Instead this question should be posed: Which brand do you prefer; X or Y?
Personal Interviews
This is essentially the same as an questionnaire, but a the interviewer rather than the interviewee fills out the responses to questions on a questionnaire. In interviews there is a tendency to use more open questions since the interviewer is recording the results and can control what is recorded to avoid useless data. The interviewer must be skilled at their job and not bias questions but there is a high usable response rate. Another advantage is that if a question isn’t understood it can be explained. Interviews can be time consuming though, and rely much upon the skill of the interviewer.
Telephone Interviews
These are similar to a personal interview, but held over the phone. The length of questions is limited but the advantages are that telephone interviews are cheaper and can cover a large geographical area easily.
These also have a lower response rate than face-to-face interviews, as they are distrusted by the public.
Postal Surveys
Postal surveys are inexpensive compared to most methods of field research, but they have a poor response rate and are time consuming, taking up to six weeks for data collection. A large advantage of postal surveys is that by taking away the interviewer they lower costs and remove bias from the data accumulation.
Observation
This involves watching consumer behaviour, it can help in the design of shops but does not explain the reasons behind the actions of the shopper which may mean problems can be identified but not solved. E.g. The direction people walk round a supermarket may be studied but how it affects the amount they purchase, what they purchase and therefore how much they spend will not be clear for each person, only a general theory can be formed.
Use of Technology
Electronic Point of Sale (EPOS) systems which calculate the cost of your shopping, are also used for stock control, which means that the item that was bought and the time and date it was bought are recorded. This can be useful for market researchers to calculate sales of seasonal item or how external factors affect purchases e.g. when it is sunny how many more cartons of fruit juice are sold.
Video cameras that can count the number of potential customers entering a shop within a shopping centre and which areas are most heavily visited. Some systems can even differentiate between adults children and pushchairs, which gives information on the age of customers as well as the number of the,. In this way different areas and different shopping centres can be compared.
Expression of views through the internet and websites is becoming more popular. Data can also be collected when orders are placed over the internet.
Loyalty cards are a valuable method of data gathering, a customer profile can be built up and discounts / special offers can be aimed specifically at them.
Often technology means that the consumer is not involved directly in the data capture for market research, this is much more convenient for them and for the researchers as they do not have to rely on the cooperation of the subject/s.
Focus Groups
This type of research uses observation but needs the cooperation of the consumers. By listening to discussion on a number of questions given to the group the researcher can gain an insight into the views of their chosen target consumer group. The disadvantage lies in that by only using a small group the views of the whole market may not be represented, but it does mean that it is inexpensive to carry out.
Consumer Panels
A group of customers are brought together on one or a number of occasions to give feedback on a product. This is relavtively cheap because there is a small number of people, but may not give a wide enough cross-section of the public, therefore it may reflect the views of the market segment in which the business is interested.
Test Marketing
By selling a product in a small section of the market (e.g. limited to a particular geographical area) it can be tested and evaluated before going into full production. It is less expensive than fully releasing a product and can lead to more research and development for products that are not as successful as hoped.
Field research can give companies an advantage over their competitors, as the firm that originally collected it owns it. It also lets the business know where they stand as the sales markets open up. It can be very expensive and time consuming to carry out but it also means that by collecting it a business has control over exactly the questions that are asked and the data that is gained. It can make customers feel valued and included in the company’s interests. It can help in looking to the future and adapting to the ever-changing business world in a sensible way without wasting money on pointless projects.
Research is very useful but not totally reliable or an accurate prediction of what will happen in reality. By the time a product is launched consumer requirements may have changed and the product may not sell. The problem of bias is a considerable one and careful sampling must be carried out to avoid this as well as how the research itself is done e.g. careful phrasing of questions in an interview. This can ameliorate statistical bias and therefore sampling discrepancies.
Secondary Research
This can take many different forms but the important factor is the research has already taken place.
From within the business it could be:
- Existing market research reports
- Sales figures showing previous performance.
- Reports from direct contact with customers, which may highlight good and bad points of existing products and services.
- Annual Reports and Accounts published by businesses, which must be produced by law, and provide an easy to understand snapshot of the health of the business.
- Stock movements, (more up to date than sales figures) are a good indicator of current performance.
- Web sites may help to establish the type of image the business is portraying to its customer base.
From sources outside the business, it could be:
- Promotional material, product specifications and price lists from competitors so the competition can be analysed- especially useful when thinking about marketing.
- Government statistical publications (e.g. census) can show how big a potential market is. E.g. how many 45 to 50 year olds there are in the UK with a income of £60,000 + - if that is your target audience for a new product.
- Data from customer services on complaints about products may help to develop existing products and correct problems.
- The European Commission produces statistical data regarding the European Economy, which is valuable to companies operating in EU countries.
- International Publications (e.g. from the World Bank) data from here may also be too general to be of use to a small company.
- Commercial Publications on the subject of various markets- for example industry specific newspapers and magazines.
- Retail Audits, which allow businesses to monitor performance of sales constantly.
- General Publications e.g. magazines, newspapers, telephone directory.
Secondary research is low cost, sometimes free, as it has already been produced but does not always contain all the information that is needed, or may not be up to date. It is difficult to use if a business is trying to investigate the potential success of a new product in development, as often the type of questions that need to be asked about colour or price may not have been asked and answered before. For descriptive and explanatory data it can be extremely useful as it is looking at the past, and two different types can be collated.
Sampling
Sampling is the process of taking some consumers’ views, clearly it is not possible to use the entire population due to the cost, time or sheer size.
Random Sampling
Random sampling allows every person in a group an equal chance of being chosen but may not mean that the sample is totally representative, as it as it assumes all people in the group are homogeneous. Systematic sampling involves taking a name at a set interval, every tenth or fiftieth for example, but this is less random.
Stratified Random Sampling
Researchers often prefer this method as it make the sample more representative of the group. The population is divided into groups and then random samples are taken from each group, making sure that there are the same proportions of the sample in each category as in the population as a whole.
Quota Sampling
The population is divided up into segments that share particular characteristics, then a target is set for each. Once the set target is reached then no more will be interviewed. This does not mean that this sample represents the population, nor is not random either. This makes the information less statistically valid, but it is cheaper than other methods.
Cluster Sampling
With this method, the population is divided into “clusters” (usually geographical areas), and then random samples are taken from each cluster. This can be done quickly so it is useful for opinion polls.
Multi Stage Sampling
This involves selecting one sample from another, eventually the level of detail will be reached. For example, one household from one street from one town in one country will be chosen.
Snowballing
This starts with one individual or group, then using this contact to develop more. This is used in extremely small, specialised industries, when no other sampling methods are possible.
