Named for Tax Code §42.26 · Equal & Uniform
Ranked over-assessment leads. The equity exhibit your case is built on.
We run the §41.43(b)(3) equal-and-uniform test on every commercial parcel in the certified roll, rank the results by cap-aware estimated savings, and deliver TDLR-registered consultants a prospect list with a one-page evidence sheet per parcel — prepared in your firm's name.
Data and software onlyWe never represent ownersCalibrated on published ARB outcomes
| Parcels | n | Reduced | Mean cut | Cut >15% |
|---|---|---|---|---|
| Flagged | 591 | 92.0% | 11.9% | 29.6% |
| Unflagged | 726 | 86.4% | 8.7% | 17.5% |
Identical screen, run on prior-year noticed values and joined to HCAD's published hearing outcomes. 2025 replicates: 92.6% vs 88.4% reduced. Full tables below.
Travis County · TY2026 · Certified roll 07-18-2026
What's in the list
Every figure is computed from the county's own certified export and rate tables — nothing scraped, nothing estimated by hand.
Cap-aware savings, parcel by parcel
F1–F5 commercial parcels ranked by estimated annual savings at each parcel's actual composite rate, with §23.231 circuit-breaker math applied so capped parcels show the taxable path and the future-year benefit separately.
Tiers you can rely on
HIGH / MEDIUM / LOW from cohort size and dispersion. Only evidence-grade cohorts — same improvement type, same market area — receive evidence sheets. Low-confidence flags showed no measured lift, so they're labeled and excluded.
The exhibit that slots into your case
Comp grid with documented size and age adjustments, the median math, indicated value, estimated savings, and full source citations — one page, "Prepared for your firm." Informal-ready as is; for formal hearings, the same lead expands to a full packet (comp detail, methodology, verbatim statute, provenance). You own the case — this is the equity exhibit inside it.
The junk is already removed
Parcels in active litigation, public and institutional owners, and assets over $25M (quarantined with a written memo each) never reach your list. Ambiguous lapse signals are triaged: likely-still-represented parcels are removed entirely, soft lapses are labelled pending confirmation, and only verified lapses make the headline counts.
Harris backtest · TY2024–2025 · Published outcomes
Measured, not promised
We ran the identical screen on two prior years of noticed values, then joined the results to HCAD's published ARB hearing outcomes. These are the numbers, with sample sizes.
| Formal ARB hearings (FC/FN), high confidence | n | Reduced | Mean cut | Cut >15% |
|---|---|---|---|---|
| 2024 · Flagged | 591 | 92.0% | 11.9% | 29.6% |
| 2024 · Unflagged | 726 | 86.4% | 8.7% | 17.5% |
| 2025 · Flagged | 581 | 92.6% | 10.1% | 23.6% |
| 2025 · Unflagged | 794 | 88.4% | 8.4% | 18.9% |
Across all protesters in high-confidence cohorts (agent-represented): 87–91% vs 79–85% reduced; mean cut 8.8–9.4% vs 7.0–7.1%; cuts over 15% in 18–21% vs 14–15% of cases. Pro-se owners (smaller samples: n=254/352 in 2024, 129/321 in 2025): 82–85% vs 69–76% reduced, 11–14% vs 8–9% mean cut. About 90% of Harris commercial protesters use an agent, and flagged parcels are represented at the same rate — no selection effect. Lift measured among parcels that protested; low-confidence flags showed no lift and are excluded from evidence sheets.
Read the full methodology →See exactly what you get — sample formal-hearing packet (PDF, watermarked) →
How we validate · ✻ checked against sources we don't control
Trust, earned the boring way
Before any list ships, the pipeline has to pass five independent checks. Each compares our output against something outside the pipeline — the county's own publications, live records, or a from-scratch reimplementation.
Our Travis roll total matches TCAD's published 2026 Certified Totals within +0.37% — and every point of that variance is traced to a documented category difference, not error.
30 randomly drawn parcels — leads, comps, and whales — verified field-by-field against TCAD's live records: 0 errors in 150 comparisons.
Top-ranked evidence sheets recomputed from scratch in a separate SQL implementation that shares no code with the pipeline: 0 mismatches across 65 figures.
A fresh copy of the code, run on the same raw export, reproduces every output hash-identical. No randomness, no hand edits, no AI anywhere in the math.
6,900 deliberate defects injected into the code, one at a time; the test suite catches every one that touches the arithmetic. Zero survivors.
Validation is re-run for every county and season before delivery. The full validation report is available to customers on request.
Packages · Travis County · 2026 season
Two products, priced like the tools they replace
A season of prospecting ammunition, or evidence production for the book you already have. Delivered within two business days of purchase.
- Full ranked list as CSV — parcel, owner, mailing address, tier, cap-aware savings
- Evidence sheets (PDF) for the top 100 ranked leads, prepared in your firm's name
- Lead classes: never-represented, recently-lapsed (with prior firm and expiry date), soft-lapsed (30, labelled pending confirmation), prior-protester, represented
- Everything in the Season Package
- Founding bonus, 2027 season: full evidence-sheet coverage of every evidence-grade lead at no charge
- Milestone billing — the second half is due only on results
No savings guarantee. No refunds after delivery. Milestone terms on the receipt.
- Send five parcels you represent; get five sheets free, no card required
- Then $30/sheet for medium- and high-confidence flags across your book
- Comp data included as CSV alongside every batch
Free sample · Personalized to your book
See your county before you spend a dollar
A watermarked top-10 for Travis, personalized to your firm
Email your firm name and TDLR license number and you'll receive the ten highest-ranked leads with full evidence detail, watermarked, in your firm's name. Three free half-list pilot seats are reserved for firms willing to share season outcomes — ask.
or write to praneeth@4226tax.com
The 20% circuit breaker on non-homestead parcels valued at $5.32M or less expires December 31, 2026, with no extension enacted as of August 2026. 2026 is the last cycle in which lowering a capped parcel's market value is a "future-year benefit" rather than next year's full exposure. Our savings figures report both paths separately.