Local SEO Keyword Research for a Small Business: Choose Searches Worth Tracking

Local SEO keyword research is the work of matching real customer searches to services you actually provide, places you actually serve, and pages or business listings that answer them. Start with customer language, gather a wider list of candidate phrases, inspect the search results in your market, then choose what to answer on your website and what to observe in Maps. A large estimated search volume cannot make an unoffered service a good target.

This guide ends with a one-page worksheet and a worked example. It is a way to make an owner decision, not a promise that choosing a phrase will improve a Google position.

1. Inventory the work before opening a keyword tool

Write three short lists: services sold, areas served, and services not sold. Add variations that matter to customers: urgent versus planned work, residential versus commercial, and in-shop versus at-home. For a business with several locations, make a separate sheet for each one. A phrase that fits one branch may be wrong for another.

Our fictional example is Riverside Appliance Repair. It visits homes in Riverside and Eastbank for household washers and refrigerators. It does not install appliances, repair commercial refrigeration or promise same-day visits. These facts are invented for the worksheet, not observed Mazuly customer data.

Before writing a page or setting up a rank check, identify the service boundary. “Appliance repair” is broad; “commercial fridge repair” is plainly outside this shop's offer. Recording that exclusion early prevents a popular but misleading keyword from becoming an inaccurate service claim.

2. Collect candidates from four useful places

Do not begin by generating hundreds of near-identical city phrases. Gather a manageable set from evidence the business can recognize:

1. Customer enquiries. Note the words callers use for the job, the symptom, and the area. Ask whether “washer not draining” usually turns into a bookable repair or an information request. Do not copy private customer messages into a public article.
2. Existing pages and services. List each real service page and the terms it already uses. A phrase without a matching answer may point to a missing page, but it can also point to a service the business does not offer.
3. Google Search Console, if the owner has a verified website. Review queries and pages with impressions or clicks, including phrases you did not expect. These are searches on which the website appeared, not a full market keyword list and not a Maps rank report. Google's Performance report explains these units and filters.
4. Current search results and optional keyword estimates. Look at the types of pages and local listings that appear for a candidate in the actual market. Notice customer wording on competing results, without copying another company's claims or inventing a service. A keyword tool's volume estimate can help prioritize two equally relevant ideas; missing data does not mean no local demand, and a national estimate does not prove demand in Riverside.

Write variants separately at first. “Fridge repair Riverside,” “fridge repair near me,” and “why is my fridge warm?” may share words but imply different locations and decisions. Group them only after inspecting the results.

3. Observe intent where the customer searches

For each promising phrase, record the date, location and device you used to inspect results. Note whether the results show a local pack or Maps listings, service pages, how-to answers, product retailers, or a mixture. Search results can vary by place and person, so one screenshot is a sample, not a universal ranking.

Do not assume a city name guarantees a Maps result or that “near me” is a separate service. A phone search for “washer repair near me” depends heavily on the searcher's location; the same words from another suburb may show a different set of businesses. An informational query such as “why is my washer leaking?” may call for an explanation and a safe route to a professional, even if some local results appear too. Record the actual result format before deciding where the answer belongs.

Google says local results mainly depend on relevance, distance and popularity. Selecting a keyword does not override those factors or guarantee inclusion in a local pack.

4. Group by customer job, then select and reject

Here is a worked, illustrative first pass for Riverside Appliance Repair. The “result pattern to verify” column is a research task; it is not a claim that we ran these searches or measured their volumes.

The first draft list had ten phrases; the useful output is two service clusters, one information question, and several clear rejections. That is better than ten thin location pages. Another business may choose different clusters after observing its own results and customer questions.

5. Map the chosen cluster to a real answer

For each cluster, decide what the customer should see and how you will assess it:

Search Console measures the website's appearance in Google Search. A local grid samples where a business listing appears for one query from chosen points. A page's average position and a grid square's position are different units; do not combine them into a single score. Google's Search Console report guide describes page/query metrics. The nine-point Maps grid guide explains the local view.

For the fictional shop, the before state might be a generic appliance page that never says “washer,” a Business Profile with incomplete service details, and an internal list of twenty city phrases. The after plan is specific: verify the real service facts, make one washer page and one refrigerator page useful, reject installation/commercial/same-day claims, and save two representative local phrases with their location and date. This is an example of a clearer plan, not a claim of improved ranking or enquiries.

Copy this one-page local keyword worksheet

Use one row per customer job or phrase you are still testing. Copy the headings into a note or spreadsheet; add more rows only when they change a decision.

The example result type in the first row is illustrative worksheet text, not a live SERP observation. Replace it with what the owner actually sees. Do not add a keyword to a page because it is on this sheet; add useful, true information that answers the customer who used it.

Mazuly's Search Visibility samples selected local searches across nine points. The owner chooses a real phrase and area; Mazuly is not claiming to discover every keyword, replace Search Console, or improve Google rank by running a check. Once the sheet identifies a phrase worth watching, use the 30-minute post-check fact audit to decide whether any customer-facing fact needs correction.