Home/Insights/What features are worth
Every feature I have measured against Dallas–Fort Worth sales, in one place. Each figure compares houses against their own neighbours in the same subdivision, so the school district, the street and the price tier are held still. Sample sizes and intervals are shown because they are the difference between a finding and a guess.
Updated September 2026 · 89,302 closings, 2023 – September 2026
| Feature | Adds | 95% interval | Sales | Costs to add |
|---|---|---|---|---|
| Swimming pool | +9.6% | +9.1 to +10.2% | 12,741 | $55,000 – $150,000 (gunite) |
| Guest house or separate quarters | +9.7% | +7.1 to +12.2% | 605 | $150 – $300 / sq ft |
| Pool house or cabana | +9.5% | +6.9 to +12.1% | 409 | — |
| RV or boat storage | +8.5% | +5.7 to +11.4% | 545 | — |
| Barn, stable or arena | +7.7% | +4.5 to +10.9% | 370 | — |
| Accessory unit — detached | +11.3% | +7.1 to +15.5% | 329 | $99,000 – $198,000 at 659 sq ft |
| Accessory unit — attached or unflagged | +5.9% | +4.1 to +7.8% | 945 | less; no local figure |
| Workshop | +5.7% | +4.6 to +6.9% | 2,125 | — |
| Solar — owned outright | +3.3% | +1.8 to +4.8% | 265 | about $29,000 |
| Residential elevator | +4.5% | −2.6 to +12.0% | 121 | $35,000 – $80,000 |
| Solar — leased | +0.2% | −2.7 to +3.2% | 91 | — |
| Solar — ownership not stated | −0.1% | −1.3 to +1.2% | 414 | — |
| Each extra 100 sq ft of accessory space | +0.30% | −0.52 to +1.12% | 753 | — |
The four rows in bold are the ones where the interval crosses zero. An elevator, a leased solar system, solar of unstated ownership and the size of an accessory unit all fail to separate from nothing, however plausible the point estimate looks.
Percentages are not decisions. Against what each thing costs:
| Feature | Roughly returns | Verdict |
|---|---|---|
| Swimming pool, $800,000+ neighbourhood | two thirds to most | Best large-project return |
| Swimming pool, $500,000 – $800,000 | half to three quarters | Strong |
| Workshop on acreage | well above cost | Best per dollar spent |
| Swimming pool, under $300,000 | a fifth to a third | Weakest pool case |
| Solar, owned outright | about 40% | Now rests on the power bill, not resale |
| Accessory unit, detached and small | half to most | Similar to a pool |
| Accessory unit, built large | a quarter to a half | Size adds nothing measurable |
| Residential elevator | roughly its cost, or nothing | Cannot be told apart from zero |
The workshop is the quiet winner of this whole body of research. A metal building on a slab with power costs a fraction of what habitable space costs to put up, returns 5.7% on a suburban lot and 8.8% on five acres or more, and nobody ever asks me about it.
Acreage does not behave like a feature, so it gets its own figures. From 5,671 closed land sales across seven North Texas counties:
| Tract size | Median price per acre |
|---|---|
| Under 1 acre | about $216,667 |
| 50 acres and up | about $17,299 |
Price per acre falls roughly a quarter every time a tract doubles, so doubling the acreage raises the total price by about 52% rather than 100%. A school district moves bare land substantially — within Parker County, Aledo ISD ran about 59% above Weatherford ISD and Poolville ISD about 38% below. And 69% of land listings never sell at all.
Both of these move the price a buyer actually pays without touching the house.
Property tax on new construction. The effective rate runs 0.38% of the sale price on a brand-new build, 0.61% at one year, and 1.93% at two years — the listed figure is the vacant lot's bill. On a $450,000 build that is about $2,054 shown against roughly $8,362 owed, a difference near $526 a month. Full analysis.
Seller concessions. About 55% of all closings carried one, and 77.3% of 2026 new-construction closings did, at a median of $10,000. New builds close at 96.8% of original asking on the recorded figure but 94.0% net of what the seller paid. Full analysis.
Every percentage above comes from the same method. Closed sales of detached single-family homes across Dallas–Fort Worth, 2023 through September 2026 — 89,302 of them. For each feature, the model compares houses that have it against houses in the same subdivision that do not, holding living area, lot size, year built and date of sale constant, with standard errors clustered by subdivision. Only subdivisions containing both kinds of house contribute.
That within-subdivision comparison is what separates these figures from the ones circulated online. The raw premium on an elevator is 35.1%; against its own neighbours it is 4.5% and indistinguishable from zero. Most published feature premiums are measuring the neighbourhood rather than the feature.
Where figures are banded by price, the bands are cut on the median price of the subdivision, not on the individual sale price. Splitting on sale price conditions on the outcome the model is explaining and biases every feature toward zero, worst at the bottom of the market. Done the wrong way, the pool premium in inexpensive neighbourhoods reads 2.4%; done correctly it is 8.4%.
They are not adjustments. A market-wide average applied to a specific property in a specific subdivision will be wrong, sometimes by a factor of three. These are the starting point for a local extraction, not a substitute for one.
The MLS records presence, not quality or condition. A $45,000 vinyl pool and a $250,000 gunite installation with a spa, slide and grotto record identically, as do a permitted casita and a converted garage.
Residual quality is not controlled. Houses with expensive features tend to be better finished in ways the MLS does not capture, so some unknown part of every premium here is that rather than the feature. Treat each figure as an upper bound.
And this is one market over one window. Dallas–Fort Worth, 2023 to 2026, a period of comparatively cheap land, cheap power and heavy new construction. None of it should be carried to another state.
Terrence Bilodeau is a Certified Residential Appraiser (TX-1360232) and a Texas real estate broker. Clearfork Appraisals provides private-client appraisals in Fort Worth and across Tarrant, Parker, Johnson and Hood counties. This page is a general market study. It is not an appraisal, not tax advice, and not a valuation opinion about any specific property.
Measured against neighbouring sales in the same subdivision across 89,302 Dallas–Fort Worth closings: a guest house or separate quarters adds 9.7%, a swimming pool 9.6%, a pool house or cabana 9.5%, RV or boat storage 8.5%, a barn or stable 7.7%, a detached accessory unit 11.3% and an attached one 5.9%, a workshop 5.7% and owned solar 3.3%. A residential elevator, leased solar, and the size of an accessory unit all fail to separate from zero.
A workshop, per dollar spent. A metal building on a slab with power costs a fraction of what habitable space costs to build and returns 5.7% on a suburban lot and 8.8% on five acres or more. Among larger projects, a swimming pool in a neighbourhood above $500,000 returns at or above its build cost.
Building a large accessory unit. Size adds nothing this data can measure, so a 1,200 square foot unit costing $180,000 to $360,000 carries the same premium as a 450 square foot one costing a third of that. A small detached unit recovers about half to most of its cost; a large one does far worse.
Each figure compares houses that have the feature against houses in the same subdivision that do not, holding living area, lot size, year built and date of sale constant, with standard errors clustered by subdivision. Only subdivisions containing both kinds of house contribute. The source is 89,302 closed sales of detached single-family homes across Dallas–Fort Worth from 2023 through September 2026.
No. They are market-wide averages and the same feature is worth materially different amounts in different neighbourhoods — on pools the spread between the strongest and weakest market is roughly three to one. They are a starting point for extracting the figure locally, not a substitute for doing so.
Every figure on this page links to the full analysis it came from, including the confidence intervals, sample construction and limitations.
Everything on this page is a market-wide average. What a feature contributes on your street is extracted from sales on your street, and it is regularly a different number. If you need that done properly for an estate, a divorce, a protest, a refinance or a sale, that is an appraisal. Written fee quote, typically within one business hour.