Short answer
- The marketing research process has six steps: define the question, plan the research, collect data, analyze it, decide, and check the result.
- Start with a decision you need to make, not with "let's do a survey".
- Small businesses can do most research cheaply: customer interviews, short surveys, Google Search Console, Google Business Profile insights, competitor checks and your own sales records.
- Research matters because it reduces expensive guesses: wrong offers, wrong prices, wrong channels.
Why market research is important
Every marketing decision is a bet: which customers to target, what to charge, which services to promote, where to spend ad money. Research doesn't remove risk, but it turns big blind bets into small informed ones. For a small business, the most valuable things research tells you are:
- Why people choose you (and why some don't). This shapes your website, ads and offers.
- What words customers use. These become your headlines and the searches you target.
- Which customers are worth the most. So you can spend more to win those.
- What competitors offer and charge. So you can position yourself instead of copying.
- Whether a new idea has demand before you spend on it.
The 6 steps of marketing research
Step 1: Define the decision and the question
Good research starts with a decision: "Should we add a maintenance plan?", "Why are quote requests falling?", "Is Facebook or Google a better place for our ad budget?" Turn it into one or two clear questions. If you can't say what you'll do differently depending on the answer, the research isn't needed yet.
Step 2: Plan the research
Decide what information answers the question and the cheapest way to get it. Some answers already exist in your data (secondary research): sales records, website analytics, Search Console, call logs, competitor websites and public industry reports. Others need new data from people (primary research): interviews, surveys, test ads. Use existing data first.
Step 3: Collect the data
Keep it small and focused. Five to ten customer conversations often reveal more than a long survey nobody finishes. If you survey, keep it short and ask about real past behavior ("How did you find us?") rather than hypotheticals ("Would you buy…?"). My guide on Market research questions has ready-to-use questions.
Step 4: Analyze
Look for patterns, not single quotes. Group interview answers into themes, count how often each comes up, and compare groups (new vs repeat customers, high-value vs low-value jobs). For numbers, simple tables and averages are enough; the point is to spot what changes the decision.
Step 5: Decide and act
Write down what you'll change: a new headline, a different offer, a budget shift, a new page. Research that doesn't change anything was just curiosity.
Step 6: Check the result
Measure whether the change worked: more calls, better close rate, higher average job value. This closes the loop and often raises the next question. See Marketing analytics for setting up the measurement.
Low-cost research methods compared
| Method | Best for | Cost | Watch out for |
|---|---|---|---|
| Customer interviews (10–15 min calls) | Why people chose you, what nearly stopped them | Your time | Ask open questions; don't sell during the call |
| Short surveys (email or text) | Counting how common an answer is | Free tools | Low response rates; keep under 5 questions |
| Google Search Console | What people search before finding your site | Free | Only shows searches you already appear for |
| Google Business Profile performance | Calls, direction requests and searches on Maps | Free | Numbers are approximate |
| Competitor review mining | What customers love and hate in your market | Free | Reviews skew to very happy or very unhappy |
| Keyword research tools | Demand for a service or topic in your area | Free to paid | Volumes are estimates, not exact counts |
| Small test ads | Whether an offer gets interest before you build it | A small ad budget | Needs tracking to read results properly |
| Your own sales records | Which jobs, seasons and customers are most profitable | Free | Often messy; clean before trusting |
A worked example
Here's a fictional example of the process for a small landscaping company:
| Step | What they did |
|---|---|
| 1. Decision | Quote requests are steady, but too many are small one-off cleanups. Should they focus marketing on design-and-install projects? |
| 2. Plan | Check last year's jobs by type and value; interview recent design clients; look at what searches bring visitors. |
| 3. Collect | Exported jobs from their invoicing tool; called 8 design clients; pulled Search Console queries. |
| 4. Analyze | Design projects were a minority of jobs but most of the profit. Clients found them through photos of past projects and a neighbor's recommendation. Few site visitors searched design-related terms because the site barely mentioned design. |
| 5. Act | Built a project gallery page with before-and-after photos, added a design consultation offer, and shifted ad budget toward design searches in spring. |
| 6. Check | Tracked design consultation requests and average job value for the next season. |
No research agency, no big budget: a spreadsheet, eight phone calls and a free Google tool.
Common research mistakes
- Researching without a decision. Interesting facts, no action.
- Only asking happy customers. Lost quotes and one-time customers often tell you more.
- Leading questions. "How much did you love our fast service?" isn't research.
- Trusting one loud opinion. Look for patterns across several people.
- Never repeating it. Markets change; a quick check once or twice a year keeps you current.
If you'd like a hand turning research into a plan, the Marketing plan template is a good next step, or book a free call and we'll look at your data together.