Market Saturation Benchmarks: The Hidden Denominators That Decide Whether a Feasibility Study Is Bankable

The Number Is Not the Saturation

Across more than hundreds feasibility studies our team has conducted during our feasibility consulting practice a single methodological pattern explains most of the cases in which a defensible-looking saturation analysis fails its first contact with a sophisticated reader. The pattern is straightforward. A saturation benchmark is a ratio. The numerator is supply. The denominator is some unit of demand. Both halves are choices, and the choice changes the verdict.

The most-quoted statistic in American commercial real estate illustrates the point. The United States has roughly 23.5 square feet of retail space per capita, more than three times the figure for the United Kingdom and ten times the figure for Germany (1). The number is real, the number is sourced, and the number is essentially useless for any specific decision. ICSC's 23.5 figure counts shopping-center gross leasable area divided by total national population. CoStar's broader retail inventory series, which includes freestanding stores, big boxes, drug stores, and standalone restaurants, puts the same country at 54.3 square feet per capita (2). The narrow grocery-only figure runs approximately 4.15 square feet per capita (2). Macy's former chief executive Terry Lundgren has publicly cited 7.3 (3). Each number is correct under its own definition. Each yields a different conclusion about whether American retail is oversupplied.

What is true of retail is true of every asset class commercial banks finance. The denominator is the decision, and the decision is rarely audited.

The two prior MMCG articles — on bankable supply-and-demand mechanics, and on demographic conversion methodology (4, 5) — establish how institutional feasibility reports construct the demand side of the equation. This piece picks up where those leave off. The question is no longer how much demand exists or how it is counted. The question is whether that demand has already been claimed by competing supply, and whether the test used to answer that question is the test a credit committee can defend.

The remainder of this article proceeds in five parts. Part I documents why the most-cited saturation benchmarks fail when audited. Part II isolates three structural questions — universe, denominator cohort, vintage — that determine whether any saturation finding holds. Part III walks through the dispositive institutional saturation tests by asset class, organized around the seven core asset families and the specialty assets where standard ratios reliably break. Part IV identifies two recurring failure modes in the feasibility literature. Part V documents the MMCG approach.

Part I — Why Saturation Is the Most Abused Number in Real Estate

The denominator problem repeats with a kind of dark consistency across commercial real estate.

In self-storage, the national net-rentable-square-feet-per-capita figure ranges from 6.07 — the Self-Storage Almanac saturation reference (6) — to 7.8 (Yardi Matrix) (7), to 9.5 (Storage Authority and Radius+) (8), depending on which vendor's facility universe is being used. The spread is roughly fifty-seven percent. A 3-mile trade area near the threshold can read as undersupplied on Yardi data and oversaturated on StorTrack data, and neither analyst is wrong. Yardi Matrix tracks roughly 32,640 completed facilities; Radius+ tracks approximately 48,000; StorTrack tracks more than 67,400 (9, 10, 11). The same geography pulls a materially different competitive set across vendors, not because the data is incorrect, but because the inclusion rules differ. Yardi skews institutional. StorTrack catches mom-and-pop and vehicle-storage hybrids that Yardi excludes. The result is that the saturation reading is a function of the vendor as much as the market.

In senior housing, the National Investment Center's own methodology paper acknowledges that the same national inventory of approximately 1.59 million units produces an 11.4% penetration rate against households age 75 and over, an 18% rate against households age 80 and over, and a 30% rate against households age 85 and over (12). That is a denominator-driven divergence of more than two-and-a-half times on identical numerator inventory. Institutional feasibility convention compounds the divergence by adding income qualification. Fitch, Ziegler, Dixon Hughes Goodman, and BB&T use age-and-income-qualified households as the denominator, where income is set to approximately 1.5 to 1.7 times the annualized monthly service fee at the project (13). That denominator can be five to ten percent the size of the all-75-plus base, which means the same inventory that reads 11.4% on NIC convention can read 22% or higher on feasibility convention — and a project rejected under one denominator can be approved under the other.

In hotels, the popular benchmark of rooms-per-thousand-residents obscures the only test that matters: STR fair-share, the ratio of a property's room-night capture to its inventory share of the competitive set (14). A demand-generator-driven submarket — a regional medical center, a research university, a port — can support hotel density that reads as abusive on a per-capita basis and remains structurally undersupplied at the property level. The dispositive institutional measure is the Market Penetration Index, with a stabilized reading of 100 to 110 the underwriting target.

In multifamily, units-per-household ratios are a research curiosity. The institutional test is vacancy versus a long-run equilibrium of approximately 5.0% (CBRE) (15) and net absorption versus completions on a trailing-twelve-month basis. Through Q4 2025, the national reading was unambiguous and bifurcated. CoStar reported 8.2% vacancy (16). Moody's reported 6.7% (17). Yardi reported 5.4% (18). CBRE reported 4.4% (15). The spread reflects universe differences — institutional-grade only versus all 5+ unit; stabilized only versus lease-up included; sixty-nine markets versus one hundred forty. All four show the same direction: vacancy above the long-run equilibrium and rent growth flat or negative.

In industrial, there is no institutional per-capita benchmark at all. The asset class is governed entirely by vacancy and the absorption-versus-deliveries flow (19). Equilibrium frictional vacancy sits at approximately five percent. Seven percent is "watch." Eight percent is CoStar's downside oversupply trigger (20). Through Q1 2026, the major brokerages clustered between 6.7% and 7.5% — CBRE at 6.7% (21), Cushman & Wakefield at 7.0% (22), JLL and Newmark at 7.5% (23, 24), Prologis at 7.4% (25).

The pattern is consistent. Lenders, sponsors, and policy advisors cite saturation numbers as if they are facts about markets. They are facts about methodologies. The number is not the saturation. The number is the saturation under a specified denominator, vintage, geography, and vendor universe — and any feasibility analyst who delivers it stripped of that context has handed credit committee a quotation, not an analysis.

Part II — The Three Denominators

Three structural questions determine whether any saturation finding will hold. They precede every asset-class analysis and they answer for the credibility of every conclusion that follows.

The First Question Is Universe

Whose supply is being counted? The vendor's facility universe controls the saturation reading regardless of the apparent precision of the methodology. In self-storage, the Yardi-versus-StorTrack universe gap of roughly two times produces saturation conclusions that contradict each other on the same trade area (9, 10, 11). In industrial real estate, CBRE reported a 6.7% national vacancy in Q1 2026 (21), Cushman & Wakefield reported 7.0% (22), JLL and Newmark reported 7.5% (23, 24), and CommercialEdge reported 9.5% (26). The cluster is wide because CommercialEdge includes flex and older stock that the brokerages' top-fifty-market panels exclude. A feasibility study that quotes "national industrial vacancy" without naming the vendor and the universe is reporting an opinion, not a statistic.

The Second Question Is Denominator Cohort

The cohort against which supply is normalized matters more than any other single methodological choice. Senior housing is the cleanest illustration. NIC publishes 11.4% penetration against the 75-plus household base; the same inventory reads 18% against the 80-plus base and 30% against the 85-plus base (12). The typical resident in senior living is 83 or 84 years old, which is why NIC uses the 80-plus or 85-plus framing in forecasting work even though it publishes the 75-plus figure in headline reports.

Institutional feasibility goes further. The Fitch project penetration ceiling is five percent against age-and-income-qualified households; BB&T's is five to ten percent (13). Market saturation ceilings run fifteen percent (Fitch) to twenty-five percent (BB&T) against the same income-qualified denominator. The result is that the same inventory NIC reports at 11.4% can read 22% or higher under feasibility convention. Both are correct under their definitions. Only one is the underwriting answer.

The same denominator problem appears in retail trade area definition. A trade area can be defined by radial ring (1, 3, or 5 miles), drive-time isochrone (5, 10, or 15 minutes), or mobility-derived true trade area (anonymized GPS pings calibrated to capture seventy to eighty percent of observed visits at the closest competitive center) (27). Each method captures a different population at a different income profile, and each will produce a different gap analysis. Sophisticated feasibility studies report all three and disclose the divergence.

The Third Question Is Vintage

Saturation benchmarks drift. The Yardi Matrix national self-storage per-capita figure was 7.1 in 2022, 7.6 by mid-2025, and 7.8 by December 2025 — the metric itself moved fifty basis points in three years as supply outran demographic growth (7). Any feasibility study citing a 2022 benchmark against a 2025 supply pull will systematically conclude oversupply in markets that have actually rebalanced. The drift applies to retail per capita (which has fallen from twenty-five to twenty-three since 2009 as inventory was demolished faster than population grew) (28), to senior housing inventory growth (now at a record-low 0.4% in Q1 2026, dramatically below historical pace) (29), and to multifamily completions (which peaked in 2024 at a thirty-eight-year high and have already fallen forty percent into 2026) (30).

These three questions — universe, denominator cohort, vintage — should appear on the cover page of every institutional feasibility study before any saturation number is reported. They almost never do.

Part III — Asset-Class Saturation Tests, Stripped to What Matters

The institutional saturation test differs by asset class. There is no universal benchmark. What follows is the working framework MMCG uses across thirty-plus asset classes, organized by the dispositive test for each.

Hotels: STR Fair-Share Is the Only Honest Answer

The popular hotel benchmark is rooms per thousand residents. The institutional benchmark is STR fair-share.

A 100-room hotel inside a 1,000-room competitive set has a ten percent inventory share. A fair share of demand means it captures ten percent of room nights — neither more nor less than its inventory entitles it to. The Market Penetration Index, Average Rate Index, and Revenue Generation Index measure exactly this: occupancy, ADR, and RevPAR relative to the competitive set, indexed to 100 (14). A subject reading 110 is outperforming by ten percent. A subject reading 85 is the problem credit committee will be asked about in month thirteen.

The dispositive saturation test is not whether the market has too many rooms per resident. It is whether the new property can achieve a stabilized MPI of 100 to 110 within twenty-four months against an honest competitive set defined by STR's panel rules — four or more properties, at least three unaffiliated, no single brand exceeding fifty percent and no single company exceeding seventy percent of the panel (31). Layered onto that is the Lodging Econometrics pipeline, which tracks projects in three stages: under construction, scheduled-to-start in twelve months, and early-planning. The Q4 2025 U.S. lodging pipeline stood at 6,146 projects representing 720,089 rooms, of which only 18.7% were under construction (32). That distribution is itself a saturation signal: a market with heavy under-construction share is about to absorb supply; a market with heavy early-planning share has volume that may never break ground.

The 2025 RevPAR decline of 0.3% was the first non-recessionary annual decline ever recorded (33). The 2026 forecast is +0.6% (CoStar) (33). The chain-scale trifurcation — luxury up three percent, economy down 4.4% — explains why the national average obscures more than it reveals (33). A new economy hotel and a new luxury hotel are not the same asset class for saturation purposes, and a feasibility study that treats them as one is producing fiction.

Multifamily: Vacancy Versus Equilibrium, Absorption Versus Completions

The multifamily saturation test in 2026 is brutally simple. The institutional reading combines three things. Stabilized vacancy against a long-run equilibrium of approximately 5.0% (CBRE) (15). Trailing-twelve-month net absorption against completions. Lease-up velocity at directly comparable new properties, measured as units leased per property per month. A submarket where absorption has lagged completions for four-plus consecutive quarters is oversupplied regardless of the units-per-household ratio.

Through Q4 2025, the national reading was unambiguous and bifurcated. The 8.2% CoStar (16), 6.7% Moody's (17), 5.4% Yardi (18), and 4.4% CBRE (15) prints all show vacancy above the long-run equilibrium and rent growth flat or negative.

The geographic bifurcation is the story. Sun Belt markets are digesting the 2024 supply peak: Austin down five percent on rent at 14.2% vacancy (34), Phoenix down four percent at approximately twelve percent (35), Denver down 3.6% (36). Coastal and Midwest gateways are tightening: New York up 5.8% (37), San Francisco up 5.6% (38), Chicago up 4.1% (39), Twin Cities up 3.2% (40). The South Census region rental vacancy reads 9.0% (Q2 2025 Census HVS) (41); the Northeast reads 5.2% (41). That regional gap is what a saturation analysis must capture; the national figure is the average of two opposite stories.

The thirty-percent rent-to-income standard functions as a demand governor, not a saturation threshold. It caps achievable rent at thirty percent of qualifying household income, which sets the income-qualified renter pool used as the denominator in capture-rate analysis (42). When the Census reports median renter rent-to-income at thirty-one percent (43) and Moody's reports national gross RTI at 29.7% (44), demand becomes inelastic to additional supply at top-of-market rents. That inelasticity is why Class A lease-ups in Austin and Phoenix are taking eighteen to twenty-four months and requiring concessions of seven percent or more, while a feasibility study using asking rents in the same market overstates revenue by five to eight percent.

The 2024 supply peak was 608,000 completions per NAHB — the highest since 1986 (30). The 2026 forecast is 333,000 units, the lowest since 2014. Moody's "inventory multiplier" reads 2.8 against a 2.0 oversupply threshold (45). The imbalance is not that supply has been excessive in absolute terms but that demand has decelerated faster, and the resulting overhang will resolve through 2027 as deliveries fall toward 200,000 units per year while household formation stabilizes.

Self-Storage: The Per-Capita Number Almost Nobody Should Be Using

Self-storage is the asset class where per-capita has the strongest hold on industry vocabulary and the weakest hold on institutional underwriting.

The popular national benchmark is seven to 7.8 net rentable square feet per capita (Yardi Matrix) (7). The Self-Storage Almanac references 6.07 NRSF per capita as a saturation threshold (6). Storage Authority and Radius+ reference 9.5 (8). TractIQ — the new official data provider of the 2026 Self-Storage Almanac, replacing Radius+ — references 8.0 (46). The high-low spread is roughly fifty-seven percent on the national figure. That is the widest cross-vendor divergence in any asset class we track.

What is producing the spread is universe. Yardi covers 32,640 facilities (9), Radius+ approximately 48,000 (10), StorTrack 67,400-plus (11). The institutional names are publicly repudiating the metric. DXD Capital states there is "a lack of correlation between higher storage prices and lower square feet per capita" at the macro level (47). Inside Self Storage notes markets at two square feet per capita with collapsing rates and markets at fifteen square feet per capita nowhere near saturation (48). The four-variable CBSA model used in the Self-Storage Almanac — population, percentage of renters, average household size, average household income — runs on Newmark's regression, but Newmark publishes no R-squared (49).

What works instead is rents, occupancy, and pipeline. National same-store occupancy ran 76.9% in Q1 2026 (Storable Industry Pulse, more than 30,000 facilities) (50). REIT-managed portfolios ran ninety-two to ninety-three percent (51). The spread between REIT and independent is itself a saturation signal — institutional revenue management is sustaining occupancy in markets where mom-and-pop facilities are dropping rents to fill space. National advertised rent fell to $16.07 per square foot annualized in March 2026, down two percent year over year (7). The 2026 forecast for new deliveries is 51.1 million net rentable square feet — approximately 2.4% of stock — with starts down twenty-nine percent year over year in Q1 2026 as the construction pipeline burns down (7).

The local adjustment factors are non-optional: climate-controlled penetration, basement geology (the Northeast naturally absorbs storage demand through residential basements that Texas and Florida lack), household mobility, the renter share, the unit-size mix (Manhattan units average thirty-two square feet, Boise units one hundred fifty-one), and proximity to demand generators (52). A 3-mile trade area in a snowbird market with a high renter share and limited basement penetration is genuinely undersupplied at nine square feet per capita. The same trade area in a Northeast suburb with high basement penetration is genuinely oversupplied at five.

Senior Housing: The Denominator Is the Decision

Senior housing is where the denominator argument is most consequential and where the gap between popular metric and underwriting metric is widest.

NIC's published national penetration rate is 11.4% against households age 75 and over (12). NIC's own methodology paper acknowledges the rate is 18% against age 80 and over, and 30% against age 85 and over — on identical numerator inventory (12). Institutional feasibility goes further. Fitch's project penetration ceiling is five percent against age-and-income-qualified households; BB&T's is five to ten percent (13). Market saturation ceilings run fifteen percent (Fitch) to twenty-five percent (BB&T) against the same denominator. The income qualification is set to approximately 1.5 to 1.7 times the annualized monthly service fee at the project (13), which means the denominator is a small fraction of the 75-plus base. The result is that the same inventory NIC reports at 11.4% can read 22% or higher under feasibility convention.

The structural undersupply is verifiable in the operating data. National occupancy reached 89.5% in Q1 2026 — the nineteenth consecutive quarterly gain, per NIC MAP's April 23, 2026 release (53). Independent living crossed ninety-one percent. Q1 2025 construction starts in NIC's primary markets fell to 1,076 units, the lowest since Q2 2009 (54). Units under construction at year-end 2025 stood at roughly 17,000, the lowest since 2012 (29). Q1 2026 year-over-year inventory growth was 0.4%, a record low since NIC began tracking in 2006 (29).

NIC's own projections put the unit shortfall at 549,000 by 2028 and 806,000 by 2030 against current penetration (55). Cushman & Wakefield's H1 2026 Investor Survey calls for 70,000 units of annual supply growth through 2036 against 2025 deliveries of fewer than 6,000 units (56). The first Baby Boomers turned 80 in 2026. The 80-plus population grows from 14.7 million in 2025 to roughly 19 million by 2030 (55).

The institutional site-selection rule that underwrites this — three hundred to five hundred age-and-income-qualified seniors per planned unit within the 3- to 5-mile primary trade area (57) — is the gate. A 100-unit assisted living community needs thirty thousand to fifty thousand qualified seniors in its primary market. Most tertiary markets in America cannot clear that threshold. Most secondary markets can, barely. The strongest primary markets clear it three or four times over and remain underbuilt.

The middle-income gap — Harvard's "Forgotten Middle" framework — is the structural caveat. Approximately 7.8 million middle-income seniors age 75-plus will have less than $60,000 in annual financial resources by 2029, including housing equity (58). That cohort cannot afford private-pay senior living at expected price points. The demographic wave is real. The affordable-to-the-product portion of it is materially smaller.

Industrial: Vacancy and Absorption-Versus-Deliveries, and Nothing Else

Industrial real estate has no institutional per-capita benchmark. The asset class is governed entirely by vacancy and the absorption-versus-deliveries flow (19).

Equilibrium frictional vacancy sits at approximately five percent. Seven percent is watch. Eight percent is CoStar's downside oversupply trigger (20). Through Q1 2026, the major brokerages clustered between 6.7% and 7.5% — CBRE at 6.7% (21), Cushman & Wakefield at 7.0% (22), JLL and Newmark at 7.5% (23, 24), Prologis at 7.4% trending down (25). The cycle inflected in Q4 2025 when net absorption exceeded deliveries for the first time since 2022 — Prologis recorded fifty-nine million square feet of absorption against substantially lower completions (25); Newmark recorded the first quarterly vacancy decline since 2022 (24).

The bifurcation within industrial is essential to any feasibility conclusion. Big-box bulk distribution (500,000-plus square feet) ran 8.7% to 11.5% vacancy through 2025 and remained oversupplied (59). Shallow-bay and small-bay product (under 100,000 square feet, especially under 50,000) ran three to five percent vacancy and commanded rent premiums of twenty to thirty percent per square foot above bulk (60). The two segments are in opposite cycles. A 75,000-square-foot multi-tenant shallow-bay project in suburban Phoenix is a different asset class for saturation purposes than a 750,000-square-foot speculative bulk distribution building in the same metro.

The build-to-suit share of the pipeline is the second under-appreciated saturation lever. Through 2025, build-to-suit projects represented roughly thirty-five percent of square footage under development (Cushman & Wakefield Q1 2025) (61), rising toward sixty-plus percent of Prologis starts in 2025 (62). Pre-committed BTS does not threaten the spec leasing market and should be netted out of forward supply when computing months-of-supply for a submarket. The 354 million square feet of national pipeline at Q1 2026 (63) looks more threatening than it is once BTS is excluded and once the approximately 190 million square feet of pipeline without committed tenants is isolated (64).

The 2024 delivery peak (608 million square feet of NAHB-tracked completions) (30) and the 2025 collapse (down twenty to twenty-eight percent depending on the measure) (65) set the supply backdrop. The 2026 forecast — 180 million square feet of Prologis-tracked completions, the lowest in a decade (25) — is the floor under the inflection. Data center demand pulled twenty-seven percent of new manufacturing leases signed in Q1 2026 (CBRE / Link Logistics) (66). That is itself a new form of demand that no per-capita benchmark would have captured.

Retail: Famous Metric, Bifurcated Market

Retail is the asset class where the public conversation and the underwriting conversation have drifted furthest apart.

The 23.5 square feet per capita figure has anchored the "overstored America" narrative for fifteen years (1). It is also the asset class with the lowest availability rate on record. CBRE reported 4.7% retail availability in Q2 2024, the lowest since the firm began tracking in 2005 (67). Q1 2026 read 4.9% (68). Cushman & Wakefield's parallel shopping-center vacancy series read 5.7% in Q4 2025, well below the 7.4% historical average (69). The market is allegedly the most overstored on earth and is simultaneously running its tightest availability in two decades.

What resolves the paradox is bifurcation. Grocery-anchored retail ran 3.5% vacancy through 2024 (JLL Grocery Report 2025) with rent growth of 3.1% — the strongest of any retail subtype (70). Necessity and service retail held similar tightness. Mid-tier malls, older power centers, and big-box department-store anchors ran ten to fifteen percent vacancy and absorbed the bulk of the roughly 7,300 store closures in 2024 (Coresight Research) and a projected 15,000 in 2025 (71).

CBRE's January 2025 framework concluded the United States is underretailed by 200 million square feet — approximately five percent of total stock — when measured against population growth and the necessary-retail-per-capita figure in fast-growing Sun Belt and high-income suburban markets (72). Austin, Orlando, Nashville, Seattle, and Charlotte ran availability as low as two to three percent (72). The country is overstored at the all-retail-square-feet-per-capita level and undersupplied at the necessary-retail level. The blended per-capita figure is the average of two opposite stories.

The institutional methodology has accordingly moved to Esri Retail MarketPlace gap-and-leakage analysis at the NAICS level, capture-rate stress tests against system-wide sales-per-square-foot benchmarks, and trade-area availability-and-rent overlays (73). The workflow is five steps: define the trade area (rings, drive-times, true trade area); pull NAICS-level retail potential and retail sales from Esri to compute leakage versus surplus; identify categories with leakage factors above twenty as recoverable demand; apply a defensible capture rate (eight to fifteen percent for in-fill anchor, three to eight percent for in-line shops) and cross-check the implied sales-per-square-foot against industry benchmarks; stress-test capture — if a two-hundred-basis-point cut breaks debt service, the deal is too thin. Per capita does not appear in this workflow.

Specialty Assets: Where the Per-Capita Metric Reliably Fails

Five specialty asset classes — express car wash, fuel-and-convenience, RV park and storage, wedding venues, extended-stay hotels — are united by one trait: population-per-rooftop ratios fail systematically. Each requires a different test.

Express car wash saturation is rooftops-in-trade-area plus AADT plus the membership curve. The defensible methodology layers four tests. A demand equation built from households times vehicles times annual washes times capture rate, with capture rates of one-half to one-and-a-half percent treated as defensible and three to five percent treated as aspirational (74). An AADT cross-check at three to five percent catch rate (74). A competitor map showing all conveyor and in-bay automatic sites within three miles. And a subscription-club ramp curve modeling the eighteen-to-twenty-four-month progression to fifty to seventy percent revenue share. The 2025 supply signal is submarket-level overbuild, not national saturation. ZIPS Car Wash filed Chapter 11 in February 2025 with $654 million in debt against $1 million in cash (75). Mister Car Wash, the only public pure-play, held membership revenue at seventy-six percent of wash sales by Q2 2025 (76) — a measure of the segment's reliance on subscriptions, not transactional throughput. PE-backed platforms still control under twenty-five percent of the express market (74), which means consolidation has further to run even as new development tightens.

Fuel station feasibility is AADT times capture times gallons-per-month, full stop. A site at 4,300 AADT on Interstate 8 in Salton City can move the same gallons as a site at 60,000 AADT in a built-out Dallas suburb, because trucking-route capture and local moratoria dominate (77). The U.S. now operates 151,975 convenience stores per NACS, of which 122,620 sell fuel — the second consecutive annual decline in total store count but the highest fuel-selling count in eight years (78). The national-average station moves 4,000 gallons per day (79). The practical floor for new SBA-financed sites is 100,000 to 150,000 gallons per month (79). EV transition pressures gallons-per-site by one to three percent per year in metro markets; BCG's scenario analysis identifies up to twenty-five percent of sites in EV-dominant markets as potentially unprofitable by 2035 (80). Shell's announced 500-site-per-year global divestiture pace (81) is the most concrete corporate signal of managed attrition rather than collapse.

RV park, campground, and RV/boat storage are demand-generator-driven, not per-capita. The relevant figures: 8.1 million RV-owning households in 2025 (Go RVing 2025, down from the 11.2-million pandemic peak) (82); 11.6 million registered boats (83); private campground occupancy around seventy-three percent (84); Yardi Matrix-tracked RV/boat storage facility count up from approximately 800 in 2023 to 1,798 in 2025 (85) against a structurally undersupplied installed base. The feasibility test is county-level RV and boat registrations, proximity to lakes and parks and snowbird routes, HOA restrictiveness, and competitor occupancy — not population per facility. Median RV usage is twenty-five days per year; median boat usage is fifty-four (86). The asset is storage, not utilization.

Wedding venues are tested by addressable weddings per year divided by competitive venue count. The United States has roughly 35,829 wedding venues against approximately two million weddings per year (87, 88). That arithmetic yields fifty-six weddings per venue as the national average — squarely inside the practitioner benchmark of forty to sixty events for feasibility break-even and the operational ceiling of fifty to one hundred (89, 90). The 2024 "engagement gap" — driven by reduced pandemic-era dating downstream from Signet Jewelers' tracked engagements — produced reported booking declines of forty to fifty percent at many venues before the rebound through 2025 (91). Average venue spend reads $12,900 (The Knot 2026 study) (88) or $8,573 (Zola 2026 index) (92), a fifty-one percent divergence that traces to bundled-versus-space-only pricing structures and must be disclosed in any feasibility report.

Extended-stay hotels are STR fair-share with length-of-stay segmentation, not population per room. The Kalibri Labs framework segments demand by 0-6, 7-14, 15-29, and 30-plus night length-of-stay buckets across 334 U.S. markets and 975 submarkets (93). The 2025 supply signal: extended-stay constitutes forty percent of all U.S. hotel projects in the Q3 2025 pipeline — 2,468 projects, 250,754 rooms (Lodging Econometrics) (94). Despite that, segment occupancy ran 12.4 percentage points above the broader industry through 2025 with demand up 2.2% year-over-year against a 0.8% decline for all hotels (95). The saturation test is the named demand-generator list: corporate transient employers, traveling-nurse contracts, insurance and displacement housing, infrastructure-project demand. A new extended-stay property with a thin or speculative demand-generator stack is oversupplied in any market regardless of population per room.

Part IV — Two Recurring Failure Modes

Across the more than eight hundred feasibility studies our team has reviewed since 2017, two methodological failures recur with such consistency they deserve their own treatment.

The first is averaging across denominators. A study reports "national vacancy" without naming the vendor and the universe. A study reports "penetration rate" without naming the cohort. A study reports "square feet per capita" without distinguishing all-retail from shopping-center-only. The numbers travel through credit committee as if they are interchangeable. They are not. The CBRE-versus-CommercialEdge industrial vacancy spread is forty percent (21, 26). The Yardi-versus-Storage-Authority self-storage per-capita spread is fifty-seven percent (7, 8). The NIC-versus-feasibility-convention senior housing penetration spread is two-and-a-half times (12, 13). Averaging across these denominators is not a conservative approach. It is an unforced error that produces a conclusion neither denominator would support.

The second is vintage drift. A 2022 benchmark cited against 2025 supply data produces a saturation conclusion that neither the 2022 analyst nor the 2025 market would endorse. The benchmarks themselves are moving. Self-storage NRSF per capita drifted from 7.1 to 7.8 in three years (7). Retail per capita drifted from twenty-five to twenty-three across the past fifteen years (28). Industrial vacancy moved from approximately 4.0% in 2022 to 7.5% in early 2026 as the supply cycle resolved (96). Multifamily vacancy moved similarly. Senior housing inventory growth fell to a record-low 0.4% (29). A feasibility study that cites the prior cycle's benchmarks against the current cycle's supply is producing an artifact, not an analysis.

The remedy for both is procedural. Every saturation number in a defensible feasibility study should appear with four pieces of attached context: the source, the universe (what is in the numerator), the denominator (the demand reference cohort and trade-area definition), and the vintage. When all four are present, the reader can reconstruct whether the conclusion follows. When any one is missing, the number is a quotation.

Part V — What MMCG Does Differently

The MMCG approach to saturation analysis differs from dominant feasibility-industry practice in five specific ways.

We disclose the denominator at first reference and at every subsequent reference where it could be misread. Every penetration rate in a senior housing study carries an explicit "x% against 75-plus HH" or "x% against age-and-income-qualified HH" tag. Every retail per-capita figure carries the universe (ICSC shopping center GLA, CoStar broader retail, grocery-only). Every industrial vacancy figure carries the vendor and the universe. The disclosure is not a footnote; it is part of the sentence.

We triangulate across vendors whenever the conclusion sits within reasonable distance of the threshold. A senior housing project with 8.5% penetration on age-and-income-qualified households is too close to the Fitch five percent ceiling to publish a single number. We run it against three denominators (NIC convention, Fitch convention, BB&T convention) and present the range. The reader can decide which ceiling matters for their lender.

We segment within asset class. A retail feasibility study that conflates grocery-anchored with mid-tier mall is producing fiction. An industrial study that does not distinguish big-box from shallow-bay is producing fiction. A hotel study that averages economy and luxury chain scales — when 2025 RevPAR moved minus 4.4% for economy and plus three percent for luxury (33) — is producing fiction. Segmentation is the discipline that prevents averaging across opposite stories.

We use the institutional test as the primary saturation conclusion, with the popular metric as illustrative context. STR fair-share for hotels, not rooms per resident. Vacancy plus absorption-versus-deliveries for industrial and multifamily, not per-capita. Esri Retail MarketPlace gap analysis for retail, not square feet per capita. NIC MAP plus income-qualified depth for senior housing, not headline penetration. Rooftops-and-AADT for car wash, not population per tunnel.

We flag the thirty-percent-plus divergences explicitly. They appear in every asset class. Naming them is the discipline; pretending the numbers agree is the failure.

These five practices add modest cost to the production of a feasibility study and substantial value to the underwriting committee that receives it. The cost is the analyst's time to triangulate. The value is that the conclusion holds six months later, which is precisely when the questions arrive.

Closing: Saturation Is a Conclusion, Not a Number

The thread that runs through every section of this article is the same. Saturation is not a number. Saturation is a conclusion produced by a number, a denominator, a vintage, and a vendor universe. Two of the four are usually missing in the studies that arrive at credit committee. The conclusion that emerges in those cases is not wrong because the analyst made an error. The conclusion is wrong because the conclusion is unconditional — and saturation is the most conditional answer in commercial real estate.

The earlier MMCG references on bankable supply-and-demand mechanics and on demographic conversion methodology (4, 5) lay out how to count demand. This piece is the corollary: how to test whether that demand has already been claimed. The combination — counting demand honestly, then testing saturation honestly — is what produces a feasibility study that withstands the questions a lender will ask in month thirteen, when the construction loan converts and the operating numbers start arriving.

For sponsors evaluating new projects, for lenders sizing exposure, for analysts producing the studies on which both depend, the discipline is the same. Disclose the denominator. Triangulate across vendors. Segment within asset class. Test against the institutional benchmark, not the popular metric. Flag the divergences. The bankable conclusion follows from the bankable methodology — and the methodology either survives a sophisticated reader's questions, or it does not.

The denominator discipline described above is not a theoretical exercise. It is the standard MMCG Invest applies on every engagement, and the difference is most visible in the two loan programs where saturation analysis most often determines the outcome. For SBA 7(a) and 504 transactions, where the feasibility study must satisfy both the lender and SBA SOP 50 10 8, the saturation conclusion has to survive a credit reviewer who is looking for exactly the denominator substitution problems this article describes. Our SBA feasibility study practice is built around that scrutiny, with asset-class-specific saturation tests prepared for the way SBA lenders actually read them.

The same discipline governs rural transactions. USDA Business and Industry, REAP, and Community Facilities loans are evaluated under 7 CFR Part 5001, and the demand and saturation questions in rural trade areas are harder precisely because the population denominators are thinner and the vendor data is sparser. A per-capita benchmark that misleads in a metro becomes actively dangerous in a county. Our USDA feasibility study practice addresses that gap directly, pairing trade-area demand modeling with the generator-driven analysis that rural saturation actually requires. In both programs, the goal is the same: a saturation conclusion a lender can defend, not a statistic that happens to be available.

May 28, 2026, by Michal Mohelsky, J.D. Principal of MMCG Invest, LLC, feasibility study company.

Reach out to discuss how our methodology supports your lending or development decision.

Michal Mohelsky, J.D. | Principal | mmcginvest.com 

Contact: michal@mmcginvest.com

Phone: (628) 225-1125

Disclaimer: This report is provided for informational purposes only and does not constitute investment advice. Data presented herein is derived from proprietary MMCG databases and third-party sources believed to be reliable; however, MMCG Invest makes no representation as to the accuracy or completeness of such information. Figures from third-party industry databases have been independently verified and, where appropriate, adjusted to reflect MMCG's proprietary analytical methodology. Past performance is not indicative of future results.

Sources

(1) ICSC, Shopping Centers: America's First and Foremost Marketplace; AFIRE Summit Journal (2024)

(2) CoStar / Brandon Svec, September 2023 commentary; CoStar food-retail series via Wall Street Journal

(3) Terry Lundgren, CNBC interview, August 2016

(4) MMCG Invest, "Supply and Demand in Bankable Feasibility Studies: How the Gap Gets Measured" (2025)

(5) MMCG Invest, "Demographic Analysis in Feasibility Studies" (2025)

(6) Self-Storage Almanac, 2023 edition (Radius+ data partner through 2025; TractIQ data partner from 2026)

(7) Yardi Matrix self-storage data: monthly national reports 2022–2026; March 2026 supply forecast

(8) Garrett Byrd, Storage Authority Franchise / Radius+ commentary (2024–2025)

(9) Yardi Matrix Self-Storage National Report, March 2026

(10) Radius+ / Union Realtime product documentation

(11) StorTrack via Storable resource pages; Inside Self Storage (2025)

(12) Beth Burnham Mace and Anne Standish, National Investment Center, Looking into the Future: How Much Seniors Housing Will Be Needed? (white paper)

(13) Dixon Hughes Goodman methodology paper hosted by LeadingAge Illinois (2011); Fitch Ratings underwriting criteria; BB&T senior housing lending standards

(14) STR Global Methodology Documentation; CoStar Group (acquired STR 2019)

(15) CBRE Research, "Looming Multifamily Oversupply Expected To Be Short-Lived"; Q3 2024 Multifamily Outlook

(16) CoStar Q4 2025 multifamily release

(17) Moody's Analytics CRE, Q4 2025 multifamily release

(18) Yardi Matrix National Multifamily Report, December 2025

(19) NAIOP / CoStar Industrial Space Demand Forecast (Hany Guirguis, Manhattan University; Joshua Harris, Fordham), Q1 2026

(20) CoStar industrial commentary, Juan Arias, February 5, 2026

(21) CBRE U.S. Industrial MarketView, Q1 2026

(22) Cushman & Wakefield U.S. Industrial MarketBeat, Q1 2026

(23) JLL U.S. Industrial Market Statistics, Q1 2026

(24) Newmark 1Q26 U.S. Industrial Market Conditions and Trends

(25) Prologis Insights & Research, Q4 2025 and Q1 2026 reports; FreightWaves coverage

(26) CommercialEdge National Industrial Report (Yardi-published), September 2025

(27) SiteSeer Technologies, "How to Define the Right Trade Area" (2024)

(28) GlobeSt, "Retail Space Per Capita Hits a Multi-Decade Low" (2023); AFIRE Summit Journal (2024)

(29) NIC MAP Vision, "Senior Housing: Five Key Trends to Watch in 2026" (December 2025)

(30) NAHB, "Multifamily Completions Climb Near a 40-Year High" (July 2025)

(31) STR competitive set panel rules

(32) Lodging Econometrics, Q4 2025 U.S. Construction Pipeline Trend Report

(33) CoStar Hospitality / STR, U.S. hotel performance commentary 2025–2026

(34) Yardi Matrix Austin Multifamily Report, March 2026

(35) CoStar Phoenix multifamily release, Q4 2025

(36) Northmarq, "Denver's multifamily development pipeline falls to five-year low"

(37) Yardi Matrix New York Multifamily Report, December 2025

(38) CoStar San Francisco multifamily release, November 2025

(39) Yardi Matrix Chicago Multifamily Report, December 2025

(40) Yardi Matrix Twin Cities Multifamily Report, December 2025

(41) U.S. Census Bureau, Housing Vacancy Survey, Q2 2025

(42) National Council of Housing Market Analysts (NCHMA), Model Content Standards v3.1, September 2025

(43) U.S. Census Bureau, "Nearly Half of Renter Households Are Cost-Burdened," September 12, 2024

(44) Moody's Analytics, Rent-to-Income Methodology Release, May 2023; Q4 2022 RTI commentary

(45) Moody's Analytics inventory multiplier; Scotsman Guide, Q4 2025

(46) TractIQ, "Some of the Most Densely Populated States Have the Lowest Per-Capita Self-Storage Square Footage"

(47) DXD Capital, "The Most Overused and Unreliable Self Storage Underwriting Metric"

(48) Kris Bennett, Inside Self Storage (2023–2025)

(49) R. Christian Sonne, Newmark Valuation & Advisory, 2025 Self-Storage Almanac (p. 143)

(50) Storable Industry Pulse, Q1 2026

(51) Public Storage, Extra Space, CubeSmart, SmartStop earnings releases Q4 2025–Q1 2026; SkyView Advisors Q4 2025 Self-Storage Industry Report (April 16, 2026)

(52) Adam Barker, Union Realtime, "Is Basement Concentration the overlooked metric in self storage analysis?"

(53) NIC MAP Vision press release, April 23, 2026

(54) NIC MAP Vision, Q1 2025 construction starts data

(55) NIC MAP, "The Impending Age Wave"; demographic projections

(56) Cushman & Wakefield, H1 2026 U.S. Senior Living & Care Investor Survey and Trends Report, February 6, 2026 (Zach Bowyer)

(57) build.inc, "Senior Housing Development in 2026: Site Criteria, Care Levels, and Where AI Fits"

(58) Health Affairs / NORC at the University of Chicago, "The Forgotten Middle," April 24, 2019 (NIC-funded)

(59) Cushman & Wakefield Q1 2026 (big-box vacancy); Marcus & Millichap 2026 Industrial Outlook

(60) BKM Capital Partners Q3 2025 Light Industrial Market Update; CBRE Shallow-Bay Industrial Brief

(61) Cushman & Wakefield Q1 2025 industrial pipeline composition

(62) Prologis Q4 2025 earnings; Markets Daily coverage

(63) CommercialSearch / CommercialEdge National Industrial Construction Report 2026

(64) CoStar via Bisnow, February 2026

(65) Cushman & Wakefield, January 12, 2026 press release on 2025 industrial deliveries

(66) CBRE / Link Logistics, "How Data Center Construction Is Driving Warehouse Demand Across the U.S.," April 2026

(67) CBRE Q2 2024 U.S. Retail Figures (record-low availability since 2005)

(68) CBRE Q1 2026 U.S. Retail Figures; Chain Store Age, May 4, 2026

(69) Cushman & Wakefield Q4 2025 U.S. Shopping Center MarketBeat

(70) JLL Grocery Report 2025

(71) Coresight Research, Deborah Weinswig, BusinessWire, January 23, 2025

(72) CBRE, Five Forces Shaping the Future of Retail, January 2025 (James Breeze)

(73) Esri Retail MarketPlace documentation; ICSC trade area methodology; Robert J. Gibbs, Principles of Urban Retail Planning and Development

(74) MMCG Invest, "Demand Analysis in Feasibility Studies: How AADT, Capture, and Mobility Data Drive the Forecast"; Innowave Studio express car wash design analysis; Car Wash Advisory commentary 2024–2025

(75) ZIPS Car Wash Chapter 11 filing, February 2025

(76) Mister Car Wash Q2 2025 earnings release; Quiver Quantitative coverage

(77) Ticon / C-Site, "From Traffic Counts to Bankable Sales Projections: Why Feasibility Studies Need More Than AADT"

(78) NACS / NIQ TDLinx U.S. Convenience Store Count, January 2026; Mass Market Retailers coverage

(79) Toast POS, "How Much Do Gas Stations Make?"; KMF Business Advisors

(80) BCG, The EV Opportunity for Fuel Retailers, July 2024

(81) Shell Energy Transition Strategy 2024, p. 28

(82) RVIA / Go RVing 2025 RV Owner Demographic Profile (Ipsos)

(83) National Marine Manufacturers Association (NMMA), 2023 registered recreational boat count

(84) Recreation USA campground occupancy benchmarks; MMCG synthesis

(85) Yardi Matrix Q2 2025 National RV & Boat Storage Report; Toy Storage Nation, May 9, 2025

(86) Matthews, "Exploring & Storing: A Look Into Boat and RV Storage"

(87) SmartScrapers / RenTech Digital, U.S. wedding venue count, October 15, 2025

(88) The Knot Worldwide, 2026 Real Weddings Study, February 18, 2026

(89) Metal-Buildings.org, Event Venue Construction Cost: Full Breakdown 2026

(90) Perfect Venue, "How Much Do Event Venues Make in 2026?"; StartupFinancialProjection KPI guide

(91) Shane McMurray, The Wedding Report, CNN Business, March 17, 2024; Signet Jewelers engagement data

(92) Zola, "Average Cost of Weddings in 2026"

(93) Kalibri Labs extended-stay methodology

(94) Lodging Econometrics, Q3 2025 U.S. Hotel Construction Pipeline Trend Report

(95) Highland Group, U.S. extended-stay hotel performance commentary 2025; Hotel Management coverage

(96) CBRE Econometric Advisors and CoStar industrial vacancy series, 2022–2026