The Assessor's Office is once again refreshing its open data, publishing the same datasets the Data Department uses to estimate residential property values, conduct sales ratio studies, and create maps.  These datasets are being shared to promote transparency and for use in analyses by taxpayers, academics, and the real estate industry.  Documentation for the datasets, their source code, and raw modeling input data can be found on the Data Department's wiki.

The Data Department is deprecating its older datasets, but not removing them. Data that will no longer be maintained and updated will be prefixed by "Archived [XX-XX-XXXX]". New datasets will be updated automatically, on either bi-weekly or monthly schedules in order to ensure ongoing support.

The 2022 to 2023 open data refresh has two major updates: disaggregating location data and adding appeals data. 

Location data, such as geography, address, proximity, and proration rates, were previously published together in the old Parcel Universe dataset. This made Parcel Universe large and difficult to work with.  In 2023, this data is separated into multiple datasets:

This update also adds a new Assessor Appeals dataset. This dataset is a rough draft and will be replaced with a finalized version once the Assessor's internal system transition is complete. Currently, appeals data is sparse pre-2021 and some columns are missing values for other random subsets of time. Please read the dataset's documentation thoroughly and make sure you are aware of its current limitations before using it.

While the data presented in this release is as accurate as the Data Department can reasonably make it, there are some cases where users should not expect the data to represent the real world perfectly. The Assessor's field team, which helps maintain residential characteristics, cannot enter homes and so must make educated guesses as to interior characteristics. Additionally, the team cannot be realistically staffed to visit every residential parcel in the county every year and depends heavily on home-improvement permits to trigger site visits. Permitting is at best an incomplete process - some home improvements are performed unpermitted, and some permits do not represent the work being done accurately. This process is improving as more digital tools become available to the office.

Feel free to contact the Data Department with any questions or suggestions about the data here – or if you'd like to share how you've used it!

A dataset containing the entire universe of parcels in Cook County dating back to 1999.  This data includes a wealth of spatial data, such as the census geographies and tax districts for a given parcel.  The main purpose of the current iteration of this dataset is to allow users to retrieve all parcels within geographic and temporal limits of their choosing. A typical use case for this spatial data might be to join it with property characteristics data (see below) for estimating its value, or to join it with sales to analyze sales trends within or across a number of possible geographies.  Spatial characteristic data, unlike physical parcel characteristics, is largely only available back to 2013, so the Data Department has filled these fields back to 1999 using the oldest available data.

Characteristics for Assessor regression-class improvements used in the Automated Valuation Model (AVM). This dataset, unlike the rest, is improvement-level rather than parcel-level; 'improvements' is assessment parlance for buildings.  A single parcel with multiple buildings in 2021 – think a coach house in the back of a single-family home – will have two rows for 2021, containing the physical characteristics for both the main house and the coach house. Rows are thus unique by PIN, year, and card (improvement).

Characteristics for condominium units used in the condominium Automated Valuation Model (AVM). Individual condo units are associated with a 14-digit parcel identification number (PIN), and share the first 10 digits of this PIN within a building.  The Assessor’s Office currently has a sparse (but growing) set of physical characteristics for condos. The characteristics in this dataset primarily relate to the entire condo building or are indicators as to whether a unit should be valued as a living or parking space.

The set of property sales available to the Assessor’s Office, populated by the Illinois Department of Revenue (IDOR). The dataset includes an indicator for sales that included more than one parcel and has been cleaned to avoid duplicate sales. Joining this dataset to characteristics is how the Data Department trains its AVMs. Joining it to the parcel universe allows for convenient spatial analysis and mapping of sales, and it can be joined to assessed values for ratio studies. 

Assessed values of the land, building, and total parcel for every parcel in Cook County dating back to 1999 for each step of the assessment process: the initial values the Assessor's Office mails to taxpayers, values after appeals to the Assessor's Office, and final values after appeals to the Board of Review.  Assessed value is not market value - assessed values must be adjusted by their level of assessment (e.g. 10% for residential parcels) to arrive at market value. The Data Department cannot currently guarantee that past and current parcel values align perfectly with what is presented on the Assessor's website. The data used for that purpose is not available to the Department in its final form. We hope to be able to confirm the accuracy of what we're making available here against what's available on the website soon.

Land, building, and total assessed values, pre and post-appeal with the Cook County Assessor’s office, for all Cook County parcels, from 1999 to present. Assessed value is not market value - assessed values must be adjusted by their level of assessment (e.g. 10% for residential parcels) to arrive at market value. Taxpayers are allowed to appeal their assessment directly to the Assessor's Office, and the outcomes of those appeals determine the Assessor's certified assessments. This data set can be used to investigate trends in appeals over time, across geographies (by joining to the parcel universe), or by property class. Please note the preliminary state of this dataset when working with it.

Mailing and situs addresses for every parcel in Cook County dating back to 1999. The addresses in this dataset are used by the Assessor's office to mail assessment notices. This data can be be used as a crosswalk if users have address-level data they wish to join to parcels, or for geocoding if parcel centroids are insufficient.

Parcels and their proximity to various spatial features, such as train stops and Lake Michigan. This data can be used to filter PINs based on their distance to a feature of interest. Unlike other parcel-level datasets, proximity will not be updated on a monthly or bi-weekly schedule, but only annually as new parcel shapefiles are made available.

An aggregated version of the Assessor's commercial valuation spreadsheets. While this data should be more consistent in the future, historically column values have changed across time, townships, and classes. This dataset is an attempt to consolidate as much information as possible from the office's commercial valuation excel workbooks into one dataset. It contains data used to value commercial parcels such as revenue estimates, rents,  number of income producing units such as apartments and hotel rooms, vacancy, and business types.

Parcels with property tax-exempt status . Properties of religious, charitable, and educational organizations, as well as units of federal, state and local governments, can be eligible for exemption from property taxes.

A shapefile containing the neighborhoods the Assessor's Office uses for thematic mapping and reporting. Neighborhood codes are just one of many location factors that the Office’s AVMs can use to estimate property values.