Discourse analysis examines how language shapes meaning, attitudes and social reality around a topic, brand, product or category. In market research, discourse analysis helps identify not only what people say, but also how they frame issues, assign responsibility, use metaphors and reproduce shared assumptions.
What is discourse analysis?
Discourse analysis is a qualitative research approach focused on language in context. It studies recurring ways of speaking and writing about a subject, including vocabulary, narratives, metaphors, contrasts, labels, arguments and implicit assumptions. Rather than treating statements as isolated opinions, discourse analysis investigates the wider meaning systems that make certain interpretations appear natural, credible or dominant.
The term originates in linguistics, sociology, communication studies and cultural analysis. In market research, discourse analysis is used to understand how consumers, customers, employees, media outlets, brands and other stakeholders construct meaning around markets and commercial decisions. It is particularly useful where perceptions are shaped by public debate, identity, trust, risk or social norms.
Discourse analysis may examine material such as:
- individual interviews and focus group transcripts,
- open-ended survey responses,
- social media posts, comments and discussions,
- media articles, podcasts and press releases,
- customer reviews and complaints,
- brand communication, advertising and product descriptions,
- internal documents, sales materials and employee communications.
The central question is not only whether a given sentiment is positive, negative or neutral. It is also which interpretive frames are used. For example, consumers may describe electric vehicles through a discourse of innovation and progress, but also through a discourse of inconvenience, cost, technological uncertainty or environmental responsibility. Each frame highlights different benefits, risks and decision criteria.
In discourse analysis in marketing research, the unit of analysis can be a word, phrase, metaphor, narrative pattern, speaker position or interaction between participants. The method also considers who is speaking, to whom, in which channel and under what conditions. The same phrase can carry different meaning in a consumer interview, a social media debate and a corporate sustainability report.
Discourse analysis in practice
Discourse analysis in marketing research is applied when an organisation needs to understand the language and cultural logic behind customer perceptions. It is especially relevant when standard survey metrics show that an attitude exists but do not explain how that attitude has been formed or why particular arguments resonate with specific groups.
The method is often used in research concerning brand positioning, reputation, communication effectiveness, category entry and sensitive market issues. It can support both B2C and B2B projects, although the sources and terminology differ between these contexts.
Typical applications of discourse analysis include:
- Brand perception research: identifying the language consumers use to describe a brand and comparing it with the language used in brand communication.
- Communication development: testing whether campaign messages fit existing audience frames or challenge them in a credible way.
- Category research: mapping how people talk about emerging categories such as artificial intelligence, renewable energy, digital health or plant-based food.
- Reputation and issue monitoring: analysing public narratives around trust, pricing, labour practices, sustainability, data privacy or product safety.
- Customer journey research: examining how customers explain barriers, doubts and moments of confidence before or after purchase.
- B2B decision-making research: identifying how buyers define value, risk, expertise, innovation and supplier credibility in procurement processes.
For example, a financial services provider may find that customers do not describe a savings product primarily through interest rates or features. Instead, they may use language associated with security, control, adulthood or fear of making a mistake. Such findings can inform message architecture, customer support content and product naming.
In a mixed-methods project, discourse analysis can explain patterns identified in quantitative research. A survey may show that one customer segment is less willing to adopt a new service. Interview and social media data can then reveal whether hesitation is framed as distrust in the provider, concern about complexity, fear of losing autonomy or resistance to a broader market trend.
Hume’s Institute can apply discourse analysis to interview materials, online discussions, media content and open-ended research responses when the objective is to understand the language behind market attitudes rather than only measure their distribution.
Discourse analysis and related methods
Discourse analysis belongs to the broader group of qualitative approaches used to interpret unstructured data. It can be combined with content analysis, thematic analysis, semiotic analysis, ethnography, social listening and survey research. These methods may use similar source material, but they answer different questions and require different analytical logic.
The distinction between discourse analysis vs content analysis is particularly important. Content analysis generally focuses on the presence, frequency or classification of defined themes, words, topics or categories. It can be qualitative, quantitative or mixed. For example, content analysis may measure how often customer reviews mention price, delivery or product quality.
Discourse analysis goes further by examining how those topics are constructed in language. It asks whether price is presented as unfair, unavoidable, evidence of premium quality, a sign of corporate greed or a necessary trade-off for sustainability. The focus is therefore on meaning, framing and the relationship between language and social interpretation, not only on topic occurrence.
The key differences can be expressed as follows:
- Content analysis identifies and classifies what is present in a body of material.
- Thematic analysis identifies recurring patterns of meaning across qualitative data.
- Discourse analysis examines how language constructs those meanings, positions speakers and legitimises particular interpretations.
- Semiotic analysis focuses more strongly on signs, symbols, visual codes and cultural associations.
- Sentiment analysis estimates the emotional tone of text, usually at scale, but may not capture irony, context or competing frames accurately.
Discourse analysis is often paired with thematic analysis. Thematic analysis can first identify major topics in interview or social media data, while discourse analysis investigates the language patterns within each topic. This combination is useful when research needs both a clear map of customer concerns and a detailed understanding of how those concerns are articulated.
When discourse analysis is most useful
Discourse analysis is most valuable when language itself is strategically important. This is the case when a brand operates in a contested category, enters a new market, addresses a sensitive issue or needs to understand why the same proposition receives different reactions from different audiences.
It should be selected when the research objective includes questions such as:
- How do customers define the problem that a product is intended to solve?
- Which words, metaphors and narratives support or weaken trust in a brand?
- How is a category discussed differently by consumers, experts, media and competitors?
- Which assumptions shape resistance to a new technology, service model or pricing approach?
- Does brand communication reflect the language of the target audience or impose an unfamiliar frame?
Discourse analysis does not estimate market size, prevalence or statistical significance on its own. Findings should therefore be interpreted as an explanation of meaning patterns within the analysed material, not as direct proof that every identified discourse is equally widespread. Where prevalence matters, discourse analysis can be followed by quantitative validation through surveys, structured content analysis or other measurement methods.
Used appropriately, discourse analysis provides a precise view of the language through which markets are understood. It enables researchers and decision-makers to move from a list of stated opinions to an evidence-based understanding of the frames, assumptions and narratives that influence brand perception and customer behaviour.