By 
Search “online focus groups” and half the results assume you want to join one for cash. The other half want to sell you a platform. Neither answers the question a founder or brand manager is usually asking: should I run one, what will it actually cost me, and is there a faster way to get the same answer.
This guide is for the second crowd. Here is what an online focus group is, the formats you can choose from, real 2026 pricing, the honest limits that the platform vendors skip over, and a decision table so you can tell in about a minute whether a focus group is the right spend or whether something quicker will do.
What an online focus group actually is
An online focus group is a moderated group discussion held over the internet instead of in a research facility. A trained moderator brings together a small group of participants, usually 6 to 9, and works through a discussion guide for around 90 minutes to two hours. Jakob Nielsen put the classic shape of it at 6 to 9 users for about two hours, run by a moderator who keeps the group focused.
The point is qualitative. You are not counting anything. You are listening to how people talk about a category, a product or an idea, watching them react in the moment, and picking up the words they use when nobody has handed them a script. Done well, a focus group is a good way to discover what people want and why, and to hear the language you should be using in your own copy.
That is the job it is built for. Most of the trouble with focus groups comes from asking them to do a different job, which we will get to.
The formats, and which one fits
“Online focus group” covers a few different setups. They are not interchangeable, and the right one depends on how much depth you need and how quickly.
Live video groups. The closest thing to the traditional room. Everyone joins a video call at the same time, the moderator runs the session live, and you can watch from the back channel. Good for energy and spontaneous back-and-forth. The catch is scheduling: getting 8 strangers free at the same hour, in the same time zone, is the hardest part of the whole exercise.
Asynchronous discussion boards. Sometimes called bulletin-board groups. Participants log in over two to five days and respond to prompts in their own time. You lose the live spark but gain more considered answers, the ability to include people across time zones, and time for participants to go away, try something, and come back. Often cheaper to run because there is no single live session to coordinate.
Mini-groups, dyads and triads. Smaller formats of two to four people. Useful when the topic is more personal, when you want more airtime per participant, or when recruiting a full group of a niche audience is not realistic.
One-to-one online interviews. Strictly this is not a focus group, but it is the format brands most often actually need. An in-depth interview (IDI) speaks to one person at a time, which removes the group dynamic entirely. If you are worried about conformity, this is the fix.
There is also a fast-growing set of AI-moderated and synthetic options, where software either runs and analyses interviews with real people or stands in for people altogether. Those deserve their own treatment, and I have written a full guide to AI focus groups that pulls the two types apart and covers how far you can trust each.
How much online focus groups cost in 2026
Here is the number people actually came for. A professionally recruited and moderated focus group is not cheap. Market research firm Drive Research puts the 2026 range at $7,000 to $20,000 or more per group, and that is per group, not per project. Published figures vary a lot by source and scope, and you will see leaner quotes of $4,000 to $7,000 for a stripped-back online group, but the shape holds: budget per group, and expect to run more than one.
Online formats sit at the lower end of that range, because going virtual removes several line items that in-person groups carry. There is no focus group facility to hire, no travel, and no catering, and recruiters can pull participants from anywhere rather than one city. What is left is the recruiting, the incentives and the moderator’s time.
Rough 2026 costs for a single online group, drawn from Drive Research’s breakdown:
| Cost component | Typical 2026 figure |
|---|---|
| Full group, all-in (online) | Lower end of $7,000 to $20,000+ |
| Participant incentive, general public | $50 to $100 per person |
| Participant incentive, business professionals | $200 to $250 per person |
| In-depth interviews (alternative) | $5,000 to $15,000 for 10 to 15 |
Two things founders underestimate. First, incentives scale with how hard your audience is to reach. A group of general consumers is cheap to recruit; a group of paediatric surgeons or B2B decision-makers is not. Second, you almost never run one group. Nielsen’s own guidance is to run more than one, because a single session may not be representative. Two groups is a realistic minimum, which quietly doubles the headline number.
So the honest budgeting rule for a proper online focus group study: think in the low tens of thousands, not the low thousands.
What focus groups are genuinely good at
I do not want to talk you out of focus groups. They earn their place for specific jobs.
They are strong at the exploratory end. Early in a project, when you do not yet know what you do not know, a group can surface the questions, objections and language you had not thought to ask about. Nielsen’s framing holds up: the proper role of a focus group is to discover what users want, not to predict what they will do. Use it to generate hypotheses and to collect the raw quotes that make a survey or a campaign sound like a human wrote it.
They are also good for reactions to something concrete. Put a real prototype, a pack design or an ad in front of a group and you will hear immediate, unfiltered responses and the reasoning behind them. That “why” is the thing a spreadsheet of survey scores can never give you.
Where they mislead
This is the part the platform pages skip. Focus groups have three well-documented failure modes, and every one of them is easy to walk straight into.
They record what people say, not what they do. This is the big one. A focus group can only capture what customers say they do, and as Nielsen put it, there are often major differences between what people say and what they actually do. People are bad at predicting their own behaviour, and worse at predicting it out loud in front of strangers. If you need to know whether people will buy, a room of people telling you they probably would is weak evidence.
The loudest voice sets the tone. Group settings breed conformity. NN/g’s own write-up of focus groups notes that one strong personality can shape what others reveal and disproportionately represent opinions. This is not a soft worry. In a controlled study, researcher William Boateng ran the same questions through focus groups and then through one-to-one interviews with the same people, and found the group answers departed quite significantly from what those individuals said alone, confirming that focus groups are not insulated from groupthink. The effect traces back to the classic Asch conformity work: put people in a group and a chunk of them will drift toward the majority view even when they privately disagree.
The numbers are too small to trust as data. Two groups is 16 people. That is a fine sample for gathering ideas and a terrible one for measuring anything. For comparison, Drive Research notes that even a proper survey needs around 400 responses to get the margin of error down to roughly plus or minus 4 to 5 per cent. A focus group is not built to be projectable, so never let a strong quote from one participant get promoted into “our customers think”.
And the online format adds a fourth failure mode the in-person room never had: you cannot always be sure who is on the other end. Online panels have a professional-respondent problem, people who join study after study for the incentive, and in 2026 a growing bot-and-fraud problem on top, with participants using AI to pass screeners they do not qualify for. In-person recruiting is not immune, but a webcam and a PayPal transfer are far easier to game than a person walking into a facility. If you run online groups, treat recruitment quality as something to actively police, not assume: real-time verification, attention checks, and a moderator alert to answers that sound rehearsed.
None of this makes focus groups useless. It makes them a discovery tool, not a decision tool. The mistake is spending $15,000 to hear eight people agree with each other and then treating that as a green light.
Focus group, or something faster? A decision table
Match the method to the decision in front of you.
| Your question | Best-fit method |
|---|---|
| What language and objections come up around this category? | Focus group or interviews |
| How does one person really reason about this, without peer pressure? | One-to-one interviews (IDIs) |
| What do people think of this pack, ad or prototype, and why? | Focus group, then validate at scale |
| Would people actually buy this, at this price? | Pre-spend concept or price test |
| Which of these three concepts should we back? | Concept test with a purchase-intent read |
| Is this true of our whole market, not just a few people? | Scaled survey (300+ responses) |
The pattern is simple. Use groups and interviews to work out what to ask. Use a scaled or purchase-intent method to work out whether the answer is real. Problems start when a group is asked to carry a decision it was never designed to hold.
Faster alternatives worth knowing
If the honest limits above describe your actual question, here are the routes that get you a cleaner answer, often faster and cheaper.
One-to-one interviews. The direct fix for conformity. You lose the group spark and gain honest individual reasoning. Better for sensitive topics and for anything where you do not want the confident participant setting the room’s opinion.
Asynchronous video-response tools. Platforms where people record short video answers to your prompts on their own time. Cheaper than a live moderated group, faster to field, and you still see faces and hear tone. The trade-off is a quiet sampling skew toward people comfortable talking to a camera, which is not the same as your average buyer.
AI-moderated interviews. Software runs the conversation with real people and writes up the themes, so you get interview-style depth at a fraction of the scheduling cost. Real human answers, faster turnaround, with the catch that the moderation is only as good as the model’s follow-ups. My AI market research guide covers where this is trustworthy and where it is not.
Pre-spend concept and product testing. When the real question is “will people buy this”, the fastest honest answer is to test the thing itself against your target audience before you spend on the launch. This is where testing an idea against real shoppers beats a discussion group: instead of asking eight people to imagine their behaviour, you put a product, pack, ad, price, claim or name in front of a defined audience and get a purchase-intent read back, with the reasons in shoppers’ own words, in days rather than weeks. TestFeed does exactly this, and gives you a directional signal and a clear next move before you commit budget. It is pre-spend signal, not a guaranteed sales forecast, and it will not tell you how something tastes or smells, so it sits alongside qualitative work rather than replacing it. If you are weighing up concepts specifically, the same logic runs through my roundup of concept testing platforms.
Scaled surveys. When you need to know something is true of your whole market and not just a vocal few, a survey of a few hundred people is worth more than any number of groups. Less colour, far more confidence.
A worked example
Say you run a DTC skincare brand and you are choosing between three names for a new serum. The default move is two online focus groups, eight people each, roughly $12,000 all-in and two to three weeks including recruiting and scheduling.
Run that, and here is the risk. In each group, one confident person says they love name B, and within ten minutes the room has quietly converged on B. You now have a $12,000 vote that was really one person’s opinion amplified by conformity, and you still do not know whether name B makes anyone more likely to buy.
The faster path: run a name test against a few hundred people in your target audience, where each person reacts independently and you get a purchase-intent read on all three. You lose the live discussion. You gain an answer that is not hostage to the loudest voice in the room, in a couple of days, for less money. Then, if you want the “why” behind the winner, book a short round of one-to-one interviews. Same insight, better order of operations.
That is the general lesson. Focus groups are a good first instrument and a poor final one. Lead with them to learn the shape of the question, and finish with a method built to predict.
Common questions
What is an online focus group? An online focus group is a moderated group discussion held over the internet rather than in a facility. A trained moderator guides 6 to 9 participants through a set of topics, usually for around 90 minutes, either live over video or asynchronously on a text-based platform over several days. The goal is qualitative: to hear how people talk about a product or idea and understand the reasoning behind their reactions.
How much do online focus groups cost? In 2026, a professionally recruited and moderated focus group typically costs $7,000 to $20,000 or more per group, according to Drive Research. Online formats sit at the lower end because they remove facility hire, travel and catering. Participant incentives usually run $50 to $100 for the general public and $200 to $250 for business professionals. Most projects need more than one group, so budget accordingly.
Are online focus groups accurate? They are accurate for exploring attitudes and language, and unreliable as a predictor of behaviour. Focus groups record what people say they will do, which often differs from what they actually do, and group dynamics pull answers toward the dominant voice. Treat the output as hypotheses and quotes, not a forecast, and validate anything important with a one-to-one method or a scaled test.
What is the difference between an online focus group and an in-depth interview? A focus group interviews several people together and relies on group discussion; an in-depth interview speaks to one person at a time. Interviews avoid group conformity and go deeper on individual reasoning, which makes them better for sensitive topics and for anything where you do not want the loudest participant setting the tone. Groups are faster for surfacing shared language and a range of reactions in one session.
What can I use instead of an online focus group? Depending on the question: one-to-one interviews for depth without conformity, asynchronous video-response platforms for cheaper reach, scaled surveys for a representative read, and pre-spend concept testing when you need a purchase-intent signal on a product, ad, price or pack before committing budget. Each trades depth, speed and representativeness differently, so match the method to the decision you are about to make.
The short version
Online focus groups are worth running when you are early, curious and want to hear people think out loud. Budget in the low tens of thousands for a real study, expect to run more than one group, and treat what you hear as the start of the question, not the answer. The moment your decision turns on “will they actually buy”, stop discussing and start testing against a proper sample. That is the difference between spending your research budget on colour and spending it on confidence.
