Market Research

Marketing research is a systematic, objective and scientific activity. It is a decision-support and problem-solving process that is applied, continuous and interdisciplinary, and it draws on both qualitative and quantitative methods to reduce uncertainty in marketing decisions.

Two marketing research problems are:

1. Product research problem: Why are sales of a product falling, and which features do customers want improved?
2. Pricing or advertising problem: How will customers react to a price change, and how effective is a current advertising campaign?

Causal research is conducted to establish cause-and-effect relationships between variables, such as the effect of a price cut or an advertisement on sales. It uses controlled experiments, where one variable is changed and the effect on another is measured while other factors are held constant. It is more rigorous than exploratory or descriptive research.

Distribution research studies the channels through which goods reach customers. It covers the choice of wholesalers, retailers and outlets, the location of stores, channel performance and costs, inventory, and transportation. Its aim is to find the most efficient way of making the product available to the target market.

Sampling is the process of selecting a subset (sample) of units from the whole population so that conclusions drawn about the sample can be generalized to the population. It saves time and cost compared with studying every unit (census).

A review of literature is a systematic study of previous books, journals, reports and research papers on the research topic. It shows what is already known, identifies gaps, and helps the researcher build the conceptual framework, hypotheses and methods of the new study.

• Open-ended question: "What do you like most about our brand of detergent?" The respondent answers in his or her own words.
• Close-ended question: "How satisfied are you with our detergent? (a) Very satisfied (b) Satisfied (c) Neutral (d) Dissatisfied (e) Very dissatisfied." The respondent chooses from fixed options.

Cluster-based segmentation is a statistical method of dividing the market into groups (clusters) of customers who are similar to each other within a group and different from other groups. It uses variables such as demographics, attitudes, buying behaviour and lifestyle, and the groups emerge from the data rather than being decided beforehand. The firm can then target each cluster with a suitable marketing mix.

Advertising research is the systematic collection and analysis of information to improve the planning, creation and evaluation of advertising. It tests the message, media choice and effectiveness of advertisements, for example through recall, attitude change and sales impact.

Simple random sampling is a probability sampling method in which every unit of the population has an equal and known chance of being selected. Units are drawn by lottery or random number tables, so the choice is unbiased. It requires a complete list (sampling frame) of the population.

Marketing research is carried out through a series of steps.

1. Define the problem and research objectives: Clearly state the decision problem (for example, falling sales) and what information is needed. A well-defined problem guides the whole study.
2. Develop the research plan (design): Select the type of research (exploratory, descriptive or causal), the data sources (primary or secondary), the methods of contact (survey, observation, experiment) and the budget and time.
3. Review secondary data and literature: Collect existing information to understand the issue and refine the problem.
4. Design the sampling plan and instrument: Decide the population, sampling method and sample size, and prepare the questionnaire or observation form.
5. Collect the data: Carry out the fieldwork with trained interviewers and supervision, to ensure accuracy and honest responses.
6. Process and analyse the data: Edit, code and tabulate the data, and apply statistical techniques to find patterns and test hypotheses.
7. Interpret findings and prepare the report: Present the results, conclusions and recommendations clearly to management.
8. Follow up: Management uses the results for decision-making and the effectiveness of the research is evaluated.

Literature review is the study of earlier research and writing on the topic. Its importance is as follows.

1. Clarifies the problem: It helps the researcher define and narrow the research problem and objectives.
2. Avoids duplication: It shows what has already been done, so effort is not wasted repeating earlier work.
3. Identifies research gaps: It reveals unanswered questions and weaknesses in earlier studies, which justify the new study.
4. Builds the theoretical framework: It provides theories, concepts and variables on which hypotheses are formed.
5. Guides methodology: The researcher learns which methods, tools, samples and statistical techniques were used and how well they worked.
6. Provides a basis for comparison: Findings can be compared and discussed against previous results, which improves the quality and credibility of the study.
7. Builds the researcher's knowledge: It gives a good understanding of the subject and of the terminology and sources available.

In probability sampling every unit of the population has a known, non-zero chance of selection. The main types are as follows.

1. Simple random sampling: Each unit has an equal chance of selection, using a lottery or random number table. It is simple and unbiased, but needs a complete list of the population.
2. Systematic sampling: Every $k$th unit is chosen after a random start, where $k = N/n$. It is easy and fast, but may be biased if the list has a pattern.
3. Stratified random sampling: The population is divided into homogeneous groups (strata), such as age or income, and a random sample is taken from each stratum, usually in proportion to its size. It ensures representation of all groups.
4. Cluster sampling: The population is divided into clusters (such as wards or districts), some clusters are selected at random, and all or some units within them are studied. It is cheap for scattered populations but less precise.
5. Multistage sampling: Sampling is done in stages, for example district, then ward, then households, using random selection at each stage. It is suitable for large populations.

Questionnaire construction is a stepwise process.

1. Specify the information needed: Based on the research objectives and hypotheses, decide what information is required and from whom.
2. Decide the type of questionnaire and method of administration: Choose structured or unstructured, and mail, personal, telephone or online administration.
3. Decide the content of individual questions: Each question should be necessary and should be answerable by the respondents.
4. Decide the form of response: Choose open-ended, close-ended (multiple choice, dichotomous) or scaled (Likert, semantic differential) questions.
5. Decide the wording: Use simple, clear, unbiased words. Avoid leading, double-barrelled, ambiguous and sensitive questions.
6. Decide the sequence and layout: Begin with easy, interesting questions, group related questions together, put sensitive and personal questions at the end, and keep the layout neat and attractive.
7. Pre-test: Try the questionnaire on a small group to find errors, confusing words and problems with length.
8. Revise and finalize: Correct the problems, then print or digitize the questionnaire and use it in the field.

A research proposal is a written plan that tells the reader what will be studied, why, and how. Its main components are:

1. Title and cover page: a short, precise title with the researcher's and institution's details.
2. Introduction / background: the context of the topic, the market or industry setting, and why the study matters.
3. Statement of the problem: a clear description of the gap or decision problem to be solved.
4. Objectives and research questions: what the study intends to find out, stated in measurable terms.
5. Hypotheses: tentative, testable statements about relationships between variables.
6. Significance and scope/limitations: who benefits from the study and what it will and will not cover.
7. Review of literature: earlier studies and theories, showing the research gap.
8. Conceptual framework: the variables and their expected relationships.
9. Research methodology: research design, population and sampling, data sources, data collection tools (questionnaire, interview) and the tools of analysis.
10. Work plan and budget: time schedule (Gantt chart) and cost estimate.
11. References and appendices: sources cited and the draft questionnaire.

Together these show that the study is relevant, feasible and methodologically sound, and they help a sponsor or committee approve it.

Post-testing measures the effectiveness of advertising copy after it has been run in the media. The main methods are:

• Recall tests: respondents are asked what they remember about the ad. Unaided recall gives no clues, while aided recall gives product or brand cues. Examples are the Starch and Gallup and Robinson tests.
• Recognition tests: respondents are shown the ad and asked whether they have seen it before. This is common for print ads.
• Attitude and brand-awareness tracking: surveys before and after the campaign measure changes in awareness, image and preference.
• Inquiry tests: the number of coupon returns, calls, letters or website visits generated by the ad is counted.
• Sales-effect (sales-results) tests: sales are compared before and after the campaign, or between test and control markets.
• Physiological and behavioural tests: eye tracking, and web analytics such as clicks and conversions for digital ads.

The best approach is usually a combination, because sales are also affected by price, distribution and competition, while recall alone does not show purchase.

(a) Role of marketing research in selecting the new name

Marketing research reduced the risk of choosing a name that would cost more than 50 million dollars to introduce. Accenture used:

• Idea generation: an employee contest produced candidate names (exploratory/qualitative input).
• Surveys of target customers: they tested the candidate names for meaning, pronunciation, recall, positive association and fit with the positioning ("future of business").
• Linguistic and legal screening: the names were checked for unwanted meanings in other languages and for trademark availability in 47 countries.
• Evaluation: the research experts compared the names against criteria and found the one that sounded most like "adventure", which supported the desired positioning.

The research therefore gave an objective basis for the decision instead of managerial opinion.

(b) Target market and the role of research in understanding its needs

The target market consists of large corporations and governments across 19 industry groups, such as communications and high tech, financial services, consumer goods, health and natural resources. These include most Fortune Global 100 and 500 firms, and the buyers are senior executives (CEOs, CIOs, CFOs) who purchase consulting, technology and outsourcing services.

Research helps by:

• identifying the problems and needs of each industry through depth interviews and focus groups with executives;
• measuring satisfaction with current service providers and the criteria used in choosing a consulting firm;
• segmenting clients by industry, size and need, and sizing the market;
• tracking changes in the needs of clients so that service offerings can be modified.

(c) Research design to increase preference and loyalty

• Problem: what determines preference for and loyalty to Accenture's services, and how can they be raised?
• Type of design: exploratory research first (interviews with clients, secondary data on the industry), then conclusive descriptive research (a cross-sectional survey, and a longitudinal tracking panel to see change over time). A causal experiment may test the effect of a service or communication change.
• Data collection: a structured questionnaire using Likert and rating scales on service quality, relationship, price, innovation, satisfaction, preference, repurchase intent and recommendation. Collection is by personal interview, telephone or online survey.
• Sampling: the population is the existing and potential clients; use stratified random sampling by industry and size.
• Analysis: descriptive statistics, cross-tabulation, correlation and multiple regression, with the loyalty measure as the dependent variable, plus factor analysis of the drivers.
• Output: a report with recommendations on service improvement, relationship management and loyalty programmes.

(d) Building brand equity research

As the researcher I would:

1. Define brand equity as awareness, perceived quality, associations and loyalty (Aaker/Keller).
2. Run qualitative research (focus groups, projective techniques) to find what the brand stands for in the minds of clients.
3. Run a quantitative brand-tracking survey to measure aided and unaided awareness, image, perceived quality, preference and willingness to pay a price premium.
4. Benchmark against competitors such as Deloitte, PwC and IBM.
5. Evaluate campaigns ("Now it gets interesting", "Be a Tiger", "High performance. Delivered.") using post-testing for recall and attitude change.
6. Estimate the financial value of the brand, and repeat the tracking each year so management can see the trend and act on weak dimensions.

Marketing research is the systematic design, collection, recording, analysis and reporting of data and findings relevant to a specific marketing situation, to help management make better marketing decisions.

Ethics is important in marketing research because it protects respondents' rights (privacy, confidentiality, informed consent), keeps the data honest and unbiased, and maintains the credibility of the research profession and the trust of clients and the public. Without ethics, results would be misleading and respondents would stop cooperating.

A research objective is a clear statement of what a study intends to achieve or find out. It comes from the research problem and guides the design, the data collected and the analysis. For example, "to determine the effect of service quality on customer satisfaction".

A review of literature shows what is already known on the topic, avoids duplication, identifies the research gap, helps to frame the problem, hypotheses and conceptual framework, and guides the choice of methods and variables. It also supports the results by comparison with earlier findings.

Two characteristics of a good research design are:

1. Objectivity and reliability: it gives consistent, unbiased results that can be repeated.
2. Validity and suitability: it matches the research objectives, and it is flexible and economical in time and cost, while controlling extraneous variables.

Quantitative research is a research approach that collects numerical data from a large, usually representative, sample with structured tools such as questionnaires and experiments, and analyses it statistically to measure, test hypotheses and generalise results.

Two sources of secondary data are:

1. Internal sources: company sales records, accounts, customer databases.
2. External sources: government publications (such as the Central Bureau of Statistics and Nepal Rastra Bank reports), trade journals, books, and published research.

Types of research questionnaire:

• Structured (closed-ended): fixed questions with fixed answer choices.
• Unstructured (open-ended): respondents answer in their own words.
• Semi-structured: a mix of closed and open questions.
• Disguised and undisguised: the purpose is hidden or revealed.

By mode of administration they can be mail, telephone, personal-interview or online questionnaires.

Advertising research is the systematic study of advertisements and their effect on the audience. It tests the message, copy, media and campaign effectiveness before the ad is run (pre-testing) and after it has run (post-testing), using measures such as recall, recognition, attitude change and sales results, so that advertising money is spent effectively.

A research proposal is significant because it:

• presents the problem, objectives and method clearly before the work starts;
• is used to get approval, funding or permission from a sponsor or committee;
• acts as a plan and guide for carrying out the study on time and within budget;
• helps the researcher to find weaknesses early;
• serves as a yardstick to judge the final work against the plan.

Marketing research decisions are affected by several factors:

1. Decision-maker's information need and the problem: the importance and clarity of the problem determine whether research is justified at all.
2. Cost-benefit: research is carried out only if the value of the information is greater than its cost.
3. Time: when a decision is urgent, a long study may not be possible, so quick methods or secondary data are used.
4. Availability of existing information: if enough secondary data exist, new primary research is not needed.
5. Budget and resources: funds, skilled staff and technology limit the scale and method.
6. Management attitude and culture: managers who trust intuition may ignore research, and those who value it will commission it.
7. Environmental factors: competition, the economy, technology, culture and legal and ethical constraints.
8. Respondent and data issues: accessibility of respondents, willingness to answer and the quality and reliability of data.
9. Research capability: whether the firm has an internal department or must outsource to an agency.

When these factors are weighed carefully, the research design, scope and timing can be chosen so that the study is useful and affordable.

A research problem is a clearly stated question, gap or difficulty that a researcher wants to solve through investigation. It states what is unknown or needs a decision, for example "why have sales of brand X fallen in the Kathmandu market?". A good problem is specific, researchable, measurable, relevant and feasible.

Importance in the research process

• It is the starting point of the entire study, because a poorly defined problem leads to wasted effort ("a problem well defined is half solved").
• It sets the objectives and hypotheses and decides the scope.
• It determines the research design, the data needed, the sampling and the analysis.
• It guides the literature review and keeps the study focused.
• It helps in judging the cost, time and feasibility of the study.
• It makes the findings useful to the decision-maker, because the results answer the question that was actually asked.

Problems are usually identified through discussion with management, exploratory research and the review of literature, and the management problem is then converted into a research problem.

Sampling is the process of selecting a part of the population to represent it. There are two main groups.

A. Probability sampling (every unit has a known chance of selection)

1. Simple random sampling: units are chosen by lottery or random numbers.
2. Systematic sampling: every $k$-th unit is chosen after a random start, where $k = N/n$.
3. Stratified sampling: the population is divided into homogeneous strata (such as age or income) and a random sample is taken from each.
4. Cluster sampling: the population is divided into clusters (such as wards), some clusters are selected randomly, and all or some of their units are studied.
5. Multistage sampling: sampling is done in several stages, for example district, then municipality, then household.

B. Non-probability sampling (selection depends on judgement or convenience)

1. Convenience sampling: whoever is easily available is chosen.
2. Judgement (purposive) sampling: the researcher selects units believed to be typical or informative.
3. Quota sampling: fixed quotas are filled for categories such as gender or age.
4. Snowball sampling: initial respondents refer further respondents.

Probability sampling allows the sampling error to be estimated and the results to be generalised. Non-probability sampling is cheaper and quicker but can be biased.

Data analysis is the process of editing, coding, tabulating and applying statistical techniques to the collected data to summarise it, find patterns and relationships, test hypotheses and draw conclusions that support marketing decisions.

Methods of data analysis

1. Descriptive analysis: mean, median, mode, percentages, standard deviation and frequency tables describe the data simply. It is easy but gives no explanation or testing.
2. Cross-tabulation: shows the relationship between two categorical variables, and is often used with the chi-square test. It is easy to read, but it cannot deal with many variables.
3. Hypothesis tests ($t$-test, $z$-test, ANOVA, chi-square): test whether differences between groups or associations are statistically significant. They depend on assumptions such as normality.
4. Correlation and regression: correlation measures the strength of a relationship, and regression predicts a dependent variable from independent variables. Correlation does not prove causation.
5. Multivariate techniques: factor analysis (reduces variables), cluster analysis (segments customers), discriminant analysis, conjoint analysis (preferences) and multidimensional scaling (perceptual maps). These are powerful but need large samples and expertise.
6. Qualitative analysis (content and thematic analysis): interprets interviews and focus-group material. It is rich in depth but subjective.

Critical view: the choice depends on the objectives, the scale of measurement (nominal, ordinal, interval, ratio), the sample size and the skill available. Simple methods are easier to understand and communicate, while advanced methods are more informative but may be misused. Most studies combine the two, and the results are only as good as the quality of the data.

A marketing research report communicates the problem, method, findings and recommendations to management. The usual format is:

1. Preliminary section

• Title page, letter of transmittal and letter of authorisation
• Table of contents, list of tables and figures
• Executive summary: a one-page statement of the problem, main findings and recommendations (often the only part senior managers read)

2. Main body

• Introduction: background, problem definition, objectives and scope
• Methodology: research design, data sources, sampling plan, questionnaire, fieldwork and analysis methods
• Data analysis and findings: results with tables, charts and interpretation
• Limitations: constraints such as sample size or non-response
• Conclusions and recommendations: conclusions drawn from the findings and the actions suggested

3. Appendix and reference section

• Questionnaire, technical details, additional tables, bibliography

Evaluation of a good report: it should be clear, concise, accurate and objective, use simple language and suitable visuals, be organised logically, and answer the original research questions so that management can act on it.

Product research is the study of products and of customers' reactions to them, in order to develop, test, improve and manage products.

Purpose

• To generate and screen new product ideas and concepts.
• To test concepts, prototypes and test-market products before launch.
• To find the product attributes, features, design, packaging, brand name and quality that customers value.
• To study customer satisfaction, usage and competing products, and to decide on modification, line extension or deletion.
• To forecast demand and identify the stage of the product life cycle.

Significance

• It reduces the high risk of new product failure.
• It keeps the product aligned with customer needs, which improves satisfaction and loyalty.
• It helps in positioning and differentiation against competitors.
• It saves cost by detecting problems early, before mass production.
• It supports decisions on pricing, branding and distribution, and increases sales and profit.

Therefore, product research is important to the success of the whole marketing programme.

(a) Research problem

The bank wants to increase its transaction volume by raising customer satisfaction, but does not know which factors drive satisfaction. The research problem is: "What are the key factors (service quality, service cost/interest, bank location, employee responsiveness, ease of service and online services) that affect customer satisfaction at the bank, and how strongly does each affect it?"

(b) Conceptual framework

Independent variables                         Dependent variable

 Service quality           ---- Service cost (interest)   ----- Bank location             ------>   Customer satisfaction
 Employee responsiveness   -----/            (and so, volume
 Ease of service           ----/              of transactions)
 Online services           ---/

The six factors are the independent variables, and customer satisfaction is the dependent variable. The arrows show the expected direct effect of each factor.

(c) Hypotheses

• $H_1$: Service quality has a significant positive effect on customer satisfaction.
• $H_2$: Service cost (interest) has a significant effect on customer satisfaction.
• $H_3$: Bank location has a significant positive effect on customer satisfaction.
• $H_4$: Employee responsiveness has a significant positive effect on customer satisfaction.
• $H_5$: Ease of service has a significant positive effect on customer satisfaction.
• $H_6$: Online services have a significant positive effect on customer satisfaction.

(d) Recommended research design

A descriptive, cross-sectional survey design (with a causal-explanatory orientation, using correlation and multiple regression) is recommended.

Reasons:

• The problem is to describe the customers' perception and measure the relationship between known factors and satisfaction, so exploratory research is not necessary because the factors are already identified by theory.
• A survey with a five-point Likert questionnaire fits the data in scale form and allows regression analysis.
• The customers are spread over three districts (60% Kathmandu, 25% Lalitpur, 15% Bhaktapur), so a stratified random sample with proportional allocation by district gives representative results.
• A cross-sectional design collects the data once, which is quick and economical for a single bank branch.
• The results can be generalised to the customers, and hypotheses can be tested statistically.