Focus Groups · Business Analysis Notes

What it is#

The BABOK Guide describes a focus group as a way to elicit ideas and opinions about a product, service or opportunity in a structured group setting. It is qualitative research: you learn why people think what they think, not how many think it.

Key features, as the training materials summarize them:

Participants: usually 6 to 12 people, selected to match the purpose of the group, who are willing to share their views and listen to others.
Roles: a moderator who runs the session and keeps it on topic, and a recorder who observes and takes notes. Neither takes part in the discussion, and combining the two roles is best avoided.
Discussion guide: prepared in advance, covering how to open, the sequence of topics and questions, how to collect comments and how to close.
Session: typically one to two hours. It feels relaxed and unstructured to participants, although the moderator is following a plan.
Analysis: done straight after the session, while memories are fresh and before interpretation drifts.

When to use it#

Use a focus group when:

you want reactions to a concept, prototype or proposed change from a defined group, such as customers or front-line staff;
you need to understand attitudes, expectations, language and concerns;
hearing others’ views will help participants articulate their own;
you want several perspectives in limited time.

Do not use it when:

you need numbers or representativeness: use a survey;
you need to know what people actually do, which often differs from what they say: use observation;
the topic is sensitive or private, which makes people hold back in a group: use interviews;
you need decisions or agreed requirements from stakeholders: run a workshop.

The training materials also flag that a group can be too similar to surface different views, that one dominant participant can skew the outcome, and that a skilled moderator is needed.

How to do it#

Set the purpose. Write the two or three questions the session must answer and how the results will be used.
Define the participant profile. Choose who you need to hear from. Run separate groups for clearly different segments rather than mixing them.
Recruit 6 to 12 participants per group, plus a few extra in case of no-shows. Avoid people who know the project well or report to each other.
Write the discussion guide: opening and ground rules, a warm-up question, main topics from general to specific, any material to show, and a closing question.
Choose the moderator and recorder. The moderator should be neutral; the product owner or system designer is usually a poor choice.
Run the session. Open with the purpose and how notes will be used. Ask open questions, invite quiet participants, and gently limit dominant voices.
Debrief immediately. Moderator and recorder list the main themes, surprises and memorable quotes within the hour.
Analyze and report themes with supporting comments, and turn them into requirements questions, hypotheses or design input.

Example#

The Workshop (a fictional chain of bike-repair workshops) plans repair-status notifications and wants to know what customers expect.

Purpose: Which status updates do customers value, through which channel, and how often is too often?
Participants: two groups of eight. Group A: commuters who depend on their bike daily. Group B: leisure riders with seasonal repairs.
Moderator: an outside facilitator. Recorder: the business analyst.

Discussion guide outline (90 minutes):

Text

0-10   Welcome, purpose, ground rules, how notes are used
10-20  Warm-up: "Tell us about the last time your bike was in for repair."
20-45  Waiting for a repair: what do you want to know, and when?
45-65  Show three sample messages (SMS, email, app). Reactions, likes, dislikes.
65-80  How many messages is too many? What would make you switch them off?
80-90  "If you could change one thing about getting your bike repaired, what would it be?"

Findings, after the debrief: commuters cared most about a reliable “ready for collection” time and wanted SMS; leisure riders were relaxed about timing but wanted a cost estimate before work starts. Both groups disliked more than two or three messages per repair. These become hypotheses for the notification requirements and for a quick prototype, not final numbers.

Common mistakes#

Treating results as statistics (“75% preferred SMS”) → report themes and reasons; size them with a survey if needed.
A moderator with a stake in the answer → use a neutral moderator and keep designers in the observer role.
Mixing very different groups → run separate sessions per segment.
Letting one voice dominate → invite others by name and use round-robin questions.
Leading questions → ask open, neutral questions and show options side by side.
Delaying analysis → debrief right after the session.

Checklist#

Purpose and expected use of the results are written down.
Participant profile is defined; groups are separated by segment.
6 to 12 participants recruited, plus spares.
Discussion guide is prepared and timed.
Neutral moderator and a separate recorder are assigned.
Consent and note-taking are explained to participants.
Debrief is held immediately after the session.
Findings are reported as themes and linked to next steps.

Sources and further reading#

IIBA. A Guide to the Business Analysis Body of Knowledge (BABOK® Guide), version 3. International Institute of Business Analysis, 2015, technique 10.21 “Focus Groups”.
Business analysis training materials (Elicitation and Collaboration module: focus groups).