Editor's note: Hy Mariampolski, Ph. D., is president of QualiData Research Inc., Brooklyn, New York.

The established focus group ritual is well known: Group discussions are two hours in length with 10 to 12 participants and are conducted at a professionally appointed facility (with two-way mirrors, comfortable seating and M & M's). Participants are recruited to specification--usually, not mixed by demographics or opinion. The discussion guide, the concepts presented to participants and other elements of the discussion are agreed upon beforehand and, except for the occasional spontaneous question passed to the moderator on a slip of paper, there is often little variation from this standard group format.

These have become the standardized ways of doing things but often there are good reasons to reject convention. Breaking the mold and seeking new insights through methodological innovation are intellectually refreshing. There are also pragmatic reasons to experiment with research design. Alternative research designs, for example, may be appropriate for intense discussions on sensitive or controversial topics.

This article describes and offers a rationale for several variations on conventional focus groups and discusses how to avoid pitfalls in implementing them. Creative experimentation and relaxing rigid rules enhances the validity and vitality of the qualitative approach. Besides, these strategies can be exciting and fun to try.

Maxi groups

Why has two hours become the optimal time frame for a group session? Spontaneous real life discussions over dinner or at a bar are not limited by this rigid cutoff and often focus groups should be allowed to linger longer as well. Discussion guides have grown in recent years making it difficult to fully explore responses to all the additional questions contributed by product and research managers in just two hours. Extra time is helpful also when the project calls for the use of creative techniques, brainstorming, and projectives.

Sensitive topics, including discussions of credit card usage or other financial issues, political or health care issues, for example, often require lengthy warm-ups. Respondents need to emerge from their shells slowly and a motivational pep talk cannot be handled in a ten minute introduction and briefing. Rapport building and a pattern of open disclosure may often require up to a half hour or more; rushing will only produce flawed and incomplete responses.

We have had considerable success recently with groups that are three and even four hours in length. Such groups prove valuable to clients who need to gain a particularly deep psychological understanding of their customers' attitudes and behaviors. Maxi-groups can challenge the moderator's energy but the absence of time pressure can also be liberating. Even participants find these stimulating- one woman leaving a recent group that lasted four hours suggested we hold a reunion next year!

The secret to conducting exciting maxi-groups is to vary the structure and tone of the group to heighten interest and participation. Continuous, staid discussion around a conference table will kill maxi-groups. Instead, the moderator should move back and forth between active exercises, discussion and product trials. Quiet periods and bathroom breaks are also needed. Larger than average, roomy facilities are necessary for the comfort of both participants and observers. Respondents must be encouraged to dress casually.

On-site studies

Who said the suburban mall or the office building is the only appropriate location for conducting focus group interviews? These facilities are certainly professional and efficient- accomplishing recruitment, hosting and taping under one roof. They are quite comfortable, especially for client observers. But they can also be inconvenient, cold and intimidating for certain types of participants such as elderly and down-scale respondents. Even among average respondents, the laboratory setting can create expectations for only "acceptable" answers. Additionally, for many clients, especially not-for-profit arts, charitable or public interest groups, facility rental can be a costly extravagance that deters them from fully utilizing qualitative research.

During a recent trip to the United Kingdom, where I had the chance to meet with many British moderators, I was surprised to learn that about 85 % of group interviews conducted there are held in individual homes. Both moderators and participants seem to prefer the more intimate living room environment to the office based facility.

In recent years, our firm has conducted focus groups and in-depth interviews in stores, offices, restaurants, schools, homes and community centers. Extra planning and coordination are required for these; provisions must be made for hosting and taping the sessions. Client observers who attend are usually encouraged to play an active role as "assistants."

Conducting research on-site- which has become the catchword for ethnographic or other research brought to the subject's home or workplace-is often the best way to study naturally occurring groups and to observe behavior while exploring attitudes in their context. This method is ideal, for example, for exploring corporate culture while observing the pattern of office allocation, or for observing hands-on use of office automation equipment in context, or for observing families as they prepare and eat meals.

How many participants

Why has ten become the conventional number of participants in a group discussion? It certainly is not always the best number for sharing ideas in a group discussion. If you subtract 20 minutes for introductions and warm-up, each of ten people in a two hour group has only 10 minutes to convey their opinions and feelings.

Participants with considerable technical expertise-and that includes most executive, managerial, business-to-business, and professional markets- often have to struggle to fully express their opinions in conventional focus groups. Moderators know that it's also easier for shy or reluctant participants to hide in a ten person discussion versus one with six or seven persons.

When the objectives of the research demand depth of response from either consumers or business executives, we encourage our clients to recruit six to eight participants rather than the conventional ten. We have also found the smaller groups to be perfect with children and teens because the session seems less like a classroom and permits the moderator to focus more attention on individuals in the group.

Confrontational groups

Who said that it's best for focus groups to reach consensus? I'm often puzzled when, upon leaving a session, clients solicit my interpretation of what "the group said" when what I heard were widely divergent individual opinions.

In this age of fragmented markets and of passionately asserted beliefs, structuring disagreement is better than striving for a false congeniality. Confrontation often inspires new creative insights that emerge above the noise in the marketplace. It can spark the discovery of new market segments and positionings. Dissent in focus groups can be natural: acceptable variations in attitudes arise during discussions. Alternatively, confrontational groups can be structured by purposely recruiting participants on opposite sides of an issue. These are particularly effective with public issues and political research and in work with "controversial" products which, now-a-days can include everything from red meat to alcoholic beverages, cigarettes to salad dressing. Debates between users and non-users of a brand are also quite effective.

A relaxed, dignified and non-judgemental moderator is particularly important in groups with structured confrontation. Respect for people with alternative opinions must be encouraged and things must not be allowed to get personal. Participants should be briefed about the potential for controversy at the time of recruitment.

Structured confrontation allows the client to hear the diverging opinions their brand or category may elicit. This provides opportunities for insight into factors that may cause a change of attitude, and helps establish the intensity of various beliefs.

Iterative groups

Why must the same sequence of questions govern a series of groups? Why can't discussion guides and test concepts evolve from group to group? The main benefit in encouraging modifications is that a run of at least four, but preferably six to eight groups, allows for the creation of quasi- experimental conditions in the research project.

Iterative groups are sessions purposely planned so that information and hypotheses gained in early groups can be tested and elaborated in subsequent groups. This approach can lead to stronger advertising concepts, positionings and even copy. Rather than leaving a project with the discouraging sense that "consumers didn't like the idea," researchers can experiment with variations in positionings, for example, that are possibly more motivating than others. The iterative approach gives ideas a chance to evolve and develop.

I have been involved in iterative projects where the client's marketing managers, copywriters as well as research managers worked actively with me before and after each group. Based on initial findings, we refined marketing strategy hypotheses and then copywriters rewrote concepts and positioning which were then tested in the next group. Planning sessions were held each morning to fine tune the discussion agenda for the groups to be conducted that evening. The result was not only a richer understanding of consumer needs and preferences, but also fully developed advertising copy to address consumer's expressed wishes.

This article has proposed variations in the composition, duration, location and sequencing of conventional focus groups. Rather than being the final word on the subject, it is intended to encourage experimentation with qualitative approaches. Breaking the boundaries of tired habits helps both respondents and moderators release creative energy and helps research users get the most from their investments in focus group research.