Market, Customer and Ecosystem Intelligence

Introduction

Product management begins with disciplined interpretation of the environment in which a product must create, deliver and capture value. A product may be technically impressive and still fail because it solves an unimportant problem, targets the wrong customer, enters an unattractive market, depends on inaccessible partners, or misjudges regulation and competitive response. Market, customer and ecosystem intelligence provides the evidence base for avoiding these failures.

Market intelligence concerns the size, structure, growth, economics and competitive conditions of a market. Customer intelligence concerns the people or organisations that experience a problem, make or influence a purchase, use a product and evaluate its outcomes. Ecosystem intelligence examines the wider network of suppliers, platforms, distributors, regulators, complementors, competitors and communities that influence product success. These forms of intelligence are interdependent: a customer need is strategically valuable only when it occurs in a reachable market and can be addressed within a viable ecosystem.

The product manager’s task is not simply to collect data. It is to convert incomplete and sometimes contradictory evidence into a defensible opportunity thesis, identify assumptions, test them, and make decisions under uncertainty.

Learning Objectives

By the end of this chapter, the learner should be able to:

1. Distinguish market, customer and ecosystem intelligence and explain their strategic relationship.
2. Design appropriate qualitative and quantitative market-research activities.
3. Segment markets and evaluate segments using attractiveness and accessibility criteria.
4. Apply Jobs-to-be-Done, personas and customer journey mapping to identify unmet needs.
5. Estimate TAM, SAM and SOM while stating assumptions and avoiding false precision.
6. Use PESTLE and ecosystem analysis to evaluate external dependencies and risks.
7. Integrate evidence into opportunity prioritisation and product strategy.
8. Recognise ethical, regulatory and methodological limitations in intelligence work.

1. The Intelligence System in Product Management

A useful intelligence system answers five linked questions:

Intelligence should be treated as a learning cycle rather than a one-time report. The product team forms a hypothesis, gathers evidence, interprets patterns, tests the riskiest assumptions, and updates its strategy. This approach reduces confirmation bias, in which researchers selectively notice information supporting an existing idea.

Three distinctions are especially important:

• Data are recorded observations, such as interview statements, search trends or usage events.
• Insight is an evidence-based interpretation that explains why an observation matters.
• Decision intelligence connects an insight to a choice, such as entering a segment, changing a proposition or stopping an initiative.

For example, “35% of trial users did not complete onboarding” is data. “Users abandon onboarding when they must configure integrations before experiencing core value” is an insight if supported by behavioural and qualitative evidence. “Offer a usable default workflow before integration setup” is a product decision derived from that insight.

2. Market Research: From Question to Evidence

2.1 Defining the research problem

Weak research begins with a solution-led question: “Will customers buy our proposed app?” Strong research begins with a decision-led question: “Which recurring workflow causes sufficient cost or frustration for a defined group that an alternative solution could address?”

A research brief should specify:

• the decision that the research will inform;
• the target population and relevant context;
• the assumptions creating the greatest uncertainty;
• the evidence required to change the decision;
• the method, sample, timing and ethical safeguards;
• how findings will be analysed and communicated.

Research questions should be neither so broad that they produce general opinions nor so narrow that they conceal important context. Exploratory research asks what is happening and why. Descriptive research measures prevalence, frequency or characteristics. Evaluative research assesses a concept, prototype or experience. Causal research tests whether an intervention produces a particular outcome; it requires stronger controls than ordinary customer interviews.

2.2 Qualitative and quantitative methods

Qualitative methods are useful when the problem is poorly understood. Interviews, contextual observation, diary studies, usability sessions and analysis of support conversations reveal language, workarounds, emotions and situational constraints. Their purpose is depth and meaning, not statistical representation.

Quantitative methods measure incidence, preference, behaviour or relationships across a larger sample. Surveys, product analytics, transaction data, experiments and market databases can estimate magnitude and compare groups. Quantitative results are only as reliable as the sampling frame, question design, measurement validity and analysis.

A mixed-method design is often strongest. Interviews may identify candidate needs; a survey can estimate how widespread they are; behavioural data can test whether reported intentions correspond to actual behaviour; and usability research can explain observed friction.

Interview questions should focus on recent behaviour rather than hypothetical approval. “Tell me about the last time you solved this problem” is generally more informative than “Would you use a product that solved it?” The researcher should probe triggers, current alternatives, time and money spent, consequences of failure, decision participants and switching barriers.

2.3 Sampling and evidence quality

A sample is not automatically representative because it is large. Product research should distinguish:

• Target population: everyone to whom the findings are intended to apply.
• Sampling frame: the accessible list or channel from which participants are recruited.
• Sample: those actually studied.
• Selection bias: systematic difference between the sample and target population.
• Non-response bias: systematic difference between respondents and non-respondents.

Convenience samples can be appropriate for early exploration but should not be used to claim market prevalence. In business-to-business research, the user, economic buyer, technical approver, procurement function and executive sponsor may all hold different perspectives. A single interview per organisation is therefore rarely sufficient for a complex purchase.

Research quality also depends on triangulation. Confidence increases when different evidence types converge, for example when interview accounts, observed behaviour and support data indicate the same failure. Contradictions are not merely inconvenient; they may reveal different segments, usage contexts or stages of a journey.

3. Segmentation and Target-Market Choice

Segmentation divides a heterogeneous market into groups with sufficiently similar needs, behaviours or economics that they can be served with a coherent proposition. Common bases include:

• demographic or firmographic characteristics;
• geographic and regulatory conditions;
• psychographic characteristics and attitudes;
• behavioural patterns, such as frequency, usage intensity or switching behaviour;
• needs, desired outcomes and problem severity;
• value-based characteristics, such as willingness to pay or service cost.

A useful segment is not merely descriptive. It should be measurable, substantial enough to matter, reachable through identifiable channels, sufficiently distinct in needs or response, and compatible with the organisation’s capabilities and constraints. These criteria are often summarised as measurability, substantiality, accessibility, differentiability and actionability.

Segmentation should begin with the problem context rather than with attributes that are easy to collect. For example, “small firms” may be less useful than “small firms processing high volumes of exception-based transactions without dedicated operations staff.” The latter describes a condition linked to need intensity and potential value.

After constructing segments, product managers evaluate them using criteria such as:

1. Need intensity: How costly, frequent or urgent is the problem?
2. Reachability: Can the organisation access decision-makers and users efficiently?
3. Economic value: Is there sufficient willingness or ability to pay relative to service cost?
4. Competitive intensity: How strongly are existing providers positioned?
5. Strategic fit: Does the segment align with capabilities, brand and mission?
6. Growth and learning value: Can serving the segment create references, data or capabilities for expansion?

A simple prioritisation score can support discussion, but it should not disguise judgement. If criteria are weighted, the weights and rating scale must be explicit. Scores are decision aids, not measurements of objective truth.

4. Jobs-to-be-Done and Unmet Needs

Jobs-to-be-Done (JTBD) theory focuses on the progress a customer is trying to make in a particular situation. A job is not identical to a product, demographic category or feature. Customers “hire” products to achieve functional, emotional and social outcomes.

A job statement commonly includes context, motivation and desired progress: “When [situation], I want to [motivation], so I can [expected outcome].” For example: “When a supplier delivery is delayed, I want to identify alternatives quickly, so that production can continue without emergency procurement.”

JTBD analysis often distinguishes:

• Functional job: the practical task to accomplish;
• Emotional job: how the customer wants to feel;
• Social job: how the customer wants to be perceived;
• Related jobs: preparation, evaluation, purchase, use and disposal activities.

The framework is valuable because it directs attention from stated feature requests to underlying outcomes. However, it should not be applied rigidly. Jobs are interpreted through culture, power, identity and organisational processes. Customers may also articulate jobs inaccurately, especially when behaviour is habitual or politically sensitive. JTBD hypotheses must therefore be tested with evidence.

An unmet need can arise from several forms of gap:

• an important outcome is poorly achieved;
• an existing solution is too expensive, slow or difficult;
• the customer cannot access an appropriate solution;
• the end-to-end process contains coordination failures;
• a changing context creates a new requirement;
• customers have adapted through costly workarounds.

The existence of dissatisfaction does not guarantee a viable opportunity. The need must be important enough to motivate action, reachable by the product, and economically addressable.

5. Personas: Useful Models, Not Stereotypes

A persona is an evidence-based representation of a meaningful user or buyer pattern. It usually includes goals, responsibilities, behaviours, constraints, triggers, objections, relevant context and success criteria. A persona should be traceable to research, not invented from demographic imagination.

Personas are most useful when they support a decision: prioritising requirements, selecting language, designing onboarding or aligning a cross-functional team. They become harmful when teams create too many, treat them as permanent facts, or reduce users to labels such as age and job title.

In organisational products, several personas may coexist:

• the primary user who performs the task;
• the buyer who authorises expenditure;
• the administrator who configures the product;
• the approver who assesses risk or compliance;
• the beneficiary who receives the outcome.

A persona should therefore state the relationship between role and decision. The same individual may occupy multiple roles in a small organisation, while an enterprise purchase may involve a complex buying committee.

6. Customer Journey Mapping

A customer journey map represents the sequence of stages through which a customer encounters a problem, considers alternatives, purchases, uses a product and seeks support or renewal. A robust map links actions to thoughts, emotions, pain points, channels, backstage processes and opportunity areas.

Typical stages include awareness, consideration, purchase, onboarding, core use, support, renewal and advocacy. The stages should reflect the actual journey rather than an assumed funnel. Customers may loop, pause, involve colleagues or abandon the process entirely.

Journey mapping is especially valuable because product value is often created across organisational boundaries. A promising interface cannot compensate for slow approval, poor delivery, unclear billing or inadequate support. The map should identify:

• customer goal at each stage;
• customer actions and questions;
• touchpoints and channels;
• evidence of friction or success;
• responsible internal or ecosystem actors;
• moments that disproportionately affect trust;
• measurable outcomes and opportunities.

A journey map is a hypothesis until validated. Teams should distinguish reported emotion from inferred emotion and support claims with interviews, recordings, analytics, service data or observation.

7. Market Sizing: TAM, SAM and SOM

Market sizing helps determine whether an opportunity is large enough, reachable enough and strategically relevant. The three-level model is:

• TAM (Total Addressable Market): the total demand for a relevant category if the organisation could serve the entire theoretical market.
• SAM (Serviceable Available Market): the portion of TAM that fits the product’s geography, customer definition, capabilities and business model.
• SOM (Serviceable Obtainable Market): the realistically attainable share within a specified period, considering competition, capacity, channels and execution.

Two common approaches are top-down and bottom-up. Top-down sizing starts with published industry totals and applies transparent filters. Bottom-up sizing builds from customers, transactions, usage and price. Bottom-up estimates are often more useful for operating plans because assumptions can be tested directly.

A basic revenue estimate is:

where N is the number of relevant customers or units, P is average revenue per transaction or period, and F is the number of transactions or periods in the chosen time horizon. Units must be consistent: customers per year multiplied by revenue per customer per year yields annual revenue.

For a subscription product, a simplified estimate is:

where N is the number of paying accounts and ARPA is average revenue per account per year. This excludes expansion, contraction, churn, discounts and implementation revenue unless those are explicitly incorporated.

Worked example: bottom-up sizing

Suppose a workflow product targets 12,000 mid-sized firms in three eligible countries. Research indicates that 40% have the relevant workflow, so the SAM contains 4,800 firms. The proposed annual contract value is £6,000.

If the company has the sales capacity and competitive position to acquire 3% of these firms over its planning horizon:

This is not a forecast of guaranteed revenue. It is an assumption-dependent scenario. The 40% incidence rate, £6,000 price, 3% obtainable share and time horizon should each be tested. A large TAM can be strategically irrelevant if the SAM is small or the SOM cannot be reached profitably.

A second perspective uses units rather than revenue. If 20,000 eligible organisations each process 500 relevant cases annually, the annual addressable volume is 10,000,000 cases. At £1 per case, the theoretical annual revenue is £10,000,000, before accounting for adoption, competition, discounts and delivery cost. Different sizing methods should be reconciled rather than averaged mechanically.

8. PESTLE Analysis and Environmental Change

PESTLE examines macro-environmental forces that can create opportunities or threats:

• Political: public policy, government priorities, trade restrictions and institutional stability;
• Economic: inflation, interest rates, employment, income, currency and funding conditions;
• Social: demographics, norms, lifestyles, trust and patterns of work;
• Technological: infrastructure, standards, automation, cybersecurity and technology maturity;
• Legal: privacy, competition, consumer protection, employment and sector regulation;
• Environmental: climate risk, resource constraints, emissions and circularity expectations.

PESTLE is not a checklist for collecting disconnected facts. Each factor should be linked to a time horizon, affected stakeholder, mechanism of impact and strategic implication. For example, a new data-localisation requirement may change architecture, hosting cost, procurement eligibility and market entry sequencing.

The framework is strongest when combined with scenarios. A product team might describe a base case, an accelerated-regulation case and a recession case, then identify indicators that would signal movement between them. This avoids treating a single forecast as certain.

9. Ecosystem Intelligence

A product rarely operates alone. Its ecosystem includes customers, users, suppliers, manufacturers, logistics providers, payment systems, platforms, app stores, distributors, regulators, standards bodies, complementors, communities and competitors. Ecosystem analysis asks who controls resources, information, access, standards and trust.

A practical ecosystem map should identify:

1. actors and their roles;
2. flows of money, data, goods, attention and influence;
3. dependencies and bottlenecks;
4. incentives and sources of bargaining power;
5. points where coordination fails;
6. possible partners, substitutes and disintermediation threats;
7. regulatory and ethical responsibilities.

The analysis should distinguish direct competitors from substitutes and complements. A substitute solves the customer’s problem through a different approach; a complement increases the value of the focal product. A platform may be both partner and competitor, creating strategic dependence.

Ecosystem opportunities often arise at interfaces: connecting fragmented actors, making information portable, reducing verification cost, enabling interoperability or coordinating a process previously managed through spreadsheets and email. Yet ecosystem products face adoption challenges because value may depend on multiple parties joining. The product manager must consider sequencing, incentives, governance and minimum viable participation.

10. Integrating Intelligence into Opportunity Decisions

The central output of intelligence work should be an opportunity thesis, not a collection of charts. A strong thesis states:

• the target segment and relevant situation;
• the job or outcome that matters;
• evidence of problem severity and current alternatives;
• the market and economic boundary;
• ecosystem enablers and constraints;
• the proposed value hypothesis;
• the riskiest assumptions;
• the next test and its decision rule.

A useful opportunity scorecard may combine need severity, frequency, reachability, willingness to pay, competitive whitespace, strategic fit, feasibility and regulatory exposure. Qualitative evidence should not be converted into fake precision. Teams can use low/medium/high ratings, confidence labels and explicit evidence links.

A disciplined decision rule might state: proceed to prototype if at least three independent evidence sources indicate a frequent problem, five target organisations agree to test a workflow, and no unresolved regulatory barrier prevents the proposed model. Such a rule makes learning actionable and reduces escalation based on enthusiasm.

Illustrative case: CareLink scheduling

Background. CareLink is an illustrative digital product concept for coordinating home-care visits. The team initially proposes an AI scheduling tool for care providers.

Problem. Early conversations reveal that providers do not primarily lack a scheduling algorithm. Coordinators spend substantial time reconciling worker availability, travel constraints, client preferences, funding rules and last-minute cancellations. Care workers use informal messaging, while clients often receive incomplete updates.

Analysis. Segmentation identifies three groups: large providers with established scheduling systems, small providers using spreadsheets and small teams, and public agencies with complex procurement requirements. The small-provider segment has high need intensity and reachability, although purchasing budgets are limited. JTBD research produces the job: “When a visit changes unexpectedly, I need to re-plan safely and notify everyone quickly, so that clients receive reliable care without exhausting coordinators.” Journey mapping shows a critical failure between cancellation and replacement assignment. Ecosystem analysis identifies care workers, clients, families, providers, local authorities, payroll systems and messaging platforms. Privacy and safeguarding requirements are material.

A bottom-up estimate assumes 2,000 eligible small providers, 30% with the target workflow, and annual revenue of £4,000 per provider.

The figure is modest relative to a national software market, but it may support a focused entry strategy. The team should not claim the entire home-care technology market as its initial opportunity.

Approach. Rather than build a fully automated scheduler, CareLink prototypes a cancellation-response workflow: identify affected visits, suggest compliant replacements, obtain confirmation and issue transparent notifications. It conducts shadowing with coordinators, usability tests with care workers and a privacy review before handling live client data. The commercial hypothesis is tested through paid pilots rather than stated purchase intention.

Outcome. In this illustrative scenario, pilot evidence shows that notification coordination, not algorithmic optimisation, is the strongest initial value proposition. The product roadmap prioritises audit trails, permissions and integration with existing systems. Automation is introduced only where users can review and override recommendations.

Learning points. The case demonstrates the importance of separating a proposed solution from the underlying job, sizing the reachable segment rather than an abstract industry, mapping ecosystem dependencies, and treating regulation and trust as product requirements rather than late-stage legal checks.

11. Ethics, Governance and Responsible Use of Intelligence

Customer intelligence involves power over personal information and representation. Product teams should collect only data necessary for a legitimate purpose, obtain informed consent where required, protect sensitive information, define retention periods and restrict access. Privacy regulations may impose requirements concerning lawful processing, transparency, data minimisation, rights of access and automated decision-making. The applicable jurisdiction and sector must be assessed with qualified legal advice.

Research participants should understand what participation involves, how recordings and quotations will be used, and whether incentives could create undue pressure. Vulnerable groups require additional safeguards. Teams must avoid manipulative recruitment, discriminatory segmentation and the exclusion of people who are difficult to reach but materially affected by the product.

AI-assisted research creates additional risks. Automated transcription, summarisation and clustering can expose confidential data, reproduce bias, erase uncertainty or produce plausible but unsupported themes. Human researchers should verify source material, preserve contradictory evidence, document model use and avoid treating generated summaries as primary evidence. Synthetic personas and simulated interviews may support brainstorming but cannot substitute for contact with real stakeholders.

12. Limitations and Best Practices

Intelligence is necessarily partial. Customers may misremember, markets may change, competitors may conceal information and early adopters may not represent mainstream users. TAM calculations can exaggerate opportunity through double counting, inappropriate prices or unrealistic share assumptions. PESTLE can become generic. Personas can fossilise. Journey maps can reflect internal beliefs rather than customer reality.

Best practice includes:

• begin with a decision and explicit assumptions;
• combine qualitative, quantitative and behavioural evidence;
• recruit across roles, contexts and levels of experience;
• distinguish observed behaviour from stated intention;
• document sources, dates, sample limitations and confidence;
• size markets with transparent, testable assumptions;
• revisit intelligence as conditions and product learning change;
• include legal, privacy, accessibility and safety considerations early;
• make evidence accessible to design, engineering, marketing, sales and operations;
• define what evidence would falsify the opportunity thesis.

The most valuable intelligence is not the most elaborate. It is the intelligence that changes a consequential decision and improves the probability of creating meaningful, viable and responsible value.

Chapter Summary

Market, customer and ecosystem intelligence enables product managers to identify opportunities in context. Market research should begin with a decision, use methods appropriate to the question, and triangulate qualitative, quantitative and behavioural evidence. Segmentation defines groups that can be served with a coherent proposition, while JTBD focuses analysis on the progress customers seek in particular situations. Personas and journey maps translate research into usable representations of roles, experiences and friction.

TAM, SAM and SOM provide progressively narrower views of market potential, but their value depends on transparent assumptions and consistent units. PESTLE highlights macro-environmental forces, while ecosystem analysis reveals dependencies, incentives, flows and coordination problems beyond the focal product. The CareLink case illustrates how these frameworks can convert a broad solution idea into a focused, testable opportunity.

Finally, intelligence work requires ethical governance, privacy protection, critical interpretation and continuous updating. Evidence should culminate in an opportunity thesis with explicit assumptions, risks and next tests—not in false certainty.

Key Terms

Market intelligence: Evidence about market structure, size, growth, economics and competition.

Customer intelligence: Evidence about customer needs, behaviour, context, roles and outcomes.

Ecosystem intelligence: Understanding of actors, dependencies, incentives and flows surrounding a product.

Market segmentation: Division of a heterogeneous market into meaningful groups.

Target segment: The segment selected for focused service and investment.

Jobs-to-be-Done: A framework focusing on the progress a customer seeks in a situation.

Functional job: The practical task a customer wants to accomplish.

Persona: An evidence-based representation of a user or buyer pattern.

Customer journey map: A representation of customer stages, actions, touchpoints and friction.

TAM: Total theoretical demand for a relevant category.

SAM: The portion of TAM compatible with the product’s scope and capabilities.

SOM: The realistically obtainable portion of SAM.

PESTLE: Analysis of political, economic, social, technological, legal and environmental forces.

Complementor: An actor whose product or service increases the value of another product.

Substitute: An alternative solution that addresses the same customer problem.

Triangulation: Comparison of multiple evidence sources or methods.

Sampling bias: Systematic distortion caused by how participants are selected.

Opportunity thesis: A concise, evidence-based statement of a strategic opportunity and its assumptions.

Data minimisation: Collecting only personal data necessary for a legitimate purpose.

Review Questions

Short Answer Questions

1. Distinguish data, insight and decision intelligence using a product example.
2. What makes a market segment actionable?
3. Explain the difference between TAM, SAM and SOM.
4. Why should customer interviews focus on recent behaviour rather than hypothetical approval?
5. Identify two risks associated with AI-assisted customer research.

Long Answer Questions

1. Explain how qualitative and quantitative research can be combined in a product discovery programme. Discuss the strengths and limitations of each method.
2. Compare demographic segmentation with needs-based segmentation. Under what conditions might each be useful?
3. Explain Jobs-to-be-Done, personas and journey mapping, and show how the three frameworks can be used together.
4. Describe how a product manager should conduct a PESTLE and ecosystem analysis for a regulated digital product.
5. Discuss the limitations of TAM–SAM–SOM analysis and propose practices for producing more credible estimates.

Analytical / Application Questions

1. A firm identifies 8,000 eligible organisations. Research suggests 25% have the target problem, annual price is £3,000, and the firm expects to reach 4% of the relevant organisations. Calculate the annual SAM and illustrative SOM, then state two assumptions that require validation.
2. Construct a journey map for a service of your choice. Identify one high-friction moment, the evidence needed to validate it, and a product or service intervention.
3. A platform depends on one dominant external marketplace for customer acquisition. Use ecosystem analysis to identify at least three risks and recommend a strategy for reducing dependency.

Multiple Choice Questions with Answers

Question 1

Which statement best describes an opportunity thesis?

A. A list of all features requested by customers

B. An evidence-based statement linking a target problem, segment, market context and testable assumptions

C. A forecast that assumes competitors will not respond

D. A demographic description without behavioural evidence

Correct Answer: B. An evidence-based statement linking a target problem, segment, market context and testable assumptions

Question 2

Which research method is generally most suitable for understanding the context and motives behind a poorly understood problem?

A. Exploratory semi-structured interviews

B. A large financial forecast

C. A final sales report

D. A market-share press release

Correct Answer: A. Exploratory semi-structured interviews

Question 3

A customer job is best understood as:

A. The product category selected by a company

B. A demographic characteristic of a buyer

C. The progress a customer seeks in a particular situation

D. A list of technical features

Correct Answer: C. The progress a customer seeks in a particular situation

Question 4

A market contains 5,000 relevant firms, 20% have the target problem, and annual price is £2,000. What is the estimated annual SAM?

A. £200,000

B. £1,000,000

C. £2,000,000

D. £10,000,000

Correct Answer: B. £1,000,000

Question 5

Which statement correctly distinguishes SAM from SOM?

A. SAM is the theoretical global market and SOM is the total industry revenue

B. SAM is the reachable market within scope, while SOM is the realistically obtainable portion

C. SAM measures customer satisfaction and SOM measures usability

D. SAM is always smaller than SOM

Correct Answer: B. SAM is the reachable market within scope, while SOM is the realistically obtainable portion

Question 6

Which factor is most directly associated with ecosystem analysis?

A. The flow of data, money and influence among participating actors

B. The colour scheme of an interface

C. The spelling used in a product name

D. The number of pages in a research report

Correct Answer: A. The flow of data, money and influence among participating actors

Question 7

What is a principal danger of using a convenience sample to estimate market prevalence?

A. It always produces too many participants

B. It may not represent the target population

C. It prevents qualitative questioning

D. It eliminates non-response bias

Correct Answer: B. It may not represent the target population

Question 8

Which PESTLE category most directly includes privacy law and consumer-protection obligations?

A. Political

B. Economic

C. Legal

D. Environmental

Correct Answer: C. Legal

Question 9

Why is triangulation valuable in customer research?

A. It guarantees that every finding is causal

B. It replaces the need to define a research question

C. It compares different evidence sources to increase confidence and reveal contradictions

D. It ensures that all participants agree

Correct Answer: C. It compares different evidence sources to increase confidence and reveal contradictions

Question 10

Which practice is most consistent with responsible customer intelligence?

A. Collecting all available personal data for possible future use

B. Using synthetic interviews instead of speaking to affected users

C. Documenting limitations, protecting data and collecting only what is necessary

D. Removing contradictory evidence from the final report

Correct Answer: C. Documenting limitations, protecting data and collecting only what is necessary