How the assessment check works

01Output

What does the check measure?#

Objective facts about public records — for example, how your building assessment per square foot compares with similar homes in your assessment neighborhood. In Illinois the result is facts only: you choose the value you request.

Reports never use the words appraisal, valuation report, opinion of value or certified. They show the qualifying pool of comparable homes, the ones selected by the published rule, and where your home falls among them.

The free result leads with one figure: how much higher the county values your home than the lowest assessed of the similar homes picked by the rule below. It is (your assessment per square foot − that home’s) × your square feet, turned into market value terms with the county’s level of assessment (Cook County: 10% for homes) and rounded to the nearest $100. The tax line applies the county’s own tax calculation to that difference, with a range for next year’s rates; the 3-year view adds the same yearly figure up over three years.

In states whose review boards don’t accept neighbors’ assessments, the figure compares your value with what the official house price index for your area supports (Federal Housing Finance Agency, your ZIP code where it has an index, otherwise your county). California: the base-year value implied by the roll (your assessed value ÷ the Proposition 13 inflation factors since your base year) × the index change from your base year to the last full year. Oregon: last year’s real market value × the index change, shown only where your tax follows the market value (below your capped value). Maryland: your previous assessment × the index change over the same three years, with the tax line following the phase-in (a third of an increase each year) and the homestead credit. The difference is shown only when it is positive, and the tax effect applies the local tax rate on record with a small range. It is an index calculation, not a valuation of your home; you choose the value you request.

02Evidence types

Which kinds of evidence does it compute?#

Uniformity (how similar homes are assessed), recent arm’s-length sales, and errors in your property record — plus a free check for exemptions you may be missing. Each county’s configuration lists which apply at each step.

03Comparable homes

How are comparable homes chosen?#

Similar homes are found with filters that never look at price or assessment. For assessment comparisons we then take the lowest assessed of them, after leaving out homes assessed far below the rest; sales are chosen by similarity alone. The whole pool is disclosed next to the homes taken from it.

Assessment comparisons (uniformity)

1. Matching homes: every home in the same class and the same assessment neighborhood (no distance limit), with living area within 15% and built within 10 years of yours, and the same number of units. The same published rule for every home; these filters never look at price or assessment.
2. Outlier guard: homes assessed below 65% of the matching group’s median (per square foot) are left out as likely damaged, vacant, part-year or record errors, and a record repeated on another parcel number counts once.
3. Lowest first: from the rest, the homes assessed lowest per square foot, nearest first on ties: the 5 lowest on the free result and up to 15 in the packet, in every plan.
4. You decide: the free result shows the 5 homes with their address, size and distance; the packet lists every home with its address, photo where available and match details, and discloses the pool size, how many were left out and how many are assessed lower than yours. You choose which homes to put on the county form (Cook County Board of Review: 3 to 5 homes; the filing guide marks the 5 lowest by default) and the value to request.

If too few homes match, the size and age limits are relaxed in the published steps below; the step is chosen on the number of matching homes alone. The similarity score shown for these homes leaves out location, which the neighborhood rule covers.

Sales comparisons

Hard filters come first: same class code, same neighborhood, same number of units, the allowed sales window and arm’s-length sales only. Similarity is then scored from 0 to 100, with each distance measured in the units below. At most 2 homes come from one block face, and we prefer one larger and one smaller home than yours.

If too few homes qualify, the filters are relaxed step by step and each step lowers the confidence score; the step used is disclosed. Neighboring areas are never used for Cook County uniformity comparisons.

Excluded sales: under $10,000, excluded deed types, resales within a year, multi-parcel sales, plus price outliers beyond 1.5× the interquartile range (trimming at most 10% of sales).

04Quality gates

What quality checks run before a result is shown?#

Every result passes, goes to manual review, or is blocked on checks such as parcel match, comparable count, similarity, dispersion and data freshness. Blocked results are never sold.

05Honest screening

When does a result say the data shows a clear difference?#

Every result shows the measured facts and all three plans. Plans are unavailable only when something rules a sale out: the filing window is not open or has closed, a legal screen-out applies, or we could not confirm the parcel record. The “assessed above most similar homes” badge and the savings estimate appear only when every condition below holds; otherwise the result is worded neutrally.

• Evidence: market over-assessment of at least 30% (low-end estimate at least 3%), or uniformity at least 1× the comparables’ median and higher than at least 3 of them.
• Estimated chance of a reduction at least 50%; overall confidence at least 0.7.
• Chance of an increase at most 5%, and the expected cost of an increase at most 20% of expected savings.
• Expected value after the price and any filing fee at least $150.
• At least 24 hours before the filing deadline: plans can be bought until then.
• Expected yearly savings of at least $150.

Cases where caps or freezes mean a lower assessment would not lower the bill fail the savings test. Whatever the result, an appeal can lead to no change, and at some stages the assessment can go up.

06Risk

Can an appeal make my assessment go up?#

In some places, yes: some review bodies may raise an assessment. The check estimates that risk from the data and shows it in words and dollars; above the threshold above, it offers no report.

Each county guide states whether and when a value can go up at each step, with the official source. County guides

07Limits

What does the method not promise?#

Results are not guaranteed. The review body decides, and an appeal can be denied. The check reports facts about public records; it is not an appraisal and not legal advice — not a substitute for the advice of an attorney.

The same inputs always produce the same output: the engine is deterministic, uses no AI, and records the data release, configuration and engine versions in every report.