Research methodology
Rigour you can see. Evidence you can trace.
Our methodology is designed around one principle: every recommendation should be traceable to evidence, and every piece of evidence should be fit for the decision it supports.
Last reviewed by the Hoog research team
The Hoog Research Engine
Five stages. One decision.
Define, Discover, Validate, Analyse, Advise — each stage has a clear output that feeds the next.
- 01
Define
Understand the business question and decision context.
- Decision & stakeholder mapping
- Hypotheses
- Scope & success criteria
Output: Research brief
- 02
Discover
Secondary research, industry intelligence, databases and AI-assisted discovery.
- Desk research
- Databases & filings
- AI-assisted source discovery
Output: Evidence base
- 03
Validate
Primary research including expert interviews, IDIs, FGDs and quantitative surveys.
- Expert interviews & IDIs
- FGDs
- Quantitative surveys
Output: Validated findings
- 04
Analyse
Market modelling, segmentation, competitive analysis, trend analysis and insight synthesis.
- Market models
- Segmentation
- Competitive & trend analysis
Output: Insight synthesis
- 05
Advise
Translate evidence into strategic options, recommendations and actions.
- Strategic options
- Recommendations
- Action roadmap
Output: Decision & actions
Primary research
Qualitative and quantitative, designed together.
We choose methods for the question — not the other way round. Most studies combine qualitative depth with quantitative confidence.
Qualitative research
- In-depth interviews (IDIs)
- One-to-one conversations exploring motivations, decisions and context in depth.
- Expert interviews
- Structured calls with industry specialists and former executives to validate assumptions.
- Focus group discussions (FGDs)
- Moderated group sessions to explore perceptions, language and reactions to concepts.
- Consumer interviews
- Individual conversations with consumers about needs, usage and choice.
- B2B interviews
- Research with decision-makers, influencers and users inside organisations.
- Dealer & distributor interviews
- Channel perspectives on demand, competition, pricing and service.
- Stakeholder interviews
- Internal and external stakeholder views to frame the question and test options.
- Ethnography (where relevant)
- Observing behaviour in context — at home, in store or at work.
Quantitative research
- Online surveys
- Scalable structured measurement across consumer and professional audiences.
- CATI
- Computer-assisted telephone interviewing for hard-to-reach or B2B respondents.
- B2B surveys
- Targeted surveys among screened business decision-makers.
- Consumer surveys
- Representative or quota-based measurement of attitudes and behaviour.
- Market sizing
- Top-down and bottom-up models, triangulated and validated.
- Customer satisfaction
- Satisfaction measurement with driver analysis across touchpoints.
- NPS
- Net Promoter Score with follow-up diagnostics to explain the score.
- Brand tracking
- Funnel and perception metrics measured consistently over time.
- Concept testing
- Evaluating propositions, products, packaging or messaging before launch.
Secondary research
A strong evidence base, before we ask a single question.
Desk research defines the landscape, sharpens hypotheses and ensures primary research focuses on what published sources cannot answer.
- Company filings and annual reports
- Industry associations and government statistics
- Trade and technical publications
- Licensed databases (as agreed per project)
- Regulatory and policy documents
- Patents, hiring and product signals
AI in our process
AI accelerates. Researchers validate.
Where AI fits in the workflow — and where human judgement takes over.
- 01Research discoveryAI-accelerated
- 02Data extractionAI-accelerated
- 03Competitive monitoringAI-accelerated
- 04Interview processingAI-accelerated
- 05Theme identificationAI-accelerated
- 06Insight synthesisAI + researcher
- 07Human validationResearcher-led
- 08Strategic recommendationsResearcher-led
AI accelerates the research process. Human researchers validate evidence, interpret context and translate findings into business decisions.
Quality assurance
How we protect the integrity of findings
Source verification
Every material claim is traceable to a source, and important figures are cross-checked.
Respondent screening
Participants are screened for role, experience and relevance; low-quality responses are removed.
Assumption logging
Model assumptions are documented so they can be challenged and updated.
Peer review
Findings and recommendations are reviewed by a second researcher before delivery.
AI output review
Any AI-assisted output is checked by a human for accuracy, context and consistency.
Ethics & consent
Informed consent, transparency about who we are, and data minimisation.
FAQ
Frequently asked questions
How does Hoog conduct primary research?
We start from the business question and define who can answer it — customers, experts, channel partners or decision-makers. We then choose the right mix of qualitative methods (such as IDIs, expert interviews and FGDs) and quantitative methods (such as online surveys and CATI), recruit and screen respondents carefully, and analyse results alongside secondary evidence.
What is the difference between primary and secondary research?
Secondary research uses information that already exists — reports, filings, statistics and databases. Primary research collects new information directly from people through interviews, focus groups and surveys. Secondary research builds the base; primary research validates it and fills the gaps.
What is triangulation in market research?
Triangulation means using several independent sources or methods to answer the same question — for example, comparing a top-down market estimate with a bottom-up estimate and with expert views. Agreement increases confidence; disagreement reveals what needs further investigation.
Start a conversation
Have a market, customer or competitive question?
Tell us about the decision you are facing. We will come back with a clear view of how research can help — and what it would take.