Market research does not require an expensive database or a large survey. With a clear question and a disciplined process, you can use public sources to understand customers, competitors, demand, pricing, regulations, and market risks before making a business decision.
Start with a specific research question
“Research the market” is too broad to guide useful work. Begin by writing the decision your research should support. For example:
- Should we launch a meal-planning app for busy parents?
- Which customer segment is most likely to buy this service?
- What price range do customers already accept?
- Which competitors serve small businesses rather than large enterprises?
- Is demand growing, seasonal, or declining?
A good research question identifies a market, a customer, a problem, and a decision. You may also define a time period and geographic area. “How large is the online accounting software market?” is less actionable than “Which accounting problems do self-employed professionals in the United States actively pay to solve in 2026?”
Create a short research brief before searching. Include:
- The decision you are trying to make.
- The target customer or user.
- The geographic market.
- The product or category being studied.
- The questions that must be answered.
- The evidence that would change your decision.
This prevents attractive but irrelevant statistics from taking over the project.
Build a source plan
Public sources vary widely in quality and purpose. Plan to use several source types because no single source usually explains the whole market. A company press release may describe its growth but omit weaknesses. A government report may be authoritative but too broad for a specific niche. Customer reviews may reveal problems but cannot establish total market size on their own.
Useful public sources include:
- Government statistics agencies and regulatory bodies
- Census, labor, trade, business-registration, and economic databases
- Company websites, pricing pages, annual reports, investor presentations, and filings
- Industry associations and professional organizations
- Academic papers, university research centers, and library guides
- Trade publications and specialist newsletters
- Search-engine results, autocomplete suggestions, and related searches
- Online marketplaces and retailer listings
- Customer reviews, discussion forums, support communities, and social posts
- Job boards and job descriptions
- Public procurement records and tender databases
- Patent databases and product documentation
- Conference agendas, exhibitor lists, and public presentations
Give priority to primary sources when possible. A regulator, company filing, official dataset, or original research paper is generally more useful than an article that summarizes it. Secondary sources remain valuable for discovering terminology, competitors, and leads, but trace important claims back to the original material.
Define the market carefully
Market boundaries determine your results. A market can be defined by product, customer, use case, channel, geography, or price. If you change the definition halfway through the research, figures that appear comparable may not be comparable at all.
Write down your working definition, such as: “Subscription-based project-management software purchased by teams of 5 to 50 employees in the United States.” Then record what is excluded, such as enterprise consulting, one-time desktop software, and personal task apps.
Separate these related ideas:
- Total potential market: Everyone who could theoretically use the product.
- Addressable market: People or organizations you can realistically reach and serve.
- Current demand: People actively searching, buying, switching, or requesting the solution.
- Revenue opportunity: The amount that could be earned under specific price and adoption assumptions.
Do not treat a published market-size estimate as automatically answering all four questions. Reports may use different definitions, include adjacent categories, or rely on proprietary assumptions.
Research customers and problems
Start with the customer’s problem rather than your preferred product. Search for the language people use when describing the situation, including complaints, workarounds, costs, delays, and desired outcomes.
Useful searches include combinations of the category with terms such as:
- “alternative to”
- “best” or “comparison”
- “pricing” or “cost”
- “problem” or “not working”
- “how do I”
- “template” or “spreadsheet”
- “review” or “complaint”
- “request” or “recommendation”
Read several pages of results instead of relying only on the first few links. Look for recurring phrases and classify them into a simple problem map:
| Evidence | What it can reveal | Main limitation |
|---|---|---|
| Search queries | Interest, wording, and possible intent | Does not prove purchase intent |
| Reviews | Benefits, frustrations, and missing features | Often biased toward extreme experiences |
| Forums and communities | Context, objections, and workarounds | Samples are self-selected |
| Job postings | Tasks, tools, and organizational needs | Employers may describe ideal roles |
| Support documentation | Common setup and failure points | Reflects one provider’s users |
When collecting customer evidence, capture the exact source, date, audience, and relevant quotation or paraphrase. Do not count every mention as a separate customer. One discussion may be copied across multiple websites, and one prolific poster may create a misleading impression of popularity.
Distinguish between a problem people acknowledge and a problem they will pay to solve. A complaint about an expensive service is evidence of dissatisfaction, not proof that a cheaper competitor will succeed.
Estimate demand without false precision
Public data rarely gives you a perfect demand figure. Instead, combine several indicators and make your assumptions visible.
Search trends can help identify seasonality, rising interest, declining interest, or differences between locations. Compare related terms rather than examining one phrase in isolation. For example, a decline in one keyword may reflect a change in terminology rather than a decline in the underlying need. Search-volume tools may also show estimates rather than actual counts, so treat them as directional evidence.
Other demand signals include:
- The number and freshness of competing products
- Marketplace listings and review volume
- Waiting lists, preorders, or public launch announcements
- Repeated requests for recommendations in relevant communities
- Hiring activity connected to the category
- Public spending, procurement awards, or import data
- Website traffic estimates and backlinks, used cautiously
- Pricing changes, product expansions, and new distribution channels
Use a confidence label for each finding: high, medium, or low. A high-confidence finding might be supported by an official dataset and multiple independent sources. A low-confidence finding might come from a single estimate or a small group of online comments.
You can create a simple demand score, but do not present it as a measured market size. For example, score each signal from 1 to 5 for strength, recency, and relevance, then explain the scoring rules. The score is useful for comparing opportunities consistently; it is not proof that one market is five times larger than another.
Analyze competitors and substitutes
List direct competitors first: products that solve the same problem for a similar customer. Then identify indirect competitors and substitutes, including manual processes, spreadsheets, agencies, internal staff, and doing nothing.
For every competitor, collect the same fields:
- Target customer
- Core problem addressed
- Main features or service components
- Pricing model and visible price range
- Distribution channels
- Geographic coverage
- Positioning and claims
- Reviews and recurring complaints
- Evidence of recent activity
- Strengths that would be difficult to copy
Use competitor websites to understand positioning, but verify pricing and availability from current pages because these change frequently. Save the page date and note whether prices exclude taxes, require a sales call, or depend on usage.
A feature checklist is helpful, but customers rarely choose based only on feature count. Compare outcomes, trust signals, setup effort, support, integrations, switching costs, and the clarity of the buying process. A competitor with fewer features may win because it is easier to understand or easier to adopt.
Look for gaps carefully. A missing feature may indicate an opportunity, but it may also be difficult, unprofitable, restricted by regulation, or unwanted by most buyers. Validate gaps against customer complaints, product updates, and competitor responses.
Investigate prices and business models
Collect prices from official pricing pages, public quotes, retailer listings, procurement records, and customer discussions. Record whether the price is one-time, recurring, usage-based, commission-based, or bundled with another service.
Avoid averaging prices without separating unlike offerings. A basic self-service plan, a customized enterprise contract, and a professional service package should not be placed in one simple average.
Ask practical questions:
- What does the customer actually pay after setup, delivery, support, or renewal costs?
- Is the cheapest option genuinely comparable?
- Are discounts temporary or standard?
- Does the category rely on free trials, freemium plans, or annual contracts?
- What costs make switching difficult?
- Does a lower price create additional labor or risk for the customer?
Public pricing is evidence of a seller’s stated offer, not necessarily the final transaction price. Treat it as a range and explain the conditions attached to it.
Check regulations, technology, and external risks
A market can look attractive until you examine the constraints around it. Search official government and regulator websites for licensing requirements, consumer-protection rules, privacy obligations, advertising restrictions, safety standards, import rules, and professional qualifications.
Also investigate external dependencies:
- Changes in platform policies or search algorithms
- Reliance on a small number of suppliers
- Exposure to currency, shipping, or energy costs
- Technology that may become obsolete
- Public funding or reimbursement rules
- Intellectual-property restrictions
- Environmental or reporting requirements
- Concentration among a few powerful buyers or distributors
Use exact dates and document the jurisdiction. A regulation applying in one country or industry may not apply elsewhere. When a rule is unclear, record the uncertainty and obtain qualified legal or professional advice before relying on the research for a launch decision.
Organize evidence in a research log
A spreadsheet is usually enough. Create columns for question, claim, source URL, source type, publication date, access date, geographic scope, key evidence, interpretation, confidence, and follow-up task.
Keep facts separate from interpretation. For example:
- Fact: Three major providers publish annual plans between a specified price range.
- Interpretation: Customers may accept recurring billing, but the category appears price-sensitive.
- Open question: Are annual plans common among the target customer segment?
Use consistent source names and save copies or archived links where permitted. Public pages can disappear, change wording, or update figures without clearly showing the revision date.
Before writing a finding, try to disprove it. Search for contrary evidence, different regions, older data, and sources with a different incentive. If two credible sources disagree, do not silently choose the more convenient number. Explain the difference in definitions, dates, or methodology.
Turn research into a decision
Finish by connecting evidence to specific actions. A useful report should answer:
- Who appears to have the strongest need?
- What alternatives do they use now?
- What evidence suggests active demand?
- How do buyers compare options?
- What price and business model seem plausible?
- Which regulations or dependencies could block the idea?
- What assumptions remain unproven?
- What is the cheapest next test?
The next test might be a landing page, a manual service delivered to a small group, a structured interview campaign, a paid advertisement with a defined budget, or a direct sales experiment. The research should identify what to test, not pretend that public data has eliminated uncertainty.
Common problems and limitations
Public-source research is fast and affordable, but it has important weaknesses. Data may be outdated, duplicated, incomplete, geographically mismatched, or collected for a different purpose. Search behavior is not the same as buying behavior, public reviews overrepresent unusually happy or unhappy customers, and company claims may emphasize favorable metrics.
If sources conflict, check definitions and dates first. If you cannot find reliable market-size data, use narrower operational measures such as the number of potential accounts, visible competitors, public purchase records, or qualified search themes. If a niche has little online discussion, that may indicate low digital visibility rather than no demand.
Do not publish personal information gathered from forums, copy protected material beyond what is necessary for analysis, or present estimates as official facts. Clearly label assumptions, cite sources, and state the date on which the research was conducted. A transparent report with acknowledged limits is more useful than a confident report built on weak evidence.