How to find comparable sales for a property tax appeal

Last updated 2026-07-09

What are comparable sales and why do they matter for property tax?

A comparable sale, or comp, is a recent arms-length sale of a property similar enough to yours that a reasonable person would use it to estimate your home's market value. That matters because most states require assessors to value residential property at or near fair market value, and market value is defined by what similar properties actually sold for. ^[1]

The assessor used comps to set your value. If they picked the wrong ones, stretched the time window, or ignored a neighborhood boundary, you can use better comps to knock the number down. That's the whole game. You're not arguing that your house is special or that the tax rate is unfair. You're showing the board that the math doesn't hold up against actual sales.

In nearly every state, the evidence standard for a residential appeal is the preponderance of the evidence, meaning your side just has to be more persuasive than the assessor's side. ^[2] Three clean comps that average $40,000 below your assessed value are usually enough to win, or at least force a negotiated reduction at an informal hearing.

Here's the gap that should bother you. The National Taxpayers Union Foundation has estimated that 30% to 60% of U.S. properties are over-assessed, yet fewer than 5% of homeowners ever appeal. ^[3] Most of that gap comes down to one thing: homeowners not knowing how to find comps themselves.

What time period do comparable sales need to cover?

Every state sets an assessment date, the single calendar day your property's value is supposed to reflect. Your comps need to bracket that date as tightly as possible. ^[1]

Most assessment boards accept a window of 6 to 12 months on either side of the assessment date. Sales older than 18 months are usually too stale unless your market barely moved, and you'd need to say so explicitly. Sales that closed after the assessment date can actually help in many jurisdictions, because they show what buyers were paying around the same period.

Here's a practical example. If your state has a January 1 assessment date and you're appealing in spring, you'd ideally use sales from roughly July through December of the prior year, plus any January and February sales from the appeal year. That gives the board a clean picture of the market on the valuation date.

Most big states anchor to January 1. California uses January 1 as the lien date. Texas, Florida, Illinois, and Georgia all use January 1. New York City runs on a January 5 taxable status date. Look up your specific state's assessment date before you pull a single comp, because if your sales all fall in the wrong period, the board can ignore them. ^[11]

State	Assessment date	Typical comp window
California	Jan 1	Jul 1 (prior yr) to Jun 30 (current yr)
Texas	Jan 1	Jul 1 (prior yr) to Jan 1 (current yr)
Florida	Jan 1	Jul 1 (prior yr) to Jan 1 (current yr)
Illinois	Jan 1	Jan 1 (prior yr) to Jan 1 (current yr)
New York (city)	Jan 5	18-month lookback common in practice
Georgia	Jan 1	Typically 12 months prior

When in doubt, call your county assessor's office and ask what comp window they used. They'll usually tell you.

How far away can comparable sales be from your property?

Most residential appraisal guidelines start with a half-mile to one-mile radius, but geography is really a stand-in for neighborhood. What you want is sales that competed with your home in the same buyer pool. ^[4]

If you live in a dense urban neighborhood where every block has similar rowhouses, half a mile might give you 40 comps to pick from. If you're on two acres in a rural county, you might have to go three miles or more to find anything comparable, and a good appraiser defends that distance by noting there simply aren't closer sales.

Your comps should sit in the same school district, same flood zone, same general price tier, and under the same infrastructure. A sale across a major highway, across a school district boundary, or in a subdivision with a homeowners association when your home has none, those are weak comps. The assessor's attorney will say so at the hearing.

For rural properties, comparable distance can run large. The Uniform Standards of Professional Appraisal Practice (USPAP), which most state appraisers are bound by, requires the appraiser to identify a market area that reflects how buyers actually behave rather than an arbitrary radius. ^[4] You can make that same argument yourself at a board hearing.

What property characteristics need to match your home?

You don't need an identical twin. You need properties close enough that the adjustments stay small. The standard appraisal framework calls for dollar adjustments on each difference, but at an informal residential hearing, boards mostly want to see that you made a reasonable effort to find similar homes rather than cherry-pick tiny cottages against your four-bedroom colonial.

The characteristics that matter most, roughly in order:

Gross living area (GLA). This drives price-per-square-foot differences more than anything else. As a rough guideline, appraisers get uncomfortable when one property is more than 20% to 25% larger or smaller than the subject. If your home is 1,800 square feet and your best comp is 2,400, you either explain the gap or find a better comp.

Bedroom and bathroom count. A three-bed, two-bath shouldn't be comped against a five-bed, three-bath without a real adjustment.

Age and condition. A 1960s ranch and a 2005 colonial in the same zip code are not comps, even at identical square footage. Age is a proxy for construction quality, layout, and deferred maintenance.

Lot size. In suburbs and rural areas, lot size can matter as much as living area. In dense urban markets it matters less, but it still matters.

Garage and basement. Finished basement space is typically valued at 50% to 60% of above-grade space in appraisal practice. ^[4] Unfinished basement is worth less. No garage versus a two-car garage is a real dollar difference.

Style. Ranch versus two-story matters in some markets, not others. Split-levels are notoriously hard to comp. Know your market.

If your comps aren't perfect matches, fine. Write a one-paragraph note on each difference and whether you adjusted for it. Boards give credit for showing your work.

Where can you find comparable sales data for free?

You have several solid free options. None is perfect alone. Use at least two to cross-check.

Your county assessor's or auditor's public records portal. This is the best primary source because it's what the assessor used. Most counties now run a searchable database of recorded sales. Search by subdivision, street range, or map parcel ID. You can often filter by date and download to a spreadsheet. ^[5] If your county posts this data, start here. (See the Los Angeles County property tax and Cook County tax assessor tax bill portals for what large county systems look like.)

Zillow's Recently Sold filter. Set the map to your neighborhood, click Sold under listing status, and filter to the past 12 months. Zillow pulls MLS data in most markets and public records everywhere else. It's reasonably complete for arms-length sales. ^[6]

Redfin's sold history. Redfin tends to be cleaner than Zillow in markets where it operates, and it shows days on market and price reductions, which flag whether a sale was distressed.

Your county recorder or register of deeds. Deed transfers are public record in every state. The recorder's site won't give you much detail, but it confirms sale dates and prices you can cross-reference against the assessor database.

State-run property transfer databases. Some states centralize this. Pennsylvania's State Tax Equalization Board publishes countywide ratio studies built on actual sales data. Illinois runs Department of Revenue ratio studies. These give context even when they don't hand you individual comps. ^[7]

FHFA House Price Index. If you need to argue your assessment date landed during a price decline, the Federal Housing Finance Agency's HPI tool shows price trends by metro and zip code. ^[8] That's not a comp, but it's supporting context.

What to skip: Zillow's Zestimate as evidence. It's an automated model, not a sale. Boards dismiss it on sight, and rightly. Active listings are not comparable sales either. Asking prices are not sale prices.

How do you actually run a comparable sales search step by step?

Here's the process I'd follow appealing my own assessment.

Step 1: Confirm your assessment date and your property's characteristics from the assessor's record. Log in to your county portal and pull your own property card. Write down the square footage, year built, bedroom count, bathrooms, garage, and basement finish the assessor has on file. If those numbers are wrong, that's a separate ground to appeal. If they're right, those are your search parameters.

Step 2: Pull all sales in your immediate area for the date window. Go broad first. Search within a one-mile radius or your subdivision for all residential sales in the past 12 to 18 months. Download to a spreadsheet if you can. You want a raw list before you filter.

Step 3: Filter for property type. Drop condos if you own a single-family home, drop commercial, drop any transfer marked non-arms-length (foreclosure, estate sale, family transfer). Most assessor databases flag these. You want fair market sales only.

Step 4: Apply your characteristic filters. Cut anything more than 25% smaller or larger than your home. Cut anything built more than 20 to 25 years apart. Flag but don't automatically cut different bedroom counts. You can keep those if they're otherwise close.

Step 5: Rank the survivors by similarity. Closest geographic match, closest square footage, sale date nearest the assessment date, ranked highest. Sort your spreadsheet by those criteria.

Step 6: Pick your best 3 to 6 comps. Enough to show a pattern, not so many that you drag weak ones in. Three strong comps beat six mediocre ones every time. Calculate the average and median sale price per square foot, then apply that figure to your home's square footage. If the result lands materially below your assessment, you have a case.

Step 7: Print property detail sheets for each comp. From the assessor's site or Zillow, print or screenshot the property facts and sale history for each comp. You'll submit these at your hearing. Some boards want a formal grid, others just want the printouts. Check your county's filing instructions.

If you want a pre-built spreadsheet template and a hearing-ready comp grid, the TaxFightBack DIY appeal kit has both, which saves a couple hours of formatting.

For county-specific filing procedures in Georgia, the Gwinnett County tax assessor, Cherokee County tax assessor, and Coweta County tax assessor pages each walk through the local form requirements.

How do you calculate price per square foot from your comps?

Price per square foot (PPSF) is the simplest way to summarize what buyers actually paid per unit of living area. It's not a perfect valuation method, but it's the language boards understand.

The math is easy. Take the sale price, divide by gross living area in square feet. Do it for each comp. Average them. Multiply the average by your home's square footage. Compare to your assessed value.

Example:

Address	Sale price	Sq ft	PPSF
14 Oak St	$310,000	1,750	$177.14
27 Maple Ave	$295,000	1,680	$175.60
8 Birch Ln	$322,000	1,810	$177.90
Average			$176.88

If your home is 1,760 square feet, the implied market value is $176.88 x 1,760 = $311,309. If your assessment is $360,000, you have a $48,691 gap, which turns into real tax savings depending on your mill rate.

A few cautions. PPSF works well for homes in a narrow size band. It breaks down when you compare a 1,200 square foot home to a 3,000 square foot home, because larger homes usually carry a lower PPSF. If your comps vary widely in size, a plain average will mislead. Run a size-adjusted comparison instead, or tighten your size range.

Apply PPSF only to above-grade living area. Finished basements, detached garages, and outbuildings carry lower per-square-foot values in appraisal practice. ^[4] If your assessor counted finished basement space at the same rate as above-grade space, flag that separately.

What makes a comparable sale get rejected at a hearing?

Boards and assessors' attorneys have seen every flavor of weak comp. The common rejection reasons:

Non-arms-length sales. Foreclosures, short sales, estate sales, and family transfers don't reflect what a typical buyer would pay. Many states exclude them from the assessor's own ratio studies for that reason. ^[7] If a sale shows up well below market, check the deed type or the assessor's notes. If it was distressed, cut it.

Sales outside the comp window. A sale from 30 months ago in a market that moved 10% a year is a misleading data point. Assessors will call it out immediately.

Meaningfully different condition. A recently renovated flip and an original-condition home from the same year and block are not comps without a real adjustment. If you can't document the adjustment, the board may discount the comp.

Cherry-picked outliers. If your neighborhood had 20 sales and you present the three cheapest while ignoring the other 17, an experienced board notices. Show your methodology honestly: here's my search, here's how I filtered, here's what came out.

Wrong property type. Townhomes are not comps for detached single-family. Condos are not comps for townhomes. Be precise about property type.

Price cuts that signal distress. A home listed at $450,000 that sold in 90 days for $340,000 after multiple cuts might not be a clean arms-length sale. Check days on market and listing history before you include it.

Do you need an appraisal, or can you use comps you found yourself?

For an informal residential hearing, self-found comps work. Boards at the informal level aren't bound by USPAP, and they regularly accept homeowner-prepared comp grids. The goal is a reasonable presentation of evidence, not a licensed appraisal.

For a formal hearing or an appeal to a state board of equalization, a licensed appraisal is stronger, but it costs money. A residential appraisal typically runs $300 to $600, more in high-cost markets. ^[9] Whether it's worth it depends on how much tax is at stake. Fighting over a $2,000 annual bill with a $500 appraisal is reasonable math. Fighting over $400 is not.

For Tax Court or circuit court litigation, you almost always need a licensed appraisal. Very few residential appeals go that far.

Here's the honest middle ground. Do your own comp search first. If the numbers clearly support a reduction and the informal hearing offer is fair, you won't need an appraisal. If you reach a formal hearing and the assessor pushes back hard, then weigh whether an appraiser makes financial sense.

For Maricopa property tax and San Diego property tax appeals, the local county boards have specific documentation standards worth reviewing before you decide on a full appraisal.

How does the assessor's comparable sales grid differ from yours?

When you request your assessment record, many counties provide the assessor's own sales comparison grid, or you can ask for it. This is the form their appraiser used to value your home. Read it closely.

The assessor's grid usually shows three to six comps with adjustments for each characteristic difference. Look for:

Sales outside the standard window. If their oldest comp is 22 months old, argue it should be discounted.

Adjustments that seem low. If your home has no garage and their comp does, and they only adjusted $5,000 in a market where a two-car garage adds $25,000, that's worth challenging.

Comps from a different neighborhood. If they used a sale from the subdivision across the highway that sold higher on better schools, flag it.

Upward adjustments that inflate the comps toward your assessed value. Sometimes assessors adjust up for condition, renovation, or lot size in ways that are hard to verify. Ask how they documented the condition rating for each comp.

Your goal isn't to attack the assessor personally. It's to show that your set of comps, chosen by a consistent and reasonable method, produces a lower value. You can acknowledge their comps exist and still argue yours are more representative.

The Bexar County tax assessor, Madison County tax assessor, and Bibb County tax assessor sites each post sample appeal forms, and in some cases sample comp grids, which are useful models.

What format should you present comparable sales in at your hearing?

Keep it simple and visual. Boards hear dozens of cases in a day. A clean one-page grid beats a 20-page packet every time.

The standard residential comp grid includes:

• Subject property details (your home) in the leftmost column
• Three to six comp columns to the right
• Rows for address, sale date, sale price, square footage, year built, bedroom and bath count, lot size, garage, basement, and any major feature differences
• A final row showing adjusted value for each comp
• A summary box: average of comps, your assessed value, the difference

Attach printouts of the assessor's property record for each comp. That proves the comp exists and matches your claimed characteristics.

Bring three copies: one for you, one for the board, one for the assessor's representative if they attend. Some jurisdictions require you to submit evidence in advance, sometimes 5 to 7 days before the hearing. Check the local rules.

Speak in the board's terms. Don't say "your assessment is outrageous." Say: "The comparable sales I've identified average $176 per square foot. At my home's square footage, that produces a market value of $311,000. My current assessment of $360,000 exceeds that by $49,000, which appears inconsistent with what buyers actually paid in my neighborhood during the relevant period."

That's a complete, professional argument a three-person board can evaluate in under two minutes. If your numbers are solid, it works. The TaxFightBack appeal kit includes a pre-formatted comp grid built on exactly this structure, useful if you want a starting template instead of building one from scratch.

Can you use online estimates like Zillow's Zestimate as a comp?

No. And I'd strongly advise against trying.

A Zestimate is an algorithmic model output. It is not a sale. Assessment boards in every state separate market evidence (actual sales) from model outputs (estimated values). The National Association of Realtors notes that automated valuation models carry a median error rate that varies widely by market and data density. ^[10] Zillow itself discloses that its Zestimate has a median error rate of roughly 2.4% for on-market homes and 7.7% for off-market homes nationally. ^[6] On a $400,000 home, that off-market range means the model could miss by $30,000 or more.

Boards know this. Lead with a Zestimate and you signal you haven't done the real work, which costs you credibility on everything else you present.

Use Zillow's Sold filter to find actual sales. Use the Zestimate only as a rough sanity check for yourself before you start. Never put it in your comp grid.

Same logic for Redfin estimates, realtor.com estimates, and any other AVM product. They're search tools, not evidence.

Frequently asked questions

How many comparable sales do I need for a property tax appeal?

Three to six is the standard range for a residential appeal. Three clean comps that clearly support a lower value beat six weak or inconsistent ones. Most state assessment boards set no minimum, but fewer than three invites the objection that you cherry-picked. Six is usually the ceiling before your presentation starts to look padded.

What is an arms-length sale and why does it matter for comps?

An arms-length sale is a transaction between two unrelated parties, each acting in their own interest, with neither under duress. Foreclosures, estate sales, short sales, and family transfers are typically excluded because they don't reflect what a willing buyer would pay a willing seller under normal conditions. Most state assessment statutes define market value using arms-length criteria. Using non-arms-length sales as comps weakens your case.

Can I use MLS listings as comparable sales if they haven't closed yet?

No. Active listings are asking prices, not sale prices. Buyers frequently pay less than list, and sometimes more in competitive markets. Assessment boards require closed sales. Pending sales don't count either. Only fully closed, recorded transactions qualify as comparable sales evidence for a property tax appeal.

What if there are no comparable sales near my property?

This is common in rural areas, unusual properties, or slow markets. Your options: extend the geographic radius while documenting why you had to go farther, extend the time window while noting stable market conditions, use the assessor's own comps but argue their adjustments were wrong, or hire a licensed appraiser who can defend a sparse-data analysis. A well-documented thin-market argument is still winnable.

Do comparable sales from foreclosures or bank sales count?

Generally no, but it depends on your state's statute. Many states explicitly exclude distressed sales from the definition of market value used for assessment. If distressed sales dominated your market during the assessment period, some states allow them as evidence of what the market actually was. Check your state's assessment statutes or ask your county assessor how they treat foreclosure sales in their ratio studies.

How do I find out what comparable sales the assessor used to value my property?

Request your property's assessment record or property card from your county assessor. Many jurisdictions let you pull this online. Others require a public records request. If the assessor prepared a formal sales comparison grid, that document should be part of your assessment file and is typically available on request before or after you file an appeal. Some counties mail it with your notice of assessment.

What is the assessment date and how does it affect which sales I can use?

The assessment date is the specific calendar day your property's value is supposed to reflect. Sales need to fall close to that date to be credible comps. Most states use January 1. Your comps should generally sit within 6 to 12 months of it. Using sales from a different market period, like a hot market two years earlier, is a common mistake assessors point out at hearings.

Can I use a sale of my own home as evidence of its value?

Yes, if you bought your home in an arms-length transaction within a reasonable period of the assessment date. A recent purchase price is strong evidence of market value because it's an actual negotiated transaction. If you bought within the past year or two, that price is often the single most persuasive piece of evidence in your appeal. Bring your closing disclosure or HUD-1 statement to the hearing.

How do I adjust for differences between my home and a comparable sale?

Common adjustments cover square footage, bedroom or bathroom count, garage size, finished basement, lot size, and condition. In formal appraisal, each adjustment is a dollar amount from paired sales analysis. At an informal hearing, use a simplified approach: note each difference and apply a reasonable market-based dollar adjustment. Many boards accept a brief written explanation of your adjustments rather than a full appraisal grid.

Is price per square foot a good way to present comparable sales to a board?

It works well for a simple, similar neighborhood. Calculate sale price divided by gross living area for each comp, average the results, and multiply by your home's square footage. If that number lands materially below your assessment, you have a clear argument. The weakness: PPSF assumes linear value scaling, which breaks down across large size differences. Use it as one part of your presentation, not your only metric.

What should I do if the assessor presents different comps at the hearing than I expected?

Stay calm and ask when those comps were provided or made part of the record. Many jurisdictions require both sides to submit evidence before the hearing. If the assessor introduces new comps you haven't seen, you can request a continuance to review them. Don't feel pressured to concede. Check each comp they present for issues: sale date, arms-length status, property type, and distance from your home.

Does an online appraisal tool give me the same result as a formal appraisal for appeal purposes?

No. Automated valuation tools like Zestimate are not accepted as evidence by assessment boards. They're model outputs, not appraisals. A formal appraisal by a state-licensed appraiser carries legal weight, follows USPAP standards, and can be defended under cross-examination. For most informal residential appeals you don't need a formal appraisal, but if you go to a formal hearing or Tax Court, a licensed appraisal is the appropriate evidence standard.

How long does a comparable sales search actually take if I do it myself?

Realistically 2 to 4 hours for a typical suburban property with good online data. That covers pulling your property record, running the search, filtering, picking your best comps, and formatting a simple grid. Rural properties or counties with poor online data take longer. Unusual property types like lakefront, historic, or large acreage can take much more time to find defensible comps.

Sources

1. International Association of Assessing Officers (IAAO), Standard on Mass Appraisal of Real Property: Most states require residential property to be assessed at or near fair market value, defined as the price in an arms-length transaction on the assessment date
2. National Conference of State Legislatures, Property Tax Assessment Appeals: The standard of evidence at most state assessment appeal hearings is preponderance of the evidence
3. National Taxpayers Union Foundation, Property Tax Assessment Study: An estimated 30% to 60% of U.S. properties may be over-assessed, yet fewer than 5% of homeowners appeal
4. Appraisal Foundation, Uniform Standards of Professional Appraisal Practice (USPAP): USPAP requires appraisers to define a market area based on buyer behavior and apply paired sales analysis for adjustments; finished basement space is valued below above-grade space
5. Cook County Assessor's Office, Property Search Portal: County assessors maintain searchable public databases of all recorded residential sales used in assessment
6. Zillow Research, Zestimate Accuracy: Zillow's Zestimate has a median error rate of approximately 2.4% for on-market homes and 7.7% for off-market homes nationally
7. Illinois Department of Revenue, Property Tax Assessment and Equalization: State ratio studies exclude non-arms-length and distressed sales when measuring assessment levels
8. Federal Housing Finance Agency, House Price Index (HPI): FHFA's House Price Index provides metro and zip-code level price trend data usable to document market value changes near an assessment date
9. Appraisal Institute, Residential Appraisal Cost Range: A residential appraisal for appeal purposes typically costs $300 to $600, sometimes more in high-cost markets
10. National Association of Realtors, Automated Valuation Model FAQ: Automated valuation models have a median error rate that varies widely by market and data density, noted in NAR's AVM guidance
11. California State Board of Equalization, Assessment Appeals Manual: California's assessment date is January 1 (lien date) and comparable sales evidence should reflect market conditions as of that date
12. Texas Comptroller of Public Accounts, Property Tax Appraisal and Appeals: Texas assessment date is January 1; comparable sales used in appraisal district valuations should reflect market conditions on or before January 1 of the tax year