Telephone interviews are a quick and cost-effective way to obtain opinions from a sample of customers. They can be used to assess customer reaction to a change in the product or service, or to gauge satisfaction ratings compared to competitors. The make-up of the sample is crucial; it is more important to know the views of valuable customers rather than those who haven't bought from you for a while or who tend not to spend a lot of money on your products and services. Do remember, though, that some people don't like being 'cold-called' and may not want to participate. It might be worth sounding out some key customers in advance by e-mail, on your website, or via a newsletter if you have one, as you'll get much more useful information from someone who actually wants to take part in your research.

The main benefit of telephone research is speed: a large number of interviews can be conducted in a short space of time, and data can be gathered and processed quickly. Telephone interviews also cost considerably less than personal interviews. As long as they are brief and to the point they are non-intrusive and easy to arrange. A customer may be too busy for a personal interview, but willing to spend a short time on the phone. Again, it may be worth sounding people out in advance.

Be careful about a senior decision-maker at another business being contacted by a junior member of your staff or an independent researcher, though—they may prefer to hear from you directly if you have a good relationship with them—and always bear in mind the effect on the customer's relationship with you.

Telephone interviews have limited scope because people may not be prepared to spend a long period of time on the telephone. It can also be difficult to get across complex concepts by telephone.

Follow industry guidelines:

• define the target audience
• draft a questionnaire and test it with colleagues
• make the calls at times likely to be convenient for the target audience, but check when you call that it is convenient; if not, offer to call back
• identify the purpose of the call and give an indication of the likely duration
• use faxes, if necessary, to send more detailed information for discussion

Questionnaires are very difficult to compile well, and those who do it well have honed their skills over many years. If you feel a questionnaire is that important to your research, it may be worth contacting a questionnaire specialist to help you out. Other types of postal surveys are delivered directly to customers and they can be a quick and relatively inexpensive method of obtaining information. They are ideal for customer satisfaction surveys or detailed surveys that take time to complete, but it can be difficult to obtain a reasonable rate of response and you cannot assume that the responses will be representative of all customers.

You can reduce the cost of postal surveys even further by using e-mail to distribute surveys and collect responses. You could also upload a research form to your website or ask respondents to answer questions by short text messages.

Postal surveys are relatively inexpensive. The costs include outward and return postage and stationery. They are precise and can be targeted at specific customers or prospects. They are also voluntary, because there is no pressure on the customer. The main problem, however, is the low response rate; incentives may be needed to improve it. There is also a risk of questionnaires and survey forms being returned incomplete or being incorrectly completed. You may face a slow response because there is no time pressure on the customer (unless you introduce one, by offering a discount or some such if customers respond within a certain time limit), so you must take this into account in your planning.

Response rates can be influenced by many different factors. For many consumer goods markets less than 5% would be normal and 15% or more would be extremely high. For highly involved customers in some sectors you might anticipate much more. If you need to improve response think about:

• offering an incentive, such as entry into a prize draw, for returned questionnaires
• simplifying the questionnaire
• reassuring the customer that information will be kept confidential
• enclosing a stamped addressed or business reply envelope

In group discussions—sometimes called 'focus groups'—existing and prospective customers are invited to discuss a particular topic, usually under the guidance of a researcher. They are ideal for exploring and, possibly, identifying issues or problems which concern customers, and assessing customer reactions to potential changes.

There is no limit placed on what the group can discuss, and this format can highlight important customer issues that the researcher may not be aware of. Many customers welcome the opportunity to discuss products and services with their colleagues and have an opportunity to contribute to change. The disadvantages are that group discussions are not representative, and they can be biased or influenced by a dominant member of the group. It is also difficult to quantify results.

You need to be cautious in planning group discussions. By and large they should not be used to identify only what is wrong with a product or service, and it may cause problems to have several important customers in the same room. Groups composed entirely of customers with grievances will also have limited value: it's unlikely that you'll be able to get past their individual complaints.

To set up a group discussion:

• invite eight to ten customers; this number is controllable, but with fewer people it may be difficult to maintain discussion
• thank customers for participating and put them at their ease
• record the discussion if possible, using a tape recorder
• advise people that their comments are being recorded and that all material will be treated in confidence
• if conducting a group amongst consumers (not your customers) you might consider providing participants with a gift for taking part
• consider using more than one group to gather a range of views

You could also consider using an independent researcher to run the discussion.

In a personal interview, a customer and an interviewer work through a series of predetermined questions. The personal interview is ideal for key customers and for obtaining detailed information on attitudes to products and services, or initiating the process of getting feedback. The interview can take place in a customer's home or office, or in a public place. It can be prearranged by telephone, post, or personal contact.

Personal interviews allow in-depth discussion of complex topics and give you more control over the response. They offer greater accuracy, and results are easy to analyse. Meeting people in a working environment can give an indication of their real purchasing intentions. The main disadvantages are the time and cost of recruiting interviewers and conducting interviews and the risk of interviewer bias.

Before you carry out personal interviews, you should identify the groups or individuals you wish to contact. You should advise the customer about the length of time the interview is likely to take and respect the confidence of the customer.

An omnibus survey is a cost-effective method of researching several topics at the same time. The same survey is used to carry out regular research on different products by telephone or personal interview. They are suitable for measuring attitudes and behaviour towards different types of products and services or monitoring changes in attitude among groups of consumers.

Before participating in the survey, you should check that the other topics in the survey are compatible with your own products and that the overall length of the survey is not excessive. It may reduce costs, but the audience may not give enough attention to all topics.