Local keyword research is its own discipline because local SERPs are their own animal. The same word with two different modifiers — "plumber Boston" vs "plumber near me" — produces different SERP layouts, different competitor sets, and different click distributions. Researching local keywords using national-keyword methods leaves enormous coverage gaps and wastes effort optimizing for queries that don't carry local intent. This chapter covers the four query patterns, how to find them, and how to map them to pages without overlap.

“Local SEO doesn't compete on one keyword list — it competes on four. Each pattern (geo-modified, near-me, implicit-local, service-area) has its own SERP, its own click logic, and its own ideal landing page. A keyword strategy that lumps them together rankles for some, optimizes for none, and quietly leaks every quarter.”

The four query patterns

• Geo-modified. The keyword carries an explicit location modifier — "plumber Boston", "best ramen Austin", "dentist 02115". The user has typed the location, so Google treats the query as definitely local. Map pack appears; organic results are filtered to local pages.
• Near-me. The keyword carries an explicit "near me", "nearby", or equivalent — "plumber near me", "coffee near me". Google substitutes the user's current location. Map pack appears prominently. The candidate set is geographically tight.
• Implicit-local. The keyword has local intent without saying so — "emergency plumber", "open now", "walk-in dentist". Google infers local intent from the query category and substitutes the user's location. The map pack appears; organic results are filtered.
• Service-area. The keyword combines a service with a service area, often a neighborhood or sub-region — "plumber Brooklyn Heights", "dentist Back Bay". Useful for service-area businesses and for businesses targeting specific neighborhoods of a larger metro.

The intent map per pattern

Each pattern correlates with a specific stage of intent:

• Geo-modified queries tend to be research-heavy. The user is comparing options, often using maps to evaluate distance and reviews. Best landing pages: location-specific service pages or category landing pages with clear NAP, services, hours, and reviews surfaced.
• Near-me queries tend to be conversion-imminent. The user wants to act now. Best landing pages: the GBP listing itself first, then a service page if the user clicks through. The map pack often gets the click directly.
• Implicit-local queries tend to be urgency-driven. "Emergency plumber" implies a problem already happening. Best landing pages: a prominent "open 24/7" or "emergency service" headline, click-to-call above the fold.
• Service-area queries tend to be granular and high-conversion when matched correctly. Best landing pages: neighborhood-specific service pages — but only when the content is substantive, not templated.

Finding the keywords

The mechanical workflow:

• Seed list from the GBP categories. The primary and secondary categories already define the head terms. Each category translates to 2–4 keyword stems.
• Geo-modify the seeds. Combine each stem with the city, the metro, the neighborhood, and the ZIP. A seed of "pediatric dentist" with three location modifiers (city, neighborhood, ZIP) produces three geo-modified queries per stem.
• Add near-me variants. Append "near me" to each stem. Volume here is large in keyword tools but heavily skewed toward mobile, so the conversion lift can be uneven.
• Pull implicit-local from People Also Ask. Search the head term and scan the People Also Ask box. Many implicit-local queries surface there ("emergency", "walk-in", "open now", "same-day").
• Pull service-area from neighborhood maps. List the named neighborhoods and sub-regions of the metro. Combine with the service stem.
• Verify volume. Run the candidate list through Google Keyword Planner with the geographic filter set to the target metro. Tools that don't support geo-filtering will under- or over-count systematically.
• Sanity-check intent. Search each high-volume candidate in incognito with the location set. If the SERP doesn't show a map pack, the keyword isn't actually local — drop it from the local-page strategy.

Volume estimation in local SEO

Three things to know about local volume:

• Tools systematically under-count near-me variants. Google Keyword Planner reports near-me volume nationally, not locally. A "plumber near me" national volume of 50,000 doesn't tell you how many of those happen in your city — assume a fraction proportional to metro population, then validate against actual map-pack clicks via GBP insights.
• Geographic explosion compounds. A national keyword with 1,200 monthly volume often translates to thousands of cumulative city-by-city instances. The micro-queries are too small to register in tools but they add up.
• Seasonality matters more locally. "AC repair" national volume looks moderate; July volume in Phoenix is 8x the annual average. Tools that report annual averages mask the local seasonal spikes that actually drive booking patterns.

Mapping queries to pages

Each query pattern has an ideal page type. The mapping prevents overlap and ensures every page has a single primary target:

• Geo-modified head queries ("dentist Boston") → the primary city service page or category landing page. One page per major city you serve.
• Near-me head queries ("dentist near me") → primarily the GBP listing; secondarily the homepage if it has strong local NAP.
• Implicit-local urgent queries ("emergency dentist") → a service page focused on the urgent variant, with click-to-call surfaced and 24/7 hours stated when applicable.
• Service-area neighborhood queries ("dentist Back Bay") → neighborhood pages, but only when the page has substantive unique content (parking notes, neighborhood-specific information, real photos). Templated content here is the thin-content trap.
• Long-tail FAQ queries ("dentist who takes Aetna in Boston") → FAQ content blocks, schema-marked, on the relevant service page. These earn AI Overview citations more than direct traffic.

The competitive overlay

Volume isn't enough to prioritize; competitive density at the local level decides where the marginal effort goes. The check:

• Map-pack saturation. Look at the top 3 of the map pack for each candidate query. If the top 3 are all 1,000+ review counts, that query is locked unless your prominence catches up over years. Pick a more granular variant.
• Domain strength of the organic competitors. If the organic top 3 are aggregators (Yelp, Healthgrades, local newspapers), entering organically is hard but the map pack might be open. Focus there.
• SERP-feature load. If the SERP has a featured snippet, AI Overview, and a map pack already, the organic clicks that remain are scarce. Optimize for the snippet or AI Overview rather than the organic position.

Multi-location keyword strategy

For multi-location businesses, the rule is one keyword set per location, scoped to that location's geography. A 12-location chain has 12 keyword sets — overlapping where the metros overlap, distinct where they don't. The temptation to share content across locations creates duplicate-content suppression and undermines the prominence signal of each location individually. The chapter on multi-location local SEO covers the architecture.

With the keyword set defined, the next chapter, reviews and reputation management, covers the prominence signal that turns a relevant profile into a top-ranked one.