What makes a comparable good enough to use
Similarity on recorded attributes, proximity, and nothing clever. Four tight ones beat twenty loose.
Boards refuse more uniformity appeals over comparable selection than over arithmetic. A comparable that does not survive scrutiny does not merely fail to help — it damages the rest of your set.
The five tests
1. Same valuation neighbourhood
Assessment models work in defined neighbourhoods, and this matters more than physical distance. A property four streets away inside your neighbourhood is a better comparable than one across the road in a different one, because they were priced by different parts of the model.
Cook County publishes a neighbourhood code on each parcel. Match it.
2. Same class code
Different classes are assessed on different bases. A class 203 one-storey and a class 205 two-storey are not interchangeable, however similar the floor areas.
3. Similar floor area
Within roughly 20%. This matters because assessment per square foot is not linear. Smaller houses generally carry a higher assessment per square foot than larger ones — fixed elements like a kitchen and a bathroom are spread over less area.
So comparing your 1,400 square foot house against 2,800 square foot neighbours flatters you in a way the office will spot immediately, and it is the single most common weakness in an amateur filing.
4. Similar age
Within a decade or two. A 1925 bungalow and a 1998 build are different products with different depreciation, whatever their floor areas.
5. Similar style and storeys
A ranch, a split-level and a two-storey colonial of identical floor area are three different buildings. Where a county records a style or model code, match it.
What does not make a comparable
- Similar sale price. You are comparing assessments, not sales.
- Similar tax bill. Different districts, different rates. Meaningless.
- "It looks like my house." The model does not see the street.
- A property that won an appeal on unusual facts. Its assessment reflects a condition issue or a vacancy, not the general level.
The disqualifiers to check before including one
Pull the record for each candidate, not just its assessment. A neighbour assessed conspicuously low may be low for a reason that does not apply to you:
- Recorded floor area smaller than yours — in which case there is no disparity to argue.
- A recent sale at a low price the office has already used.
- A condition or vacancy adjustment.
- An exemption reflected in the figure you are reading — make sure you are comparing assessed values before exemptions, not taxable values after.
That last one catches people constantly. Comparing your pre-exemption assessment against a neighbour's post-exemption taxable value produces a gap that does not exist.
The test that decides it
Before including a comparable, ask: if the assessor's analyst wanted to throw this one out, what would they say?
If you have an answer — it is 40% larger, it is in a different neighbourhood, it sold last year at a distressed price — leave it out. Four comparables you can each defend in one sentence beat a dozen that invite argument.
Presenting them
One table. PIN, address, class, floor area, improvement assessed value, and the per-square-foot figure calculated for each. Your own property in the same row format, and the median of the group as the level you are requesting.
The person reading it has a very large stack of files. A table they can verify in two minutes is one they can act on without further work.