Market research before a product launch: what actually works

Market research: where to start when you have an idea but no data

You have an idea for a product or service. You feel the market needs it because a few clients mentioned the problem. And here comes a question worth asking out loud: is that enough of a premise to spend a substantial budget on development and market entry?

Market research is a structured process of collecting data, not a one-off client interview or a review of competitor websites. It answers three questions: who has the problem, how they solve it today, and whether they are willing to pay for a different solution. Market research methods fall into two main groups: secondary data (already existing, collected by someone before you) and primary data (collected specifically for your needs). Order matters, because starting with the most expensive method is the simplest way to burn through your budget.

In practice, a good process looks like this: first desk research, then conversations with people in the market, then a survey on a larger sample, and finally a concept test. Each stage narrows uncertainty and lets you decide on the next step.

Desk research: what you can establish without leaving the office

Desk research is collecting data that already exists. It sounds unexciting, but in practice it lets you filter out ideas that have no chance before you spend a single unit of budget on field research. Typical sources include: industry reports, national statistics office data, chamber of commerce publications, competitor financial reports (if they are public companies), industry association analyses, and data from tools such as Google Trends or Senuto.

What specifically is worth establishing at this stage:

• Market size and structure - how many potential buyers are in your segment, how they are distributed geographically.
• Competition - who already sells a similar solution, at what price, with what distribution model.
• Barriers to entry - regulations, certifications, initial costs, customer loyalty to current suppliers.
• Purchasing cycle - how long the purchasing decision takes in your industry and who is involved in it.

It is worth remembering that desk research has a limitation: it shows what is already happening, but it does not answer whether customers will buy something that does not exist yet. That is why after this stage you move on to conversations with people.

B2B interviews and surveys: how to ask so the answers have value

B2B interviews and surveys follow different rules than consumer research. In B2B, the purchasing decision rarely belongs to one person, and the respondent often does not want to admit that their company has a problem, because it might reflect on their own negligence. That is why questions need to be formulated carefully.

In-depth interviews: 8-12 conversations that change the picture of the market

An in-depth interview (IDI) is a 30-60 minute conversation with a person who actually makes purchasing decisions in your category. You do not sell during it. You ask about the past, not the future. Instead of "would you buy our product?" you ask "when did your company last look for a solution for X and how did that process go?". The difference is huge: the first question triggers a polite "possibly", the second gives a concrete story with tool names, budgets, and decision-makers. In one project for a packaging manufacturer, the client spent three months comparing offers from four suppliers before signing a contract, and only in the interview revealed that the key criterion was not price but delivery time during peak season.

The minimum is 8-12 conversations in one segment. Below that number, conclusions are anecdotal. Above 15-20, answers usually start repeating, which means you have reached information saturation.

B2B surveys: short, specific, with questions about behavior

A survey makes sense only after interviews, because then you know what to ask about. In B2B the rule is: the shorter, the better. Use closed questions with one answer, comparative scales, and questions about frequency and budget. Avoid hypothetical questions like "would you be interested". Instead, ask about facts: "how many times in the last quarter did the company commission an external market research study?", "what was the budget range for that?".

In B2B, a large sample is hard to get, so 40-80 completed surveys in one segment is already a solid basis for conclusions. It is worth providing an incentive to complete: a report with results for participants works better than a coupon or voucher.

How to research the market before launching a product: concept testing on a small sample

Concept testing is the stage where you show respondents a preliminary product description, mockup, or prototype and check their reaction. In B2B, the question is whether they see a place for the product in their budget and purchasing process.

A practical way to run a concept test:

1. Prepare a short description - one page: problem, solution, target audience, indicative price.
2. Gather 15-25 people from the target segment, preferably outside your current client base.
3. Ask about their reaction - what is clear, what raises doubts, what is missing, how much they would be willing to pay.
4. Compare with alternatives - show competing solutions and ask which they would choose and why.
5. Ask about the next step - whether they would agree to a pilot, a conversation with a salesperson, a test in one department.

This last point is the most important. Declarations are cheap; a declaration of readiness for a pilot costs the respondent something real. If after 20 conversations no one wants to move to the next step, it is a signal that the concept needs redesigning, not that the presentation needs polishing.

A respondent who talks about how they solved a problem last quarter is far more credible than one who assures you they would "definitely buy".

How much it costs and how long it takes: realistic ranges for a B2B company

You can run market research in three ways: in-house, with a freelancer, or with an agency. Each comes with different costs and different risks.

Indicative market estimates for a small or medium B2B research project in Poland (2026), based on our own projects and conversations with vendors:

• Desk research - 2-5 business days, done in-house or outsourced for a low four-figure fee.
• In-depth interviews (10 conversations) - 3-5 weeks, with recruitment and a report included.
• B2B survey (50-80 responses) - 2-4 weeks.
• Concept test (20 conversations) - 2-3 weeks.

The full process, from desk research to concept test, usually takes 8-12 weeks and a mid-range budget. For comparison: the cost of a failed market entry with a product that has no demand often runs into the hundreds of thousands and several months of team effort. Market research is therefore relatively cheap when you treat it as insurance against a bad decision.

If you do not have someone in your company who can run this kind of process, it is worth considering external support. You can read more about what a marketing audit involves on the Growth Group site.

The most common market research mistakes and how to avoid them

From more than a decade of working with B2B companies, we see the same missteps again and again. Here are the ones that cost the most:

• Asking customers for opinions instead of behavior. A customer who says "interesting idea" is not the same as a customer who will sign an order.
• Researching your own customer base. Your current customers feel obliged to you and will be polite. For a concept test, recruit people from outside your base.
• Too large a sample at the start. Start with 10 interviews, then decide whether to run a survey at all. Not the other way around.
• Not asking about budget. If a respondent will not say what they pay today to solve the problem, you do not know whether your product has a chance at any margin.
• Treating research as a one-off event.Markets change, so it is worth repeating the research every 12 to 18 months, especially before any major change to your offering.

One last thing: market research does not give you a "yes" or "no" answer. It gives you a set of data that lets you make a decision with greater confidence. If you are looking for a way to structure this process in your company, a good starting point is a conversation about what you already know about the market and what is missing. Sometimes it turns out that half the data already exists inside the company, but nobody has gathered it in one place.