Why a five-location shop needs a different Google Business Profile cadence than a single-location shop

A single-location shop has one GBP listing and one owner who reads the morning digest before the desk opens. The cadence is a thin one: weekly digest, Monday retro, a polarity-aware reply that fires same shift on a 0.3-star drift. It holds because every signal the desk reads is the same signal Google reads — one listing, one address, one review queue, one owner-response thread.

At three or more locations the cadence starts breaking. The desk time is no longer the bottleneck — there is whole-team time on the manager-of-managers handoff, on duplicate-listing cleanup, and on per-location reply-rate decay that an aggregate dashboard never surfaces. Google reads each listing independently; the local three-pack at each address is its own ranking calculation; an aggregate (brand-level) reply rate is not a substitute for a per-listing reply rate. The weekly digest that worked at one location camouflages the drift on the slowest-responding location until that listing has fallen out of its local three-pack.

The fix is structural, not heroic. It is a per-location feed that scores polarity on each listing independently, a per-location reply rate that compounds on each address, escalation rules that route when one location crosses an SLA, and a manager-of-managers handoff that lets the same operator own every listing without inheriting one listing's quiet decay. The four operations below are the operating cadence that holds across N listings — without growing the operations team in proportion to the listing count.

NAP — Name, Address, Phone — is the single most-leveraged GBP signal Google reads at every location. The address has to match the address on the brand site, on the covering-site aggregators, and on every citation across the web, character for character. The abbreviate-vs-spell-out question (Street vs St, Suite vs Ste) is the one that creates the cleanest duplicates — Google reads the slight variation as a separate business and rewards it with its own listing page, its own photo threshold, its own review queue, and its own local three-pack slot.

The duplicate-listing trap is the operational one. A duplicate that has been indexed for two years is, by Google's reckoning, the canonical location — even when the owner's actual listing is the one Google created. The duplicate out-ranks the brand page because it has more reviews, more photos, and a longer history on the address. The owner's actual listing sits at a handicap every other signal has to overcome. The fix is to claim the duplicate first, mark it as a duplicate of the real listing, and ask Google to merge them through the support channel — but the merge itself can take sixty to ninety days, and the mapping cluster around the address can survive.

Most multi-location dashboards report reply rate at the brand level — a single number, averaged across every listing, presented in the Monday retro. The number reads well; it tracks up over time; it makes the multi-location cadence look healthier than it is. Google's indexer does not read it that way. The local three-pack at each address is its own ranking calculation, built on its own listing's reply rate, reply velocity, and review volume over the trailing ninety days. A location with a 12% reply rate competing against three sister listings at 95% does not pull its weight up by averaging; it pulls its own listing's prominence score down on every reply the other three locations earn.

Polarity scoring compounds the gap. A polarity-aware feed on a single listing scores every review on arrival against the themes the listing cares about — wait, billing, response time, follow-up. At N listings, the same feed has to score per listing; the polarity on "response time" at Location A and the polarity on "response time" at Location B are independent signals, and the themes that drift at Location A (a transparent-quoting gap on a $4,800 compressor swap) are not the themes that drift at Location B (a response-time gap on a Sunday no-cool Sunday). The aggregate polarity is a false composite, and the per-location theme scores are the only signal that scales linearly with the listing count.

The single-location cadence reads the polar themes as a single feed — one theme, one threshold, one rule. At N listings the same themes drift independently. Location A drifts on "response time" because its on-call tech is overloaded on weekends; Location B drifts on "transparent quoting" because its dispatcher quotes over the phone without a written line-set; Location C drifts on "follow-up" because its follow-up call never reaches the customer. The themes are not interchangeable, and a centralised feed that fires the same rule across every listing will mis-spend attention on Location B's response-time theme when the actual gap is on Location B's transparent-quoting theme.

The structural answer is one centralised feed with per-location escalation rules — the feed routes the same polarity score to the same operator, but the rule set underneath is per listing. Location A gets the "response-time" rule with the Sunday escalation path; Location B gets the "transparent-quoting" rule with the quote-template rebuild; Location C gets the "follow-up" rule with the call-list reattempt cadence. The operator watches one feed; the per-listing rule sets stay separate; the same alertness that catches a 0.3-star drift at one listing now catches a different drift at the next, with the threshold tuned to the local reality.

The compounding effect is substantial. A centralised feed with per-location rules produces the same same-shift lever at every address — the operator owns the alert regardless of which listing fired it, the four-part reply cadence ships unchanged, and the ranking credit compounds per listing rather than per brand. The cost is a thin rule set per location (one threshold per theme, per listing) and a manager-of-managers escalation that hands a listing off to the local manager when the alarm crosses a threshold the centralised operator cannot move.

The structural question every multi-location owner reaches by the third listing is: who owns each location at 7am, when the reply crosses an SLA, and how the manager-of-managers handoff works? The centralised answer (one operator reads every feed) works for the first thirty minutes of the morning and collapses when three locations cross the SLA at the same hour. The fully distributed answer (each location's manager owns everything) works for the largest operators and is unaffordable for anyone below a hundred listings. The cadence that holds for the middle — three to twenty listings — is a hybrid: one centralised operator owns the feed across all listings during business hours, the per-location manager owns the alert that the centralised operator cannot move, and the escalation chain sleeps during the off hours but wakes on a same-shift trigger.

The Northgate case study ships pre-scoped to the HVAC vertical — four-truck shop across five service locations, per-location reply-rate decay caught on the listing that lags, manager-of-managers escalation rule set tuned to the Sunday no-cool surge. Open the vertical to watch the per-location feed rebuild ranking credit on the slowest-responding address in real time.

When the same per-location cadence is what the listings need on every address, the pricing page lays out what the multi-location feed costs once the per-listing reply rate + per-location escalation rules are wired up — theme scoring per listing, manager-of-managers escalation, and the four-part reply cadence that compounds ranking credit on every address independently. The multi-location tier is the line that wires per-location reply rate + per-location escalation; the single-location tier stops being the right answer at three listings.