Every chapter in this cluster has built toward this one. GBP, citations, reviews, on-page, schema, multi-location architecture — they all feed the candidate set and the ranking signals that decide who gets the three slots in the local 3-pack. This chapter covers the tactical layer: what specifically moves a profile from "in the candidate set" to "at position 1, 2, or 3" for the queries that matter most.

“For a storefront business, the map pack is the conversion ranking. The standard organic ten-blue-links result below it gets clicked at a fraction of the rate. Winning position 3 of the 3-pack outperforms winning position 1 of the organic results for most local-intent searches. The whole local-SEO stack exists to win those three slots.”

What the 3-pack actually shows

A typical map pack result includes:

• A small map at the top with three pinned locations.
• Name, average rating, review count, primary category, address neighborhood, hours-status (open / closed / opens at X).
• Buttons or links: website, directions, click-to-call.
• Sometimes attributes (wheelchair accessible, outdoor seating, online appointments).
• Sometimes a snippet from a recent review.
• A "View all" link that opens the full local finder with 20+ results.

Each of those surface-level elements is also a ranking factor or an engagement-feedback signal. Click-through rate on the rich profile depends on every one of them — review count, primary category match, hours status, photos. The optimization work is partly about ranking and partly about converting the impression once you have it.

The signals that decide position 1, 2, 3

Within the candidate set, three signals decide the order:

• Proximity to the searcher. The single largest factor for "near me" and most generic local queries. Proximity is weighted less for specialty and explicit-location queries.
• Engagement on the profile. Click rate on the listing in the SERP, click-to-call rate, direction-request rate, photo views, post views. A profile that converts at higher rates than its candidate-set peers gets boosted within the pack.
• Composite relevance + prominence score. The relevance signals (primary category, services, on-page alignment, review content) and prominence signals (review count, recency, rating, citation breadth, brand search volume) combine into a quality score that sets the candidate-set ordering before proximity adjusts it.

Tactical moves that lift map pack position

• Photos in the profile. The single most under-invested category. Profiles with 100+ photos see meaningful click-rate lift, which feeds engagement which feeds ranking. Add 2–3 new photos per week, geotagged when possible.
• Review velocity above competitors'. Recency is more weighted than count above 50 reviews. If competitors are pulling 10 reviews/month and you're pulling 3, the gap closes only when you exceed their velocity, not when you match it.
• Owner responses to reviews. 100% response rate within 7 days is the visible signal. Specific, helpful responses lift conversion (CTR + directions + calls) which feeds engagement.
• Posts at the right cadence. Once a week minimum. Posts surface in the right-hand panel and contribute to relevance + freshness signals.
• Q&A seeded by the owner. Pre-populated with the top 10 likely customer questions plus owner-marked answers. Unanswered Q&A is a visible negative; seeded Q&A is a small positive.
• Services and product fields filled completely. Each service line is keyword-matched against query intent. Empty service lists leave relevance on the table.
• Attributes for every applicable filter. Each attribute is a query filter; profiles missing applicable attributes are excluded from filtered queries entirely.
• Citation graph cleaned and consistent. Inconsistent NAP across the citation graph weakens the prominence signal. The chapter on NAP consistency and citations covers the cleanup.
• On-page schema and content matched to GBP. Mismatch erodes confidence in the entity. The chapter on local on-page and schema covers the alignment.
• Local backlinks. Links from locally-rooted sites (city press, chamber, partners, sponsorships) lift prominence specifically for local-pack purposes more than generic links do.

Geo-grid rank tracking

A single-point rank pull from "Boston, MA" doesn't tell you the truth about your map pack position. The truth is that you might rank #1 from one neighborhood and #6 from another — the searcher's location moves the ranking. Geo-grid rank tracking samples your position from a grid of locations across your service area (typically 7×7 or 9×9 for a metro), producing a heat map of where you rank well and where you don't.

The diagnostic value: geo-grid reveals proximity gaps that explain otherwise-mysterious ranking variations. If you rank top 3 from your neighborhood and bottom-of-pack from across the city, the gap is proximity, not signals. If you rank inconsistently across the grid with no clean proximity gradient, the gap is signal volatility (citation drift, review-velocity dips, profile changes) — actionable signal-quality work.

The AI Overview overlap

Google's AI Overviews increasingly surface for local-intent queries, pulling representative businesses, ratings, and review snippets. The candidate pool the AI Overview draws from is larger than the 3-pack but uses similar signals — well-cited businesses with strong review profiles dominate both surfaces. Optimizing for the map pack tends to also optimize for AI Overview citation; the two layers are converging.

The implication: don't treat AI Overview optimization as a separate program. The same on-page work, the same review discipline, the same NAP consistency, the same schema correctness — all of it feeds both layers. The one extra requirement for AI Overview specifically is that the page content has to answer the user's question directly, in the first 200 words, with structured data confirming the entity claims. Stacked together, the local-SEO stack is also the AI-Overview-citation stack.

The 90-day operating rhythm

A map pack program that wins isn't optimized once — it's operated. The 90-day rhythm:

• Weekly: respond to all new reviews, post one update, add 2–3 photos, scan Q&A for new questions.
• Monthly: review GBP insights (search queries, calls, direction requests, photo views), audit attributes for new options, refresh services/products text.
• Quarterly: compare against the top 3 competitors in the pack — review count delta, citation breadth delta, content alignment delta. Run the geo-grid rank to identify proximity blind spots and signal volatility.
• Annually: reassess the primary category against the queries you're now seeing in insights. Categories that made sense 12 months ago may not match the queries that actually drive traffic today.

Closing the cluster

Local SEO in 2026 is a stack: a verified and optimized GBP, a clean citation graph, a disciplined review program, well-researched local keywords, on-page and schema alignment, multi-location architecture when the business has scale, and tactical map-pack work to win the three slots that drive bookings. None of these eight chapters is sufficient on its own; together they describe the operational discipline that decides whether a storefront-based business compounds local visibility or competes against itself in the same SERP every quarter.

Pair this cluster with the On-Page SEO cluster for the underlying page-level discipline that location pages depend on, the Off-Page SEO cluster for the link and mention work that lifts prominence beyond local citations, and the Academy hub for the rest of the disciplines.