Editor's note: Reyn Kinzey is the senior research analyst with RIVA Market Research's Richmond, Va., division.

Clients are increasingly pressuring focus group moderators for quick turnaround of reports. Moderators, in turn, are increasingly concerned about these requests. Articles like Naomi Henderson's article, "Qualitative report writing - is faster better?" in Quirk's Marketing Research Review (December 1992), and Alice Rodgers' sarcastically titled article, "Better, Faster, Cheaper Research," in QRCA Views (Summer 1993) testify to this trend.

I respect Naomi Henderson, who is my boss (as well as one of the best moderators in the business) and Rodgers, who is my editor at QRCA Views. And I share some of their fundamental concerns: pushing a moderator to churn out a thorough, analytic report in two to five working days of the last groups in a project is likely to produce research that is quicker but not necessarily better. On the other hand, I'm here to suggest that, done properly, faster is better.

The rub is that it's not cheaper. It's more expensive, but not necessarily much more expensive. I'll explain that usually takes only three to five days to later, but for now, let me say that rapid write, if nothing else is pending; often reports can be produced without compromising quality, bankrupting the client or forcing the researcher to take a loss on the project.

The problem

Before we get into the economics of the issue, let's isolate the real problem. Everyone agrees that clients often have legitimate reasons for wanting qualitative information as quickly as possible. One project for which I recently wrote a rapid report concerned a cancellation of a "frequent user" program. The clients wanted to see how badly customers would react to cancellation of the program. The first night of the project, group participants complained that they were reading about the cancellation before the company had notified them. Given that kind of rapidly changing situation, the clients didn't want to wait three weeks for a report.

And why should they? The main reason we normally tell clients that reports require two or three weeks to write is that it normally takes from five to eight days to get the transcripts back from our transcription service. The report itself the report can be written in two days. If we could get transcripts as soon as the groups finished, reports should never take more than a week, because we'd lose no time waiting for transcripts. But of course, a moderator can't moderate and transcribe at the same time.

Solutions

An obvious solution is to write the report without a transcript, a "top-of-mind" report, in which the moderator just summarizes the main points that stick out in memory. Naomi Henderson discussed this possibility in her article, but she also has reservations, which I share. In fact, I refuse to do "top-of-mind" reports unless the client absolutely insists. I don't trust my top-of-mind thinking. By its very nature it cannot be analytic: you don't have to be a trained psychotherapist to realize that what sticks out in the memory of even a trained moderator may not be the most important points from the perspective of the group or the client. We may tend to remember what interests us as individuals, rather than the trends or themes that emerge from an analytic review.

Of course, some moderators write reports without transcripts by listening to the tapes over and over again. That process can produce good reports, but it takes almost as much time as having transcripts made.

In any event, without a transcript or repeated listening to tapes, it's almost impossible to place the most important findings in a meaningful context. Reports written without either one of these tools - in short, without analysis - can be very valuable if the groups' purpose is very limited and well-defined (for example, does the new product warrant further research? Does the advertising pass a "reality check" kind of test?). But such issues barely need a report at all anyway. If, however, the purpose is to truly gather consumers' perceptions, opinions, beliefs and attitudes - a more opened-ended goal, which requires an analytic report - then transcripts are necessary, so the analyst can grasp the entire scope of the sessions.

So, back to square one: How do you provide the transcripts? Another obvious solution is to have a second moderator/analyst in the observation room who makes a handwritten transcript as the groups proceed. (If you do this, we recommend that the observer be a trained moderator: Making an accurate and complete transcript manually is difficult, and it helps if the person understands the focus group process). With this method, the moderator working with the participants can concentrate on what he or she is doing. As soon as the groups are finished, so are the transcripts, and writing the report can begin. That's how I do it, and, doing so, I promise a report within two to five working days of the last group.

Advantages and disadvantages

Having done things this way for two years, I'm now confident that the process produces reports that are not only faster but better. However, I haven't forgotten my own original concerns about quality. I was concerned that handwritten transcripts might not be as comprehensive as those made from tapes, by professional transcription services I've been happily surprised to find that they are generally more comprehensive.

The human ear picks up secondary conversation better than the microphones in most facilities. Of course, if clients talk, laugh, whistle and generally misbehave, and I'm transcribing, I may have trouble hearing what's going on; that's actually the biggest problem I've had with this procedure. But it's not like I'm working without a safety net: we still record everything. If I know I've missed something, I simply make a note of it, and go back later and find it on the tape. That's still a lot less time consuming than transcribing entirely from a tape.

But the biggest advantage of the handwritten transcript over a tape transcript is that the person on the scene can keep track of who makes which comments. With a tape transcript, it may be impossible to tell if the same person voices the same opinion 10 times or if 10 different people are voicing the same opinion over the course of the evening, which ii' obviously an important distinction for the analysis. Some people may argue that a moderator will naturally recall such nuances, but remember: If the moderator has to wait a week or possibly two to get the transcripts, he or she may have conducted eight or even more other groups since the ones in question. I don't trust the human memory that much.

And, of course, this procedure has the "two heads are better than one" advantage. Not only are there two moderator/ analysts who have actually witnessed the groups (one concentrating his or her efforts on conducting the group, the other on absorbing everything that I said), both can begin analyzing the group immediately, as soon as it's over.

How long does analysis take?

Here's the really interesting question: How long does analysis take? There's no easy answer. Part of the Puritan heritage of our country is to believe that the longer you spend on something, the better job you do. I'm not sure that kind of reasoning should be applied to analysis. In fact, much of the language we use to describe what analysis unveils are visual terms: we "see" patterns in the data; we have "insights." And the human eye is even faster than a camera. Analysis can sometimes be almost instantaneous.
Sometimes it's not, of course. I don't even try to analyze the findings from a group until I've slept on it. Still, if you have a transcript in your hands, five days shouldn't press anybody to complete a 20- to 40- page report. You can always write the sections on purpose and scope and methodology before the groups. The findings go into the detailed summary and the executive overview.

After all, the purpose of the report is to discuss the groups. All the data has been gathered. (A lot of research - academic research, for instance - is very time-consuming because it's almost impossible to know when enough secondary data has been collected, but that's not an issue here.)

If the project includes more than two groups, I write the detailed summary of each night's groups as we go along. This can be draining, but it provides an extra advantage. We can use those summaries to help refine the moderator's guide as we move from city to city. In fact, we first developed this procedure for a project where we did a series of groups on five successive Thursday nights: Each Monday, the clients received a summary of the previous Thursday night's group, so they had three or four days to modify the project before the next two groups.

The biggest disadvantage

The challenge to developing this procedure has been to make it cost-efficient for everybody involved. At first, it looks impossible. The procedure ties up two moderators, and if you have to double your prices, the enterprise becomes too expensive. However, that's not really the situation. Generally, when we do this, the project work is divided: One moderator works as the moderator, the other as the primary analyst who writes the report. For example, the "moderator" writes the screener and the topic outline. After the analysis is completed, however, this moderator goes on to other projects, and returns only to read the final report.

With this division of labor, the only parts of the project that are "double teamed" are the actual four hours of the groups and the time for analysis. Even this time is efficiently used, because the analyst who witnessed a group is able to read through and analyze his own handwritten transcript more quickly than a typed transcript of a group he has never seen. Another cost savings is that we don't have to send out tapes to be transcribed.

For all these reasons, we are able to provide rapid turnarounds at only slightly higher professional costs. Travel costs are more of a problem, since two people need to travel, doubling air travel costs, but clients have not seemed overly troubled by this. When they need something in a hurry, they'll pay the travel cost. About one-third of the projects our division does are rapid turnarounds.

The final concern about "cost" that people raise about this procedure is the "loss of opportunity" cost: Since two moderators need to be at one site, the second cannot be booked at higher moderating rates. That is theoretically true, but consider: a moderator could theoretically moderate 400 groups per year (two each night, Monday through Thursday, 50 weeks a year). But based on an average of four groups to a project, that would require 100 reports a year, two every week.

My contention is that writing the reports would choke the moderator faster than moderating the groups. Our rate-determining factor is producing the reports, not moderating the groups, so the "loss of opportunity" issue isn't realistic. It's to our advantage - as well as to our clients' - to get reports out rapidly.

Just do it

Faster can be better, and only slightly more expensive, if it's done right. There's no trick to the procedure. But it does requires a good moderator, a good analyst, and a lot of hard work. Every research project requires that. Faster is simply a matter of doing it.