Focus groups are stereotypically the domain of big brand Fortune 500 companies and U.S. election campaigns.

“It tested well in focus groups” is a phrase often thrown around in movies and TV shows, leading to a general perception that major commercial decisions are made as a result of something that maybe just a handful of people said.

For this reason, they are also heavily associated with B2C vs. B2B research.

But, that need not be the case. They can be used for both.

In a traditional setup using external market research resources, they can also be prohibitively expensive to conduct for startups and smaller companies.

Again, this need not be the case.

If you work in an early-stage B2B or B2C startup that lacks the budget for a focus group study, or, if you work in a department that lacks the budget, this avenue is still an option.

You will just need to approach it differently to bring it within budget, by conducting lean focus groups. Which, I will present to you below.

But, why would you want to?

Let’s first explore that.

The Purpose of Focus Groups

The comfort zone for face-to-face primary customer research at startups is customer interviews.

It’s straightforward in principle. You get a prospect on a call or a coffee shop chair, then hit them with a torrent of questions.

This is super valuable, of course, but it has its trade-offs vs. focus groups.

These trade-offs can be thought of as ‘blind spots’ in how prospects think and perceive.

How so?

You never get the ‘full picture’ from speaking 1:1 with prospects.

Here’s a bunch of reasons why:

Readily-available knowledge. You are only accessing readily available knowledge: schema (units of knowledge) in the mind that are either stimulated regularly or are connected with other schema that are stimulated regularly.

Structured thinking: You are mostly accessing already structured thinking. Or, accessing perceptions and thoughts that are easy to structure on the fly. There’s a lot of unstructured data and perceptions in the prospect’s mind that can’t be readily orchestrated into a coherent thought or opinion without non-trivial cognitive computation (but still affect buying behaviour).

Subconscious decision making. Harvard Business School professor Gerald Zaltman says that 95 percent of the decision-making processes associated with purchases take place in the subconscious mind.

Groupthink Bias. You only get an indirect view of how groupthink bias affects how the prospect behaves and makes conclusions and choices, which can make it difficult to isolate and understand the social mechanics behind it.

Misperception. Prospects won’t always offer up information readily on particular topics because they perceive it is not relevant to the conversation (when in fact it is). Or, they just don’t think of it in the first place. The connection is not drawn.

Heuristics. The prospect’s mind uses heuristics (mental shortcuts) to evaluate and make choices. These are not always easy for the prospect to recognise, internalise, or explain because they are not purely logical or even conscious functions. It is difficult for prospects to introspect into this.

Emotion. Prospects do not always feel comfortable sharing specific perspectives and behaviours 1:1 that they might otherwise do in a social context with peers.

Context. People perceive and behave in different ways depending upon the context. The context you are interviewing them in is radically different from most other scenarios in their lives. Context effects behaviour.

Focus groups are a research tool you can use to peer a little into some of these domains.

How so?

A group dynamic stimulates the minds of ‘respondents’ — the term for prospects participating in a focus group — in a way that 1:1 interviews do not.

Respondents ‘bounce’ off each other, mentally.

A group brainstorming session is a cousin to this dynamic. Except, instead of brainstorming ideas, the objective is to tease out perceptions.

Perceptions shared by one respondent can trigger associations and memories in another that would not have surfaced otherwise.

The diverse range of feedback given can provide more confidence to share points of view the respondent may have thought were irrelevant, or, simply were not triggered by the 1:1.

The dialogue can help structure unstructured perceptions and fragments of cognitive data that were previously not organised because they were not consciously probed.

You also get insights into ‘priming’ and ‘groupthink’ — how ideas and perceptions can spread from one person to the next in a social context and cause biases.

Overall, a runaway effect can occur: one respondent stimulating the next, and so on, generating insights that may not have otherwise occurred in a 1:1 interview because the startup did not know the right questions to ask in the first place.

You don’t know what you don’t know.

Trade-offs

There are of course trade-offs with focus groups vs. 1:1 customer interviews.

The big ones are:

• Biases. Some respondents are swayed by the opinions of others.

• Signalling. People behave differently in social settings than in private.

• Limited Depth. Due to time constraints, you can’t go deep on topics.

• Dominant Voices. One or two voices can dominate the group.

The Value of Focus Groups

Steve Jobs once said “It's really hard to design products by focus groups. A lot of times, people don't know what they want until you show it to them.”

This is why it is important to be aware of the limitations of focus groups as a research device and make sure they are designed appropriately so that useful commercial insights can be drawn.

Focus groups do not serve as a ‘silver bullet’ solution from which to get all the answers to design a product and build an entire go-to-market strategy — as they are sometimes held up to be.

If that were the case, everyone would conduct focus groups and everyone would have commercial success with everything.

The value of focus groups is to generate ideas and establish directions as to what a company could explore to be more competitive. To formulate hypotheses and tease out hidden opportunities.

It’s idea exploration and generation, not idea validation.

They are exploratory platforms designed to gather a wide array of sentiments and opinions, providing an additional dimension to become aware of the potential challenges and opportunities a company faces.

It’s less about asking ‘what people want’ and more about discovering ‘what people perceive’ — probing into the mental background behind a category of product and its usage.

For example: What are the underlying forces and perceptions driving decisions at the category level? How is a category and the behaviours around it perceived? How is a newly introduced idea within a category perceived?

But, caution should be taken here. The value of this research approach lies in discovering what is going on in the minds of prospects, not whether you should do anything about it.

Focus groups exist to discover the range of perceptions that inform choice and guide decision-making processes.

Once these have been discovered they can be combined with other data points and commercial insights. Then, hypotheses can be developed based on the merits of the most promising directions holistically.

Since it is a qualitative form of research, the output of a focus group study is not to develop a consensus upon which a commercial decision should be made in isolation. Instead, it provides clues as to how to move forward.

No value should be assigned on whether one perception in a group is more important than another. Within the group, it is impossible to know whether the view of one person is representative of them alone or a zillion prospects. The real value is in knowing it exists in the first place.

Put another way: the objective is not to quantify or determine the widespread nature of certain attitudes and viewpoints, but, to surface the various decision-making patterns and the spectrum of perceptions that exist.

By the end of the study, the startup should be in a more informed position to make a list of initiatives that have the potential to make its offering more competitive.

So, what’s a “lean focus group” and how do you go about doing it?

Lean Focus Groups

A lean focus group is a method to obtain similar results to a regular focus group but without the proportionate level of expense.

To be clear, I am not suggesting lean focus groups deliver the same quality of data, just that they can provide valuable insights inexpensively by following a similar process.

Whilst this approach may not work for established consumer-packaged goods brands and the Procter and Gamble’s of the world, if you aren’t P&G you shouldn’t act like P&G.

Speed is of the essence in startups, not perfect research hygiene, so it’s a better approach to learn as much as possible as fast as possible. Focus groups are a way to do that.

Erroneous insights can be discarded quickly by cross-referencing them with other learnings, their relative strategic fit, and by testing them in growth experiments.

After the focus group study has finished, you can recruit respondents to join 1:1 sessions at a later date, going deeper on certain topics and building a richer mental picture.

Since focus groups are not typically the domain of early-stage startups, they represent a novel opportunity to build bridges with prospects that may decline other forms of dialogue and research.

This can be a powerful knowledge and relationship-building trojan horse — particularly in B2B.

Psychologically, people like to feel their points of view matter and seek to have that need satiated. A focus group can deliver the appearance of an official and peer-endorsed forum for which to do this.

📘 Playbook

OK — so here’s how to do it.

Preparation

Here’s a list of things you need:

🕛 Agenda. To define what you want to learn and achieve. This informs what respondents you will recruit and what questions to go through.

For example, if your product is a sales tool for SDRs, maybe you want to learn more about the daily behaviours and habits of SDRs and how they perceive both the category in which your tool sits and adjacent categories.

🚪Private room. This is the location where you will conduct your focus group. This should be quiet, comfortable, and relaxing — with a table and chairs, and drinking water.

Hopefully, you have a spare office room you can use or borrow, if not a room in a co-working space is an option.

If you can’t do it in real life — which is preferable — remotely will suffice.

👥 Respondents. At least 12 respondents, split into two separate focus groups (6 in each). You will need to recruit them yourself.

In B2C market research, respondents are typically paid $50-$100 or so each (this is the bait). If your budget is super tight I suggest being creative and trying to recruit respondents based on the merits of the focus group itself.

In B2B, a small cash incentive is unlikely to entice a prospect to join the group. They are much more likely to be engaged by pitching the focus group as a forum to discuss something relevant and important to their role and position.

For both of these scenarios, using the ‘founder card’ can help a lot, so lean into this.

Here are some ideas on how to reach respondents:

B2C: social media, online communities, online classifieds, Reddit, job boards, events, customers, and leads.

B2B: cold outreach (email, LinkedIn), networking, events, social media (LinkedIn, Twitter X), online communities, customers, and leads.

Aim to achieve a much higher ratio of non-customers to customers. They should fit your ideal customer profile and represent the range of personas that exist in that profile for each buyer role.

As part of the onboarding process, you can potentially request each respondent fill out a questionnaire before the focus group takes place. This can ask questions about their circumstances, behaviours, and habits.

By the time you have finished recruiting, you want to have a reasonable level of confidence that the known range of perceptions that exist are present in the room

🗣️ Moderator. You will need someone to moderate the group. This can be someone from your team, whoever is most comfortable doing it.

The vibe of this role is usually quite formal and clinical, but in a startup setting it could be more relaxed. The main requirement is not too heavily bias the respondent’s answers, so it is essential to foster an approachable and open relationship with the group — they are not there to ‘please’ the moderator.

📋 Discussion Guide. This is a set of questions the moderator will take the group through. There are a range of structures you can use to build your discussion guide. I suggest the ‘funnel’ technique because it enables you to start with broader, general questions, and then narrow down to more specific topics.

This is a useful format to tease out background perceptions and behaviours that both consciously and unconsciously flow into and inform decision-making closer to and around the specific category your product sits in.

It can help surface insights that conflict with tacit assumptions you have made about your category and value proposition.

It is also a good format to ‘warm up’ respondents, transitioning them from higher-level topics to specific ones with a dialogue flow that feels natural — like a conversation.

How you can build your discussion guide on this structure:

What this means:

• Background. Questions that tap into supporting behaviours and perceptions. If your product is a sales tool for SDRs, this would be their workflow.

• Category. Questions that tap into perceptions around the category in which your product sits and adjacent categories.

• Value Proposition. Questions that surface perceptions around the area your value proposition sits. You can introduce product ideas, concepts, and prototypes here.

For the entire discussion guide, you need to make sure that every question you add is not a leading question. Doing so pollutes the results.

This is a lot easier said than done. If you are not confident doing this, I suggest checking out a book called The Mom Test — prompting ChatGPT about it and asking for practical examples should suffice.

💻 Laptop/camera. A device to record the discussion from start to finish. Don’t forget to hit record!

📛 Name tags. Folded up pieces of paper or card, with each respondent’s name on it. Place this on the table in front of where they will sit.

📝 Consent forms. To be signed by the respondents. Here is an outline of what you need (in the UK). You could get ChatGPT to produce it.

🤗 Product. This is your product direction, concept, prototype, or demo that you seek to generate insight into. You will need to organise this logistically, according to the nature of the product and the group.

✍️ Questionaire. A questionnaire, for respondents to provide their feedback on the product they have been exposed to. You will need to design the questions yourself, based on what you want to learn. On this, you can ask both open-ended and close-ended questions (more context here).

Moderating

Once you are fully prepped it’s time to assemble the respondents and kick off the focus group. To get a sense of what that looks and feels like, check out this University of Minnesota video.

In a professional setting, moderating is hard to do proficiently. Making the group feel comfortable to engage, whilst balancing the need to work through a set discussion guide without introducing moderator bias is non-trivial.

Since this is a ‘lean focus group’ we can modify the expectations accordingly.

Hubspot has a guide on how to run a focus group. Here are some principles to follow:

• 🎯 Set expectations from the beginning

  • The moderator can’t be upset or pleased by anything said in the group. A neutral party.

  • All answers are valid. No answer given is right or wrong, dumb or smart.

  • The moderator may need to ‘cut off’ a respondent mid-answer, to progress to the next question or stay on track.

  • Don’t dramatise the situation. Keep it chill.

• 💬 Stimulate discussion. It’s not a Q&A.

• 🔙 Use “think back” questions — take respondents back to an experience.

• ⏱️ Pacing is important. You need to get through the discussion guide.

• 🚫 Refrain from asking leading questions.

• ❓ Questions should be open-ended, not yes or no, except when probing to learn if a perspective is held by more than one person.

• 🛑 Refrain from using examples (e.g. a brand) to stimulate dialogue.

  • Questions like “How else would you describe that?” can probe respondents to elaborate without introducing bias.

• 🔎 If a respondent’s answer is vague, probe into it.

  • If something is “practical”, what does “practical” mean?

• 🛡️ Avoid asking “why?” — it can make respondents feel defensive.

  • Reference their answer. “What about X made you feel…”

• 🗣️ Referring to respondents by name will help build rapport.

• ⚡ Be energetic. The moderator’s enthusiasm is mirrored by the group.

• 🏃Try to build a sense of momentum, bouncing dialogue from one respondent to the next.

• 🥼 Don’t get emotional or defensive. Be more like a curious scientist.

• 📊 Try to ensure every respondent speaks to an equal degree.

• 🛤️ If a respondent goes off-topic, bring them back on track.

• ↔️ When a respondent gives a point of view, ask if anyone does not share the same view. Bounce perspectives off each of them.

• ⚖️ Try to recognise if a respondent has given two separate perspectives that conflict with one another or build on each other, cross-referencing them both and putting it to them.

• 🧠 Try not to make mental leaps and let this influence your questioning. Only work with what respondents have said, like nothing else exists in your mind.

• 🗺️ You want to work your way through the discussion guide but not stick to it religiously if relevant organic dialogue develops in the group. But, don’t dive too deep.

• 🤫 Respondents should answer their product questionnaires in silence, to not bias the responses of the other respondents.

• 🔄 If respondents start to give you their perspective on the market and what others are likely to think, guide them back to share their views and perceptions.

Once you have reached the end of the discussion guide you can thank people for attending and wrap it up.

Oh. And never… never… ask “Would you buy this?” 😉

That’s it for today. I’ll be back in your inbox soon.

Martin 👋

Did someone forward you this email? Happens allll the time.

Subscribe to receive more content like this, weekly(ish). 👇