Tony Wang13 min readGrocery Is the Most Regional Business in America: Mapping the Chains That Never Went National
We mapped Meijer, Publix, Wegmans, and WinCo state by state. Michigan and Florida each run ~2x their runner-up — real regional grocery monopolies, measured.
Our last piece on this dataset mapped seven chains that are, with one partial exception, present in essentially every state — Starbucks, McDonald's, Dollar General. Grocery is a different animal. Outside a handful of companies that grew by acquisition (Kroger, Albertsons), most regional grocers stayed regional on purpose — the format needs deep local supply chains, and a chain that works in Michigan doesn't automatically work in Florida. That makes grocery the cleanest test case for a question the last piece could only gesture at: when a company genuinely never expanded past its home region, what does that look like in the data, and how do you tell it apart from a company that just undercounts badly?
How concentrated is "regional," exactly? A five-way ranking
Every chain below has a state where it's strongest. The question is how much stronger — is the #1 state double the #2 state, or basically tied with it? We ranked all five data points by that single ratio:
| Chain | HQ state | #1 state (per 100k) | #2 state (per 100k) | #1 ÷ #2 | Share of total in #1 state |
|---|---|---|---|---|---|
| Meijer | Michigan | Michigan — 1.27 | Indiana — 0.62 | 2.05x | 46.9% (129/275) |
| Publix | Florida | Florida — 3.47 | Georgia — 1.78 | 1.95x | 62.4% (814/1,304) |
| Wegmans | New York | New York — 0.25 | Virginia — 0.19 | 1.32x | 43.2% (51/118) |
| Kroger (literal name only) | Ohio | West Virginia — 3.51 | Kentucky — 3.04 | 1.15x | 3.9% (62/1,588) — WV isn't even Kroger's HQ state |
| WinCo Foods | Idaho | Oregon — 0.66 | Idaho (HQ) — 0.64 | 1.03x | 19.4% (28/144) |
Two patterns jump out immediately. Meijer and Publix are real, clean regional monopolies — in both cases the home state isn't just ahead, it's the outright center of gravity for the whole company, and the HQ state matches the per-capita leader exactly. Kroger and WinCo are the opposite case: their per-capita "leaders" are near-ties with a neighbor, and in WinCo's case the actual headquarters state (Idaho) doesn't even win its own ranking. Wegmans sits in between — a real, if modest, home-state lead, but at a density so low everywhere that "monopoly" undersells how thin the chain actually is.
Meijer and Publix: the two real monopolies
Publix has the single densest state/chain pairing in this piece — 3.47 stores per 100,000 Florida residents, nearly double Georgia's 1.78:
| Rank | State | Listings | Per 100k residents |
|---|---|---|---|
| 1 | Florida | 814 | 3.47 |
| 2 | Georgia | 201 | 1.78 |
| 3 | Alabama | 92 | 1.77 |
| 4 | South Carolina | 65 | 1.17 |
| 5 | Tennessee | 44 | 0.60 |
| 6 | North Carolina | 58 | 0.52 |
| 7 | Virginia | 22 | 0.25 |
| 8 | Kentucky | 8 | 0.17 |
That's not a new finding on its own — everyone already knows Publix is a Southern chain, and it's been public knowledge for decades that Florida is its home base. What the data adds is precision: 62% of every Publix store in existence sits inside one state, and its second-best market (Georgia, at 1.78/100k) is barely half as saturated. It's the cleanest possible illustration of what "regional and proud of it" looks like in listings data.
Meijer's story is structurally the same, just a notch less extreme:
Meijer's 2.05x ratio is actually the highest of the five data points — narrowly beating Publix's 1.95x — even though Publix's absolute density (3.47/100k) is nearly triple Meijer's (1.27/100k). Those are two different senses of "regional monopoly," and they don't have to agree: Publix is denser everywhere it operates, while Meijer's lead over its nearest competitor state is proportionally larger. Both are real signals; neither one alone tells the whole story.
The parent-company trap: what a literal "Kroger" search actually misses
Kroger Co. is the third-largest retailer in the US by revenue, and if you search this dataset for the literal word "Kroger," you get a plausible-looking Appalachian/Midwest map: West Virginia leads per capita (3.51/100k), Kentucky is close behind (3.04), then Ohio — Kroger's actual Cincinnati headquarters state — in third at 2.23. That's a clean, sample-verified number (30/30 checked, all genuine "Kroger" or "Kroger Marketplace" stores). It's also a serious undercount, and we can prove it directly rather than just assert it.
Kroger Co. doesn't operate under one name. It owns roughly 20 regional banners — Ralphs in Southern California, Harris Teeter across the Carolinas and Virginia, King Soopers in Colorado, Fred Meyer and QFC in the Pacific Northwest, Smith's, Dillons, Mariano's, and more — each trading under its own name, none of them containing the word "Kroger" anywhere in their listing. A literal-name search structurally cannot see any of them.
| What we searched | Result in California |
|---|---|
| q='Kroger', category=grocery_store | 1 listing statewide |
| q='Ralphs', category=grocery_store | 177 listings, all California, 30/30 sampled clean, all linking to ralphs.com — a real Kroger Co. banner the literal-name search cannot see at all |
We tried to go one step further and quantify Kroger Co.'s true combined footprint by adding up every banner, and hit the same contamination problem our prior piece ran into with hyphenated and common-word chain names: Harris Teeter is unusable as a plain search. "Harris" is a common American surname, and the query pulls in real-estate agents, dentists, doctors, and attorneys named Harris in numbers that swamp the actual grocery stores — the opposite failure mode from a hyphen breaking tokenization, same practical result. We didn't force a number here. The honest version of this section is: the true Kroger Co. footprint is materially larger than 1,588, we can prove that directly with one clean banner, and getting an exact combined total would require verifying all ~20 banners individually — a project for its own piece, not a number we're willing to publish today.
The flip side: regional without one state owning it
Not every non-national grocery chain has a Meijer- or Publix-style center of gravity. WinCo Foods and Wegmans are both genuinely regional — WinCo to the Pacific Northwest and Mountain West, Wegmans to the Northeast and Mid-Atlantic — but neither one is dominated by a single state the way Meijer and Publix are.
WinCo is headquartered in Boise, Idaho, and Idaho doesn't even lead its own per-capita ranking:
28 listings
13 listings
28 listings
4 listings
9 listings
Wegmans is the thinnest chain in this entire piece. Even in New York, its actual home state (founded in Rochester in 1916), it tops out at just 0.25 listings per 100,000 residents — roughly a fifth of Publix's Florida density, and well under half of Meijer's Michigan density:
| Rank | State | Listings | Per 100k residents |
|---|---|---|---|
| 1 | New York (HQ state) | 51 | 0.25 |
| 2 | Virginia | 17 | 0.19 |
| 3 | Maryland | 10 | 0.16 |
| 4 | Pennsylvania | 19 | 0.15 |
| 5 | District of Columbia | 1 | 0.14 |
| 6 | Delaware | 1 | 0.09 |
| 6 | New Jersey | 9 | 0.09 |
| 8 | Massachusetts | 5 | 0.07 |
| 9 | North Carolina | 4 | 0.04 |
| 10 | Connecticut | 1 | 0.03 |
Wegmans isn't thin because it's failing — it's a large-format supermarket chain with famously loyal customers, and its low per-capita numbers partly reflect that each store serves a wide catchment area by design, the same effect we found made Costco's numbers look sparse in our prior piece. The point isn't that Wegmans is a weak chain; it's that "regional" doesn't automatically mean "dense in its region." WinCo and Wegmans show that a chain can stay geographically contained for decades without ever producing the kind of single-state concentration Meijer and Publix show.
How we did this, and the caveats
Every number above comes from a live search/facets call against Crawlora's Google Maps business dataset, 2026-08-26 snapshot, cross-referenced against US Census Bureau Vintage 2025 (July 1, 2025) state population estimates.
| Chain | US listings | Verification method | Result |
|---|---|---|---|
| Publix | 1,304 | q="Publix" + category=supermarket, 30/30 sampled clean | Clean |
| Meijer | 275 | q="Meijer", categories supermarket (264) + grocery_store (11) summed; supermarket 15/15 sampled clean, grocery_store too small a slice to individually sample | Clean (supermarket subset), grocery_store unsampled |
| Wegmans | 118 | q="Wegmans", categories supermarket (22) + grocery_store (96) summed, 25/25 sampled clean across both | Clean, multi-category |
| WinCo Foods | 144 | q="WinCo", categories supermarket (122) + grocery_store (22) summed; 24/25 sampled clean (one grocery_store hit, "WinCo Trails end MARKET place," links to a different site and reads as an independent business using the name, not a verified WinCo Foods location) | ~96% clean |
| Kroger | 1,588 | q="Kroger" + category=grocery_store, 30/30 sampled clean | Clean, but a confirmed undercount of Kroger Co.'s true footprint (see Ralphs proof above) |
| Ralphs (Kroger banner, proof only) | 177 | q="Ralphs" + category=grocery_store, 30/30 sampled clean, all California | Clean — used to demonstrate the Kroger undercount, not counted as its own chain |
| H-E-B | — | q="HEB", all categories: dominated by unrelated atm hits (274 of ~450 total); category=supermarket + state=Texas returned 0 results | Dropped — token too short/noisy to trust |
| Harris Teeter | — | q="Harris Teeter": top categories were real_estate_agent, dentist, doctor, mortgage_lender — "Harris" as a common surname swamps the real grocery listings | Dropped — same contamination class as Chick-fil-A in our prior piece |
Caveats worth stating plainly: Meijer's grocery_store subset (11 of 275 listings, 4%) wasn't individually sampled — it's too small a slice to justify a separate verification pass, but it means the 275 total carries slightly less certainty than Publix's fully-sampled 1,304. WinCo's total includes one likely-independent listing we chose to disclose rather than silently exclude or silently include. The Kroger figure is a known, proven undercount, not a best estimate — we're confident the true Kroger Co. multi-banner footprint is larger, and equally confident we don't have a trustworthy combined number for it yet. And as with every piece on this dataset: it's live-growing, so a re-pull months from now will show different totals.
If you want to run your own chain-by-chain or banner-by-banner cuts on this data, that's exactly what Crawlora's Google Maps business API is for — the same structured-data approach behind our chain-saturation and busiest metro areas studies.
Query 132M+ Google Maps business listings directly
Crawlora's Google Maps dataset turns business listings into structured, queryable JSON — filter and facet by category, country, state, county, city, rating, and website status. 2,000 free credits a month, no card.
Frequently asked questions
Which regional grocery chain is the most concentrated in its home state?
Meijer, by ratio: Michigan runs 1.27 listings per 100,000 residents versus Indiana's 0.62, a 2.05x gap over the runner-up — the highest ratio of any chain checked. Publix is close behind at 1.95x (Florida 3.47 vs. Georgia 1.78), but Publix's absolute density is nearly triple Meijer's, so which chain is 'more concentrated' depends on whether you mean the ratio to the runner-up or the raw density.
Does a plain 'Kroger' search on Google Maps data undercount Kroger Co.'s real footprint?
Yes, provably. Kroger Co. operates under roughly 20 regional banners (Ralphs, Harris Teeter, King Soopers, Fred Meyer, and others), none of which contain the word 'Kroger.' A literal 'Kroger' search finds only 1 grocery_store listing in California, while a search for just 'Ralphs' — a verified Kroger Co. banner — finds 177 clean, sampled-genuine California stores the Kroger search cannot see at all.
Why were H-E-B and Harris Teeter left out of this analysis?
Both failed verification. H-E-B's 3-letter query is too short and noisy — most hits were unrelated ATM listings, and a category-and-state-scoped search returned zero results. Harris Teeter collided with the common surname 'Harris': the top matching categories were real-estate agents, dentists, and attorneys, not grocery stores. Both were dropped rather than published with a misleading count, the same standard applied to Chick-fil-A in our prior chain-saturation piece.
Which grocery chain is the least concentrated in its own home state?
WinCo Foods. Its headquarters state, Idaho, doesn't even lead its own per-capita ranking — Oregon narrowly edges it out, 0.66 listings per 100,000 residents versus Idaho's 0.64. Wegmans has a real, if modest, home-state lead (New York, 0.25 per 100k) but is the thinnest chain in the piece by absolute density everywhere it operates.
How is this different from the earlier chain-saturation study on Starbucks, McDonald's, and Dollar General?
That piece covered national or near-national chains — most are present in most or all 50 states. This one is deliberately scoped to grocery chains that stayed regional by design, which makes for a cleaner test of what a genuine, still-regional company's saturation pattern looks like, and surfaces a new methodology lesson (parent-company multi-banner undercounting) that didn't come up in the earlier piece.