Gatlinburg short-term rental investment report

Market: Gatlinburg and Sevier County, Tennessee
Subject property: 832 Resort Way, Gatlinburg TN 37738, MLS 312395, asking $797,000
Trailing twelve months (TTM): September 2025 to August 2026
Date: 7 September 2026
Prepared by: Alaa Gaafar

Sample report, built entirely from public data, not commissioned by any party. The subject property is a publicly listed for-sale property used as a worked example.

1 Recommendation

2 The market at a glance

Sources: three free vendor market pages, captured 2026-09-07. S18 covers the trailing twelve months to August 2026; S21 covers August 2025 to July 2026; S22 is a rolling twelve months to 27 April 2026. The comp set median is derived here, method in section 11.

The free market data for Gatlinburg is not reliable enough to underwrite against. Three vendors put active supply 3.2x apart. Two put occupancy at 57.0% and 41.1%. One reports supply growth of +0.8% and another +43.1% for the same market. A third’s headline occupancy of 29% cannot be reconciled with its own revenue and ADR, which imply 58.1%. This table is the sanity check, not the answer: section 5 is built from eight individually read listings.

3 Seasonality

Sources: revenue index from S22 (Gatlinburg mean revenue per listing, rolling twelve months to 27 April 2026); park visits index from S1 (NPS recreation visits, CY2025). Both indexed to their own twelve month average. Occupancy and ADR are modelled: the twelve months distribute the base case rather than regenerating it, so they total $70,450. The twelve printed monthly figures are each rounded to the dollar and sum to $70,451.

Every peak and trough has a named, dated cause. January and February are the trough because the town closes: Anakeesta, Gatlinburg’s main paid attraction and seven minutes from the property, shut every day from 5 January to 8 March 2026, and Dollywood scheduled no festival between 4 January and 12 March. July peaks on school holidays and Dollywood’s summer season to 2 August. October is the second peak, on foliage: the Park Service states it is the busiest month in the park.

December is where a naive model breaks. Park visitation indexes at 74.4, well below average, while Pigeon Forge’s realised lodging tax receipts index at 106.2, above it. December demand here is not park demand: it is Winterfest’s six million lights to 31 January and Dollywood’s Smoky Mountain Christmas from 6 November. A model built on park visitation alone would price December 31.8 index points too low. Four months carry 46.6% of the year and the four weakest carry 20.8%. January and February together project $5,715 of gross, so a buyer needs working capital to sit through two months at 54.7% occupancy and a $177 average rate.

4 The comp set

The selection rule, stated exactly as applied:

Comps are active whole-home cabin listings on Airbnb within 2 miles of the subject property, within one bedroom of its 4 bedrooms, carrying at least 10 reviews and a review inside the last 90 days, so that dormant listings do not pollute the set. Sampled on 7 September 2026.

Sources: S41 (listings read individually on 2026-09-07), S42 (derived statistics). Occupancy A is blocked nights over an 85 night forward window; Adjusted applies a stated 0.90 owner-block factor. ADR is host-side, net of the Airbnb guest service fee and the 12.75% lodging tax, and excludes the cleaning fee.

Mean estimated revenue is $75,350 and the median is $70,450. The median is used, because the mean is pulled up by two comps above $126,000. Revenue runs from $32,100 to $126,700, a factor of 3.9 across eight cabins inside a two mile radius. In this market, execution beats the asset.

No comp is a statistical outlier and none was excluded. The 1.5 interquartile range test was run on ADR, occupancy and revenue. On revenue, Q1 is $54,525 and Q3 is $89,950, giving fences at $1,388 and $143,088, and every comp sits inside. One candidate was excluded before the set was formed, a five bedroom cabin with seven reviews, below the ten review minimum. AB07 is flagged but kept: it prices January nights at 2.3 times its November midweek rate, which is backwards here and unexplained, but excluding a comp for pricing unconventionally would bias the set toward hosts who price conventionally.

Eight comps is thinner than intended. The fences on a set this size are wide, so the outlier test is weak evidence and section 7 is directional at best.

5 Subject property

832 Resort Way is a 4 bedroom, 6 bathroom log cabin of 2,496 sq ft, built 2008, in a resort community, inside Gatlinburg city limits, which decides every permit and tax answer below. It carries a private hot tub, a recreation room with pool table and arcade console, a ten seat theatre room, mountain views, and a full bathroom per bedroom. On the market 79 days.

What it lacks is disclosure. The HOA fee is not published, only that there is one, billed monthly. Maximum occupancy is not stated, leaving an open regulatory question in section 9. Revenue for 2024, 2025 and 2026 is not disclosed. On a property marketed on its rental history, three missing years is the most conspicuous gap in the file, and the disclosed years fall every time: $92,950 in 2021, $76,331 in 2022, $71,785 in 2023, down 22.8%.

Sources: S31, S32 for the property; S42 for the scenario quartiles, which use the inclusive definition the workbook computes.

6 The cost side

Sources: as marked per line. Six lines are stated assumptions rather than sourced figures and say so. No financed illustration is shown: no rate or terms could be sourced, and an illustration built on an invented rate is worth less than nothing.

Two things drive it. Short-term rental use puts the parcel in Tennessee’s 40% commercial assessment band, not the 25% residential band. The county’s own 2026 estimators apply 0.40 for a short-term rental and 0.25 for a primary home at identical rates, so the taxable assessment rises 60% on the same appraised value. A model assuming residential treatment would understate property tax by about $1,740 a year. And the county appraises this property at $1,142,100, 43.3% above the asking price. A buyer paying $797,000 is taxed on $1,142,100 unless the assessment is appealed, and 2026 is a reappraisal year, so that window is live.

7 Amenities

There is no meaningful amenity gap at this property, and that is the finding. Split into thirds by revenue, every difference between the top three comps and the bottom three is one listing or zero, and three of the seven amenities are more common in the bottom third. The highest earner, AB08 at $126,700, carries only a hot tub and a game room. The lowest, AB07 at $32,100, carries a hot tub, a private pool, a game room, a theatre room and mountain views, more than any other comp. Amenity count does not track revenue here. Occupancy does: the top third runs 71.8% to 85.9% on method A, the bottom third 20.0% to 37.6%. No revenue effect is claimed for any amenity: with three listings a side, a one-listing gap is noise.

Sources: S48 (amenity split derived from S41 and S42), S47 (EV charger costs), S22 (market penetration: hot tub 83%, pets 24%), S32 (parcel record).

The upside case cannot rest on adding amenities, because there is nothing left to add. To beat the $71,785 the seller disclosed for 2023, it has to come from pricing, occupancy and management.

8 Location

Source: S45, driving distance and time, fastest route, no traffic weighting, captured 2026-09-07.

What helps. Everything a Gatlinburg guest books for is inside a fifteen minute drive, and the park entrance is fourteen minutes away. The park took 11,527,939 recreation visits in CY2025. This is a drive-to market: Knoxville one hour, Atlanta 4.6 hours, Nashville 4.2 hours, so most of the southeast is inside a weekend drive.

Wildfire is a specific, recent risk at this address. The November 2016 Chimney Tops 2 fire burned approximately 11,000 acres inside the park plus 6,000 acres outside it in Sevier County, with gusts to 87 mph throwing embers that started new fires. The subject property sits in the same wildland urban interface. This is live risk management, not history: Sevier County’s Community Wildfire Protection Plan was approved by the state on 3 August 2026. No insurance quote was obtained and none is estimated here. A buyer must get a bindable short-term-rental cabin quote before committing.

It is 2.1 miles from the Parkway, not on it. Six minutes by car, but not walkable, and two comps market walkability as their differentiator. Both of those comps sit in the bottom third by revenue, at $59,500 and $32,100, so on this evidence walkability is not what drives revenue here.

Two things were not verified and should be. The road surface and the gradient of the final approach were not inspected, so nothing is claimed about whether a two wheel drive rental car makes it up in winter ice. Cell coverage and broadband at the address were not measured; comps within 1.3 miles advertise verified speed tests of 220 and 454 Mbps, which is evidence about neighbours, not this property. Both are cheap to check in person and both can cost a cabin its winter bookings.

9 Regulation and tax

The property is inside the City of Gatlinburg, confirmed on the county record, which carries City: GATLINBURG, City #287. That decides the following.

Sources: as marked. The city portion uses the 2025 rate of 0.125700 because Gatlinburg’s 2026 certified city rate was not published in any readable source. That portion is 12.4% of the total.

Two points a buyer must act on. Gatlinburg reclassifies a tourist residency as R-1 or R-3, losing the state fire sprinkler exemption, if it exceeds three storeys, 5,000 gross square feet, or 12 occupants. The property is two storeys and 2,496 sq ft, inside two thresholds, but maximum occupancy is not disclosed, so the third is open. With four bedrooms and five full baths it is plausible it is marketed above 12 guests, which would trigger a sprinkler requirement and a real capital cost. Separately, grandfathering does not transfer: the Tennessee Short-Term Rental Unit Act of 2018 protects properties already operating against later local rules, but that protection ends when the property is sold.

10 Risks and what would change this

1. The undisclosed HOA fee. A resort unit on an undivided parcel with an HOA of unknown size, and the largest unknown in the model. Bites if: the fee exceeds $200 a month, which takes the yield from 4.0% to 3.7% on its own. Watch: the covenants and fee schedule, both obtainable before contract.

2. Insurance in a wildfire interface. Bites if: a carrier declines short-term-rental cabin cover here, or prices it well above a standard homeowner policy. Watch: a bindable quote, not an estimate. With a $200 HOA, a $2,500 premium takes the yield to 3.4%.

3. Three undisclosed revenue years. Disclosed revenue fell 22.8% from 2021 to 2023 and then stops. Bites if: 2024 and 2025 continued the decline, putting the property below this base case rather than at it. Watch: platform statements or tax returns in diligence.

4. The assessment sits 43.3% above the asking price. Bites if: an appeal fails or is never filed, leaving the buyer taxed on $1,142,100 rather than the price paid. Watch: the 2026 Board of Equalization window.

5. The 12-occupant sprinkler threshold. Bites if: the property is or becomes marketed above 12 guests. Watch: the current permit and the marketed maximum occupancy.

11 Method, sources and limitations

Date range. The trailing twelve months are September 2025 to August 2026, used throughout. Where a source offers only a different window, the table says so: park visitation is CY2025, Pigeon Forge lodging revenue is CY2024, one vendor covers August 2025 to July 2026.

Occupancy. Blocked nights over an 85 night forward window, 7 September to 30 November 2026, identical for every comp. Airbnb exposes only three months of calendar to a logged-out reader, so 85 rather than 90 is a stated deviation applied consistently. The method overstates occupancy, because an owner block, a maintenance hold and a real booking are indistinguishable from outside, so a stated 0.90 factor is applied on the assumption that 10% of blocked nights are not bookings. That factor is an assumption, not a sourced figure, and the unadjusted numbers are published so a reader can substitute their own. A second method, review velocity, was run and discarded: it returns occupancies above 100% on most comps, because neither the market review rate nor booking-level stay lengths are obtainable without a login. It is reported rather than dropped quietly.

Nightly rates. Airbnb shows only an all-in total to a logged-out reader here. Querying one check-in date for a 3 night and a 4 night stay isolates the nightly rate as the difference and the fixed fee as the remainder; both are then divided by 1.2805 to strip the guest service fee and the 12.75% lodging tax. Sampling ran in November 2026 and January 2027, not July and October, because the true peaks are booked out or beyond hosts’ open calendars, so the sampled high-to-trough spread understates the real one.

Annual revenue. Blocked nights over 85, times 0.90, times the seasonally weighted host-side ADR, times 365, rounded to the nearest $100. Both inputs are published per comp, so every revenue figure can be rebuilt from the columns beside it.

Rounding. Currency in whole dollars. Percentages to one decimal. Annual revenue to the nearest $100.

The data constraint. This report was built from public sources without a paid AirDNA seat; only the free public market pages were read. A seat would sharpen two things: comp-level realised revenue instead of calendar-derived estimates, removing the 0.90 owner-block assumption entirely, and a comp universe wider than eight listings, which would narrow the scenario range and make the amenity analysis testable rather than directional.

Sample size. The amenity comparison rests on three listings against three, reported as counts, with no revenue effect claimed.

Quality control. A formal QC pass ran before publication and found fourteen defects, thirteen fixed: revenue figures that did not reproduce from their own published formula, quartiles reported on two different definitions, a method note describing a sampling approach never used, two vendor rows carrying the wrong date window, and an occupancy range mixing adjusted and unadjusted figures. The log is published with the working.

The working. Every figure, formula and source sits in the Google Sheet: [GOOGLE SHEET LINK TO BE INSERTED BEFORE PUBLISH]. Ten tabs, including the live cost model, the full comp set, and a Sources tab holding all 49 source rows with URLs and capture dates.

This is analysis, not investment advice.