Editor’s note: Naomi Henderson is founder of and a master moderator at RIVA Market Research, Bethesda, Md.

As a little girl, I often helped my dad when he worked around the house. He answered all my questions, never missing an opportunity to teach me practical life lessons related to the chores at hand. One of them was "Use the right tool to get the job done correctly." The first practical lesson that accompanied this sage wisdom related to screwdrivers. Dad patiently explained the difference between screws with one slot and screws with two slots that form a cross. He told me that the flathead screws were fine for regular carpentry when many were used to anchor one piece of wood to another. However, when you wanted to make sure that the two pieces were held tightly or there wasn’t space for more than one or two screws, then a Phillips screw served best, primarily because the extra slot allowed the carpenter to really tighten down the screw. Two different screwdrivers were needed: a flathead screwdriver with a squared off tip for regular one-slot screws and a Phillips screwdriver that had a tip that looked like an "x."

My dad’s advice can also be applied to qualitative research. To get the right results, you need to use the right research tool for the job.

More appropriate

Focus groups are what come to most people’s minds when they think of a qualitative research tool. But in-depth interviews (IDIs) can be just as handy and can, in many cases, be a more appropriate approach than focus groups. In past years, IDIs were also called one-on-ones, but as the technique improved, they got a much better name! In IDI sessions the moderator has a guide and a planned flow to the questions. Comments from respondents can move away from the planned path and take the interaction in different directions. For example, a respondent is asked which of three package designs is most appealing. While answering, he digresses to talk about the wastefulness of American packaging, and that tangent might be briefly explored before returning to the planned line of questions. Or, a respondent may answer in such a way that a later question is answered before being even asked! This is the very nature of IDIs -- while there is a plan, it is not rigidly followed as long as the questions keep leading toward uncovering the perceptions, opinions, beliefs and attitudes (POBAs) of respondents.

There can be in-depth interviews in the survey arena as well. However, in survey research the questioning format remains the same in that the interviewer asks each person the same questions and does not vary from the proscribed sequence except in the case of skip patterns.

An IDI is the exception in qualitative research in that it looks like a survey experience with one interviewer and one research subject. The difference is in the set up of the interviewing experience and in the way questions are asked. Survey research relies heavily on closed-ended questions, while IDIs rely on open-ended questions. This key difference is important, not only in the interviewing process but in the analysis as well. Moderators have to stay vigilant in IDIs to avoid a cookie-cutter approach across a series of interviews. They must find ways to streamline questions so that the respondents have the widest range of answers as they share POBAs. This is the challenge of IDIs: keeping the research rigor that puts this technique squarely in the qualitative arena while avoiding sliding into quantitative scorings in an attempt to categorize data from a string of individual interviews.

Qualities of a proficient IDI researcher

A survey can be completed by a trained interviewer, while an IDI is best handled by someone who has mastered a key set of qualitative research skills. Experience as a moderator is a good training ground as is related work in anthropology, sociology and psychology. However, specific academic degrees do not necessarily confer competence in this area.

Key qualities of good in-depth interviewers include:

  • good interviewing skills (e.g., listening without judging, asking clear questions, etc.);
  • the right mix of intelligence and good common sense;
  • good voice tone, pacing, pitch and volume;
  • the appropriate combination of critical reasoning and imaginative thinking;
  • an eye for detail and the ability to hold big picture at the same time;
  • the ability to appear genuinely interested (as a person) and truly detached (as a researcher);
  • an appropriate blend of empathy and neutrality in word and deed;
  • the ability to think analytically and live without a sense of closure.

Appropriate applications of IDIs

While focus groups may be conducted more often, from time to time, IDIs are the right tool. These single-person sessions with a moderator can last as little as 15 minutes or as long as two hours, depending on topic and purpose. Some classic applications for IDIs include:

  • communication checks (review of print, radio or TV advertisements or other written materials);
  • sensory evaluations (e.g., reactions to varied deodorant formulations, viscosity of hand lotions, sniff tests for new perfumes, or taste tests for a new frosting, etc.);
  • exploratory research (to help define baseline understanding of products, services or ideas);
  • new product development - prototype stage;
  • packaging or usage research (IDI research is used when clients want to "mirror" personal experience and obtain key language descriptors).

This is not a comprehensive list of the applications, just the most common ones. Our firm has been asked to test tapes of talent for radio and TV stations in IDIs to help broadcast clients tease out insights about the image of on-air personalities. We have conducted individual interviews with terminally ill patients about plans they are making for their families now and for the year following their death. We have talked to liquor store owners about a controversial issue regarding liquor consumption.

In past years, IDIs were commonly used for sensitive topics like feminine hygiene, AIDS treatment, incontinence, sexually transmitted diseases, and hemorrhoids. These days, with an increase in support groups, these topics and others are no longer kept quiet or private and respondents are motivated to tell their stories and be heard in a research setting with others who share the same condition. While sensitive issues can still be explored in IDIs, the tool has been expanded to cover communication checks, sensory testing and to mirror situations of individual use of products or services as well as for traditional applications (e.g., exploratory research, new products and packaging studies).

IDI benefits and drawbacks

Returning to the screwdriver analogy, we could say that focus groups require a flathead screwdriver because the interaction between respondents outweighs the risk of "thought leaders" who might sway others in the group. A moderator just needs to make sure each screw is tight enough to hold the wood in place. By contrast, IDIs need a Phillips screwdriver because insights from each respondent have to be "locked in tight" to insure a clear understanding of key research issues. The chart outlines some of the classic benefits and drawbacks of this powerful research tool.

Benefit of IDIs

  • Singular viewpoints from a respondent without influence from others
  • Can follow or track a theme with one respondent from start to finish (e.g., from initial purchase decesion to current satisfaction levels)
  • Optimizes low incidence rates in recruiting
  • Flexible structure – can quickly modify or expand topics as learning emerges
  • Can probe specific issues
  • Can be conducted in places other than a mirrored research facility (e.g., at a convention, on-site at a factory, etc.)
  • Opportunity to discuss personal or intimate topics in confidence
  • Some segmentation can be done in IDIs that cannot be handled in focus groups (e.g., four users, four non-users, and four trier-rejectors cannot be included in one focus group)

Drawbacks of IDIs

  • Costly in time, money and effort
  • Easy to fall into the trap of analyzing data qualitatively (i.e., "How or in what ways?")
  • Time required to conduct interviews often results in less client involvement at the observation stage
  • Risk of moderator burnout if too many are held in one day or too many days in a row – latter IDIs often not as rich as early ones if moderator becomes tired
  • Physically exhausting for moderator and clients to listen for nuances and insights with fresh ears when IDI series last several hours (e.g., 10 30 minute interviews, six 45-minute interviews or two 90-minute interviews)
  • Misconception that IDIs do not require the same skills as moderating a group discussion (Fact: more skill is needed in IDIs because of the high level of involvement with respondents)

Elements typically present in IDI research

A number of key factors or elements are present in an IDI study. They include having a trained researcher, a qualified respondent and an appropriate setting for the interview. In addition, there are payments to respondents, a guide for the discussion and a subjective report of findings across the series of interviews conducted. Interviews are usually audiotaped. Videotapes are made if there is a requirement for a historical record of the interviews or if there are many instances of show-and-tell events in the interview.

Sometimes, clients make a request for IDIs, thinking they will have better research if there is no group influence on the conversation. Sometimes, clients make the choice because they are heirs to previous bad research techniques where a poor moderator didn’t allow for individual distinctions. In group settings, a good moderator can mitigate the influence of "thought leaders" with such techniques as "private writing before public disclosure" so that respondents can anchor their viewpoint first. Another technique is to let respondents know at the outset that you value a difference of opinions rather than consensus.

Trained researcher

The training models for in-depth interviewers can range from learned techniques in the social sciences (ethnographic research skills learned in anthropology courses, interviewing techniques from sociology and psychology) to specific course work in corporate workshops, seminars and training programs. Some graduate programs in colleges and universities also have course work in group dynamics and interviewing. Regardless of the basis of the training, it should incorporate these abilities: to write effective questions; to probe for clarity; to establish and maintain authentic rapport with respondents; to pace the interviewing session; to use interventions and projective techniques appropriately; and to analyze data to support client objectives.

Recruiting respondents

The same rules for recruiting qualified respondents for focus groups apply to IDIs. An appropriate screener is used to find respondents who match research specifications and respondents are paid for the time they spend. Respondents are also rescreened when they first arrive at the facility. IDI research requires that a respondent is waiting for the next interview rather than hoping each respondent will show up on time. Facilities are asked to have a "floater" recruited for each set of four interviews to be conducted. For example, if four 30-minute interviews are planned between 9 a.m. and noon, one respondent is paid a higher fee than the standard stipend to wait for that whole time period in case one of the four recruited is a no-show. That way, the moderator is never waiting for the next interview. The floater is given refreshments and reading materials (or they are encouraged to bring their own). Sometimes floaters can be seen watching TV or doing their knitting!

Setting

It is a good idea to set up the interviewing room to achieve as intimate a setting as possible. If the only space available is a traditional focus group room with a big conference table then it is best for the moderator to sit in one of these two positions:

1. The traditional chair (moderator’s back to the mirror) with the respondent on the right side, slightly facing the moderator and slightly facing the mirror.

2. No one in the traditional chair. Moderator on left side of table (first chair) and respondent on right side of table, first chair.

In both models, allowing the client to see the respondent’s face and keeping materials to be shown in easy reach are the critical elements.

If you only have a large conference table, create a "small table effect" by placing something about five feet from the front of the table (like a row of blank paper to make a line, or a length of string, or even a strip of masking tape) - anything that makes a demarcation - the "end of our space." This artificial line creates a sphere of intimacy in a large conference table room. In newer facilities, special and smaller IDI rooms have been constructed to create that intimacy. Chairs with wheels allow both moderator and respondent to wiggle around while talking or listening, adding to personal comfort levels that are a key factor in full communication.

Creating relationship

Some moderators like to excuse a respondent at the end of the interview and then wait for the facility hostess to bring in the next one. RIVA prefers to go out and get the next respondent and bring them to the room, chatting with them on the way and creating initial rapport to put the respondent at ease. When time is short between interviews, RIVA moderators use the walk down the hall to give guidelines and a context for the discussion:

"Sometimes these interviews are done in a group of eight people, but today we are talking to consumers one at a time to get personal reactions (to ads, products, ideas, etc.). I’m really interested in your viewpoint and will be tape recording and videotaping [if true] this session so I can compare your comments with others and find out the trends and themes to put into my report. Your name won’t be in the report. You get paid in cash at the end of this session and I want to say, right now, how happy I am you could fit this session into your schedule. There are no wrong answers today, just your opinions about x, y, z."

After entering the room, point out where they will sit, reference any beverages available, and indicate the source of the microphone (and/or videotape camera). Indicate that observers are behind the mirror because "they can’t wait for the report and they want to see the trends as they emerge, person to person, day by day." Recap any key guidelines such as "speak loudly, say what you believe, it’s O.K. to change your mind" etc. Make sure the first question is an easy one. This gives them time to talk aloud and hear their own voice in the room and allows the moderator to demonstrate listening. Questions that seem to work best include: "Tell me about . . ." or "What is it like to . . ." or "What have you seen, heard or been told about . . ." or "When was the last time you did ‘x’ and what was that like?" These questions open the door to an interchange that gets the interview off to an involved start.

Conducting the interview

Keeping a little clock to see the movement of time and pre-thinking the timing for each portion of the guide helps avoid rushing the respondent through the process. Give instructions for any procedures (e.g., "Open the package and tell me what you are doing at each stage," or "Pick up each of the prototypes, feel them and be ready to tell me the benefits and drawbacks of each one before moving on to the next one.") It is a good idea to write instructions down and give a copy to the respondent so they are clear about what they are supposed to do and in what order.

When a respondent is made to feel like a research partner rather than a research subject, the benefit is a deeper level of communication and more fuel for the research furnace. Whenever possible, make sure that the respondent is never confused or murky about what they are supposed to do besides just answer questions posed to them.

When designing the interview process for IDIs it is good to remember that efficient use of time and respect for respondents should be paramount. Some guidelines to consider:

  • Respect respondents by being honest about generic topic content and time set aside for the discussion.
  • Avoid rushing through the set-up (i.e., disclosures about taping or mirrors and ground rules for participation) and allow time for respondents to settle into the research environment.
  • Plan the discussion in a logical format so the respondent sees a pattern or evolution of the relationship of your questions and their answers. In interviews longer than 30 minutes, it is a good idea to provide a generic road map of key items for respondents to look at. For example: "Today we’ll talk for a few minutes about products, then I’ll show you some materials and then ask you to pick a favorite."
  • Let the respondent know they are doing a good job by occasional praise, eye contact or a smile.
  • Encourage participants to express big thoughts and small thoughts equally and to avoid editing comments because they feel either the answer is obvious (in their minds) or they feel it is not worthy of comment.
  • Allow enough time for respondents to think before responding, especially in the later interviews when the moderator is now able to anticipate the kind of answers that emerge.
  • Telegraph (by word or deed) that the interview is coming to a close so that the respondent can volunteer additional information they may have been holding onto while waiting for a question to give them an opportunity to comment.
  • At end of session, thank the respondent for their contribution to the research study and for the "difference they made" by participating.

Organizing materials

This function is what separates good IDIs from poor ones. Typically an IDI process has "manipulatives," i.e., physical items that can be moved about by either the moderator or the respondent. They may include checklists or items housed in clear plastic sleeves (e.g., brand names of soap powder). Items such as prototypes of products or pictures or architectural drawings of the "gas station of the future" might be shown. Actual product items such as six brands of cat food might be displayed. Visual items such as storyboards for TV spots or sample brochure pages or a rough edit of a TV ad could be presented. Flattened boxes of pasta could be shown or incentives intended as give-aways for buying the two-liter size of a popular beverage.

Usually, it is prudent to keep these items hidden or covered until the appropriate time in the interview. If there are multiple manipulatives and very short lead times between respondents, the organization of materials becomes critical to saving time and presenting items in a neat and orderly way. If items are thrown in a box and the moderator has to sort them out during the interview process, that takes up expensive time. If the moderator is showing three commercials on one reel and there is insufficient time between ads to insure that each ad gets full play with no cut-offs, it might be best to have each ad on a separate videotape. Paying that kind of attention to how materials are organized creates more successful IDIs.

RIVA has some procedures to help moderators when lots of paper items are presented. By using colored Xerox paper for each new item shown (e.g., checklist, brand names, worksheet, sample names for new hair coloring, etc.), visual interest is created for the respondent and moderators have a visual anchor when they set up the flow of the conversation and execute it. It is easy to see the yellow versus the green document and avoid rummaging through a stack of all white documents to make sure the right one is presented in the correct order.

If we are showing tipped-in sample ads or existing ads in a magazine, the pages are pre-tabbed to find them easily. Respondents are less likely to become engaged in the other magazine content if they know exactly where to look. If the IDI includes showing video tapes, the machine should be close enough to the moderator to insert the tapes without standing.

RIVA moderators find it useful to conduct several mock interviews with staff or family members to perfect the flow of the discussion and to test the manipulatives. Timing those mocks will also point to where streamlining is needed. It is not a good idea to use the first two actual interviews for this process because it creates tension for the moderator and dissatisfaction for the client, who is excited about the IDIs at the outset.

Recording data

Depending on the type of interview, moderators can create a cheat sheet to record a few key answers to support the analysis phase of the study. It is not a good idea to take expansive notes because it distracts from the intimacy and takes on the look of a survey. The cheat sheet might be a chart to check off plus or minus features of a prototype, or it might be boxes for key words that describe a product, or it might be a copy of key questions from the guide, formatted to allow 1-2 inches of space between questions for jotting some quick notes.

When clear trend data can be easily collected over the series of IDIs, use an easel pad to mark columns or boxes in a pre-made chart with respondent order numbers down the side and options across the top. For example, if respondents are asked to choose a favorite among three options labeled P, Q and R, mark an "X" in the row for the winning option as respondents are leaving the room, and then cover the chart so that incoming respondents can’t see the trends! Position the easel so that it can only be seen by observers. From time to time show the chart to the clients so they can see the trends emerge. Showing trend data like this helps keep the client focus on the key issues of the IDI.

Pricing IDIs

There are two primary models for pricing focus groups: the flat rate all-in-one cost for all services related to conducting a series of focus groups, and the line-item method where individual costs are provided for each service (e.g., recruiting, room rental, moderating, analysis, travel, etc.). Pricing IDIs is more difficult because more variables are at play. Focus groups are traditionally two hours long and traditionally two are held per day. IDIs vary in length (15 minutes to 120 minutes) and different numbers of them can be completed in a work day that may extend from 9 a.m. to 9 p.m.

In conversations with moderators about this issue, the following models for pricing IDIs emerged:

Type A pricing: per head costs

Type B pricing: per day costs

Type C pricing: focus group parameters

Pricing IDIs is more a matter of personal accounting style than any prescribed industry standard. The figures given below are intentionally low so as to provide the mathematical thinking behind the pricing approach, rather than a suggestion of actual costs.

Type A: Moderators who set fees based on a per head model indicate that they set a fee for each interview and that fee includes all the estimated costs expended to complete an interview and analyze the findings. For example, if they charge $10/head for each one-hour interview, that $10 has to cover the costs of recruiting, room rental, interviewing, paying stipends, feeding clients, moderator travel and preparing a report. By having a per head cost, an estimate can be prepared for client review and project costs can be adjusted as budget constraints come into play.

Type B: Moderators who charge a per day cost (for example $150/day) indicate that this flat day rate allows the client to choose any model they want (i.e., 10 30-minute interviews or six 45-minute interviews or five one-hour interviews). The costs are calibrated to cover all the same costs from recruiting to final report. Moderators indicate that the benefit of this model is that they don’t get into nickel-and-dime project pricing.

Type C: Some moderators calibrate IDI fees based on the costs of doing focus group research since that pricing approach is familiar to clients. By apportioning costs on a traditional research model for two focus groups in a day, a per head rate is constructed. For example, if a moderator charges $1,000 for a focus group (for all costs from recruiting to final report for that group) and divides that number by eight respondents (the ideal number for a focus group), the per head cost would be $125 per person. If eight respondents were interviewed in a day in an IDI format, the same money is charged as if they had participated in a focus group.

Analyzing IDI data

Two types of reports can be written based on IDI research: memo reports recapping plus-and-minus aspects or a detailed report indicating insights about broad themes across the series. When IDI sessions are short, i.e., 15 minutes, a memo report is probably sufficient to capture the "tops of the mountains." Longer sessions, with more detailed activities, may require a longer, more detailed report with illustrative quotes, or tables that compare options presented. Analytic text blocks to help the reader understand the results and what they mean, along with implications or recommendations, complete this type of report.

For the detailed report option, RIVA recommends the use of written transcripts to avoid the tendency to report head counts and to step away from selective memory of what happened. Since IDIs are conducted to illuminate nuances, those nuances need to be carefully reviewed and reported in a final document that is intended to support client understanding. As with focus group reports, it is good to include, in the appendices, a copy of the guide and worksheets or descriptions of stimuli used in the interviews.

Deep and rich

In the hands of a skilled interviewer, who shows real respect for respondents and the research rigor to use interviewing time effectively, IDIs can help clients get a revealing look into the thinking of their target market and the language used to describe usage and attitudes. Insights can be deep and rich.

Focus groups require "flathead screws" to hold in place a stable platform of ideas. Because there are more respondents in a series, the insights tend to be broader than in IDIs. On the other hand, IDIs are best held in place with "Phillips screws" due to the precision needed to lock down the research planks. Using the right tool to anchor those screws is important, just like my dad taught me.