97.9% of franchise locations have claimed their profile; only 35.9% appear in the Google 3-pack

Across roughly 350,000 US locations, 97.9% of profiles are claimed. Claiming is finished as an industry project. Only 35.9% of those locations appear in the Google 3-pack for their own category in their own geography, which means roughly two thirds of a franchise estate is correctly claimed, correctly branded and invisible at the moment of local intent. Facebook coverage sits at 53.4%, so nearly half of locations have no presence on the second-largest local surface, which is also the surface the brand's paid social is already buying against. Reviews get a response 46.9% of the time, and when they do, it takes an average of 4.3 days.

Claiming a profile is 97.9% complete as an industry project and 35.9% effective as a marketing one.

1. 01

   Claimed is a prerequisite, not a position

   The 62-point gap between claiming and ranking is the whole opportunity, and no part of it is closed by claiming more profiles. A programme still reporting claim rate as a headline metric is reporting a solved problem, which is comfortable precisely because it is finished.

2. 02

   Pack ranking is decided per location, so it cannot be moved brand-wide

   Category precision, completeness, proximity, review count, review velocity and review recency are all location-level signals. A brand-level campaign moves none of them. The model works location-by-location and triages, because the location already in the pack and the location three positions outside it need opposite work.

3. 03

   Half the estate is missing from the surface the brand is already buying

   At 53.4% Facebook coverage, a brand running paid social is sending local audiences toward locations that have no local presence to land on. Closing that gap is a data task rather than a content task, which makes it cheap relative to almost everything else on this list.

4. 04

   A 4.3-day average response is outside every window that matters

   It is outside the window in which the reviewer is still paying attention, and it sits awkwardly against the published consumer finding that 74% consider only reviews from the last three months — a corpus read for recency, answered on a lag. Below a 46.9% response rate, more than half the estate reads as unattended to anybody scrolling.

Profiles claimed **97.9%** · 3-pack appearance **35.9%** · Google review response rate 46.9% (avg 4.3 days).

The model scores every location individually rather than reporting an estate average: pack presence for its own core terms, primary and secondary category accuracy, completeness, review count, rating, recency, response rate and response latency, and Facebook coverage. It then triages into two groups that get different work. Locations already in the pack get review velocity and response latency, because that is what holds a position. Locations outside it get category accuracy and completeness first, because those are the gates before anything else registers. Response latency is set as a service level with a named owner per location rather than as a brand-wide target, since a brand-wide target on a location-level behaviour is a number nobody is accountable for. Facebook coverage is closed as a data exercise.

1. 1

   Weeks 1–4

   Location-level scoring

   Every location scored individually on pack presence, category, completeness, reviews and coverage

   Owner: OmniFlow

2. 2

   Weeks 4–5

   Triage

   Estate split into in-pack and out-of-pack, with different work assigned to each

   Owner: OmniFlow

3. 3

   Weeks 5–10

   Category and completeness

   Out-of-pack locations corrected at the gate before anything else is attempted

   Owner: OmniFlow

4. 4

   Month 3 onward

   Review velocity and latency

   Per-location service level with a named owner, reported as latency not as volume

   Owner: OmniFlow + operators

5. 5

   Months 3–4

   Facebook coverage

   Missing locations created and populated as a data exercise

   Owner: OmniFlow

6. 6

   Month 6

   Review

   Pack presence measured against the brand's own baseline, not against 35.9%

   Owner: OmniFlow + brand

The model reports pack presence as a share of locations rather than as an average rank, because average rank across an estate hides the shape of the distribution entirely. It would expect the out-of-pack group to move on category and completeness work and the in-pack group to move on review velocity, and it reports the two groups separately for that reason. Response latency is reported in days beside response rate, since a high response rate delivered slowly and a low one delivered quickly are different problems with different fixes. No estate-wide average is published alongside them, for the same reason.

Modelled. Inputs: SOCi, Local Listings Benchmarks for Franchises 2026 (2,751 brands, ~350,000 US locations); BrightLocal, Local Consumer Review Survey 2026 (1,002 US consumers). Modelled outputs are not a forecast or a guarantee of results. Cited: SOCi *Local Listings Benchmarks for Franchises 2026*.

Every figure below is published research, not a client result. They are the inputs to the arithmetic above, listed so it can be checked rather than taken on trust. The bracketed number points to the full citation at the end of this page.

These are the surfaces this engagement is run and measured from, shown with representative figures built around the benchmarks cited on this page. Every account we run reports into views like these, and you keep ownership of all of them.

Each figure on this page, the system it is read from, and the definition and window it is measured over.

Not applicable — this is a modelled engagement. Its weakest input is measurement method: 3-pack presence is captured against a defined query set at a point in time, and pack composition varies by query wording, device and the searcher's exact position, so a brand's own measured presence will differ from 35.9% in both directions and by a wide margin at some locations. The scoring phase exists to establish the brand's own baseline before any budget is moved on the strength of a published average.

A modelled capability, not a client account. Figures illustrate what the model produces and are labelled as modelled wherever they appear. See all our work

Full citations for everything cited on this page, with the sample and period each study covers, so you can go and read the original.