
Avoid: [what turns them off]
Proof they need: [what evidence convinces them]
PERSONA_DEVELOPMENT_PROCESS¶
STEP 1: GATHER RAW DATA - Customer interviews (5-8 per segment minimum) - Support ticket analysis (patterns in complaints and requests) - Sales conversation patterns (what questions do prospects ask?) - Website/product analytics (behavior flows, drop-off points) - Competitive review comments (what do competitor customers complain about?)
STEP 2: IDENTIFY BEHAVIORAL CLUSTERS - Group by BEHAVIOR, not demographics - Look for patterns in goals, frustrations, and decision-making - Typically 3-5 distinct personas cover 80%+ of the audience
STEP 3: SYNTHESIZE INTO PERSONAS - Each persona represents a behavioral cluster - Use real quotes from research (anonymized) - Validate with the team who interacts with customers
STEP 4: VALIDATE AND ITERATE - Test personas against new customer interactions - Update when patterns shift - Retire personas that no longer match reality
RULE: personas based on assumptions are WORSE than no personas — they create false confidence RULE: update personas at least annually or when the market shifts significantly
JOBS_TO_BE_DONE_FRAMEWORK¶
Source: Clayton Christensen, "Competing Against Luck" (2016); Tony Ulwick, "What Customers Want" (2005).
CORE_CONCEPT¶
People don't buy products. They "hire" products to do a job. Understanding the job reveals the real competitive set and the real value proposition.
THE_MILKSHAKE_EXAMPLE (Christensen): A fast-food chain wanted to sell more milkshakes. Demographics said "target 25-40 males." JTBD research revealed: 40% of milkshakes were bought at 6:30am by commuters who needed something to make a boring drive interesting, keep them full until lunch, and could be consumed one-handed. The "competition" was not other milkshakes — it was bananas, bagels, and boredom.
JOB_STATEMENT_FORMAT¶
EXAMPLES:
When I realize my business needs custom software,
I want to find a reliable development partner I can afford,
so I can get the tool I need without risking my company's budget.
When I'm evaluating software agencies,
I want to understand exactly what I'll get and what it will cost,
so I can make a confident decision without technical expertise.
THREE_DIMENSIONS_OF_JOBS¶
| Dimension | Question | Example |
|---|---|---|
| Functional | What do they need to accomplish? | "Get custom software built" |
| Emotional | How do they want to feel? | "Confident I made the right choice" |
| Social | How do they want to be perceived? | "Smart business owner who found an innovative solution" |
RULE: emotional and social jobs are often MORE important than functional jobs in brand positioning RULE: the functional job defines the category; the emotional/social jobs define the positioning
OUTCOME_DRIVEN_INNOVATION (Ulwick)¶
For each job, identify desired outcomes:
DIRECTION + MEASURE + OBJECT OF CONTROL + CONTEXT
Minimize the TIME it takes to GET a cost estimate for custom software.
Minimize the RISK of choosing an agency that delivers poor quality.
Increase the CONFIDENCE that my data is handled securely.
Rate each outcome on IMPORTANCE and SATISFACTION: - High importance + low satisfaction = OPPORTUNITY (unmet need) - High importance + high satisfaction = TABLE STAKES (must match, can't differentiate) - Low importance + high satisfaction = OVER-SERVED (potential to reduce and cut cost) - Low importance + low satisfaction = DON'T BOTHER
EMPATHY_MAPPING¶
Source: Dave Gray, XPLANE (2010); updated by Stanford d.school.
EMPATHY_MAP_TEMPLATE¶
EMPATHY MAP — [PERSONA NAME]
SAYS
[Direct quotes from interviews, support tickets, reviews]
[What they tell others about the problem]
[How they describe their needs]
THINKS
[What occupies their mind — concerns, aspirations]
[What they think but don't say out loud]
[Internal monologue during decision-making]
FEELS
[Emotional state — anxious, excited, frustrated, hopeful]
[What worries them about this category]
[What excites them about potential solutions]
DOES
[Observable behavior — what actions do they take]
[Workarounds they currently use]
[How they research and evaluate options]
PAINS
[Fears, frustrations, obstacles]
[What could go wrong from their perspective]
[What has gone wrong in the past]
GAINS
[What success looks like]
[What would make them delighted (not just satisfied)]
[What would they tell a friend about a great experience]
HOW_TO_FILL_AN_EMPATHY_MAP¶
BEST: real customer interviews and observation GOOD: support ticket analysis, sales call transcripts, review mining ACCEPTABLE: competitive review analysis, industry forums, social listening BAD: internal brainstorming without data WORST: making it up and calling it "intuition"
RULE: every cell in the empathy map should be traceable to a real data source RULE: "thinks" and "feels" are the hardest to source — use behavioral indicators and direct interview probing
BEHAVIORAL_SEGMENTATION¶
WHY_BEHAVIORAL_OVER_DEMOGRAPHIC¶
Demographic segmentation groups by WHO people are (age, gender, income, location). Behavioral segmentation groups by WHAT people do (purchase patterns, usage, decision style).
| Approach | Example Segment | Strategy Usefulness |
|---|---|---|
| Demographic | "Male, 35-50, Netherlands, €100K+ income" | Low — describes millions of very different people |
| Behavioral | "Risk-averse first-time buyer who researches for 3+ months before deciding" | High — directly informs messaging and channel strategy |
BEHAVIORAL_SEGMENTATION_DIMENSIONS¶
| Dimension | Question | Segments |
|---|---|---|
| Purchase occasion | When/why do they buy? | Planned vs. triggered, seasonal vs. ongoing |
| Benefits sought | What value do they prioritize? | Price, quality, speed, security, status |
| Usage rate | How much/often do they use? | Heavy, moderate, light, non-user |
| Loyalty status | How committed are they? | Loyal, switcher, new, lapsed |
| Readiness stage | Where in the buying journey? | Unaware, aware, considering, deciding, buying |
| Decision style | How do they choose? | Analytical, emotional, social proof, authority trust |
| Risk profile | How do they handle uncertainty? | Risk-averse, balanced, risk-tolerant |
SEGMENTATION_PROCESS¶
- Choose 2-3 behavioral dimensions most relevant to the category
- Gather behavioral data (analytics, interviews, transaction history)
- Cluster respondents by behavioral similarity
- Profile each cluster (add demographics AFTER behavioral grouping)
- Size each segment (is it large enough to target?)
- Evaluate attractiveness (can you reach them? Will they pay?)
- Select primary and secondary segments
- Build personas from segments
RULE: segment by behavior FIRST, then overlay demographics for media targeting RULE: the most valuable segment is not always the largest — it's the most underserved AND reachable
EXTRACTING_INSIGHTS_FROM_COMPETITIVE_ANALYSIS¶
COMPETITOR_CUSTOMER_RESEARCH¶
You can learn about your audience by studying how they interact with competitors.
SOURCES: - App store reviews — real users describe what they love and hate. Filter by 2-3 star reviews for the most nuanced feedback. - G2/Capterra/Trustpilot reviews — B2B gold. Sort by "most helpful" for substantive reviews. - Social media comments — how people talk about competitors publicly. - Reddit/forum threads — unfiltered opinions, comparison requests, complaints. - Competitor support forums — what problems are customers having? - Job listings — if a competitor is hiring for a specific role, it reveals strategic priorities.
ANALYSIS_METHOD¶
- Collect 50-100 reviews/comments across competitors
- Code each for: praise themes, complaint themes, unmet needs, switching reasons
- Tally frequency of each theme
- Identify themes with HIGH frequency + HIGH emotional intensity = opportunity
- Cross-reference with your own positioning — do you address these unmet needs?
INSIGHT_TEMPLATE:
INSIGHT: [one-sentence observation]
EVIDENCE: [what data supports this — source, frequency, intensity]
IMPLICATION: [what this means for brand positioning]
ACTION: [specific recommendation]
SURVEY_DESIGN_FOR_BRAND_PERCEPTION¶
SURVEY_DESIGN_PRINCIPLES¶
- START WITH RESEARCH QUESTIONS — What decisions will this survey inform? If you can't name the decision, don't ask the question.
- KEEP IT SHORT — 5-10 minutes maximum. Every additional minute drops completion rate 5-10%.
- ORDER MATTERS — General before specific. Unaided before aided. Behavior before opinion.
- AVOID LEADING QUESTIONS — "How much do you love our brand?" is not research. "How would you describe your feelings toward [brand]?" is research.
- MIX QUESTION TYPES — Open-ended for discovery. Closed-ended for measurement. Scales for tracking.
BRAND_PERCEPTION_SURVEY_STRUCTURE¶
SECTION 1: CATEGORY BEHAVIOR (warm-up, establishes context)
- How often do you [category behavior]?
- What solutions do you currently use for [category need]?
- What prompted you to start looking for [category solution]?
SECTION 2: UNAIDED AWARENESS
- When you think of [category], which brands come to mind? (open-ended)
- Which of these have you used in the past 12 months? (open-ended)
SECTION 3: AIDED AWARENESS AND PERCEPTION
- Which of the following brands have you heard of? [list including yours]
- For each brand you know, select the words that describe it: [attribute list]
- On a scale of 1-7, how likely would you be to consider [brand] for [need]?
SECTION 4: BRAND ASSOCIATIONS (for your brand specifically)
- What three words come to mind when you think of [your brand]? (open-ended)
- How would you describe [your brand] to a friend? (open-ended)
- What does [your brand] do better than alternatives? (open-ended)
- What does [your brand] do worse than alternatives? (open-ended)
SECTION 5: NEEDS AND PRIORITIES
- Rank the following factors by importance when choosing [category]: [list]
- What is the single most important thing a [category] brand could offer you?
SECTION 6: DEMOGRAPHICS (at the end, not the beginning)
- Role, company size, industry, etc.
SURVEY_ANTI_PATTERNS¶
- Asking double-barreled questions ("Is our product fast AND reliable?")
- Using jargon the respondent might not understand
- Having no "I don't know" or "Not applicable" option
- Asking for opinions on things people have no experience with
- Surveying only existing customers (survivor bias)
- Making all questions required (causes abandonment or random answers)
SAMPLE_SIZE_GUIDANCE¶
| Confidence Level | Margin of Error | Required Sample |
|---|---|---|
| 95% | ±10% | ~100 |
| 95% | ±5% | ~385 |
| 95% | ±3% | ~1,068 |
For brand perception: 200-400 responses is a practical sweet spot for SME budgets.
RESEARCH_DELIVERABLES¶
AUDIENCE_INSIGHT_REPORT_STRUCTURE¶
1. EXECUTIVE SUMMARY
- Key findings (3-5 bullets)
- Primary audience definition
- Strategic implications
2. METHODOLOGY
- Research methods used
- Sample size and composition
- Limitations and caveats
3. AUDIENCE SEGMENTS
- Behavioral segments identified
- Segment sizes and characteristics
- Primary and secondary segment selection rationale
4. PERSONA PROFILES
- Full persona for each primary segment
- Empathy map for each persona
- JTBD for each persona
5. KEY INSIGHTS
- Numbered insights with evidence and implications
- Unmet needs mapped to positioning opportunities
6. COMPETITIVE PERCEPTION
- How the audience perceives key competitors
- Perceptual map
- White space opportunities
7. RECOMMENDATIONS
- Positioning implications
- Messaging priorities
- Further research needs
RESEARCH_QUALITY_CHECKLIST¶
Before using research to inform strategy:
[ ] SAMPLE: is the sample representative of the target audience (not just convenient)?
[ ] SIZE: is the sample large enough for the conclusions drawn?
[ ] RECENCY: is the data recent enough to reflect current attitudes?
[ ] BIAS: have leading questions, selection bias, and confirmation bias been checked?
[ ] TRIANGULATION: do multiple data sources point to the same conclusion?
[ ] BEHAVIORAL: does the research include WHAT PEOPLE DO, not just what they say?
[ ] COMPETITIVE: does the research include perceptions of alternatives, not just your brand?
[ ] ACTIONABLE: can the findings be translated into specific strategic decisions?
ANTI_PATTERNS¶
ANTI_PATTERN: building personas in a workshop without talking to real customers FIX: personas MUST be grounded in research. No data, no persona.
ANTI_PATTERN: creating 10+ personas that nobody can remember FIX: 3-5 personas cover most audiences. If you have more, your segmentation is too granular.
ANTI_PATTERN: demographic-only segmentation ("millennials who earn over €50K") FIX: lead with behavioral segmentation. Demographics are a targeting overlay, not a strategy foundation.
ANTI_PATTERN: asking customers what they want (they often don't know) FIX: observe what they DO and ask about their PROBLEMS. Henry Ford: "If I'd asked people what they wanted, they'd have said a faster horse."
ANTI_PATTERN: conducting research once and treating it as permanent truth FIX: audiences evolve. Refresh persona research annually or when market dynamics shift.
ANTI_PATTERN: ignoring competitor customers as a research source FIX: competitor reviews, forums, and social mentions are free research gold.
ANTI_PATTERN: survey questions that confirm what you already believe FIX: include questions that could disprove your hypothesis. Research that can only confirm is not research.
ANTI_PATTERN: treating the loudest customers as representative FIX: vocal customers are often power users or complainers. They matter, but they're not the whole audience.
AUTHORITATIVE_SOURCES¶
| Source | Work | Key Concept |
|---|---|---|
| Clayton Christensen | Competing Against Luck | Jobs-to-be-done theory |
| Tony Ulwick | What Customers Want / Jobs to Be Done | Outcome-driven innovation, job mapping |
| Alan Klement | When Coffee and Kale Compete | JTBD for product/brand strategy |
| Dave Gray | Gamestorming / XPLANE | Empathy mapping methodology |
| Byron Sharp | How Brands Grow | Category entry points, mental availability |
| Jenni Romaniuk | How Brands Grow Part 2 | Brand salience, distinctive assets |
| Indi Young | Mental Models | Research-based mental model mapping |
| Erika Hall | Just Enough Research | Practical research methods, anti-patterns |
| Steve Portigal | Interviewing Users | Customer interview techniques |
CROSS_REFERENCES¶
STRATEGY: domains/creative/brand-strategy.md — positioning canvas that research feeds into BRIEF: domains/creative/strategic-brief.md — audience definition in creative briefs COMPETITIVE: domains/creative/competitive-intelligence.md — competitive research methods BRAND_MGMT: domains/marketing/brand-management.md — audience understanding for brand consistency CLIENT_COMMS: domains/client-communications/ — how audience insights inform client interactions
Audience research domain loaded. Research precedes all strategic decisions.