.svg)
On April 22nd of 2026, Apartments.com made a major update to how their site calculates property ratings. Previously, scores were heavily influenced by a pre-assigned initial CoStar rating. (worth noting: CoStar Group owns Apartments.com).
Before any resident reviews were even taken into consideration, CoStar first assigned a building-quality score. Resident reviews were then added on top, and the displayed “average” rating blended both.
CoStar ratings were based speculatively on finishes, amenities, architecture, landscaping, and construction quality for example. This framework benefited new luxury properties – they could open with 4.5+ stars without any resident reviews.
Further, when reviews did start coming in, they were slow to move a property’s score. When users reverse-engineered the score some years ago, it was found that the CoStar rating could have as much weight as ~50 resident reviews. This created very noticeable discrepancies between score and actual review content.
Many negative reviews were slow to move a property’s ratings if it had a strong initial CoStar Building Rating. The opposite was also true – positive reviews would not move the needle much on properties with low Building Ratings.
Now, as a consequence of client feedback, Apartments.com has updated their scoring model to reflect renter ratings only – removing the CoStar building rating component. J Turner believes this to be a highly commendable change and even updated our ORA Score algorithm to account for it.
Here’s the score distribution before and after the change:
Pre-update, the average score was 4.18 stars. Now, the average sits at 3.43 stars, a significant difference of -0.75. This change is observed across 65,820 properties listed on Apartments.com with at least one review, with 880,859 total reviews taken into consideration.
Let’s further break down the amount of properties that changed and the severity of those changes:
This change supports the hypothesis that the building-quality scores created (more often positive rather than negative) inertia on a property’s ratings. This behavior was different from other consumer-facing review sites, e.g. Google, Yelp, etc. The new model post-April 22nd, 2026 intends to improve transparency and to better align with user expectations.
Previous models of ORA and its interpretation of Apartments.com ratings have evolved over time. In August of 2025, Apartments.com switched to a pay-to-play structure wherein reviews from non-paying clients were removed entirely. This change skewed scores to be higher, resulting in J Turner building the ORA Score model to have significant bias against it.
Need a refresher on how ORA® is calculated? Learn more about the ORA® Score
The April 22nd update removes the blended rating construct and instead focuses entirely on sentiment. This is much more aligned with J Turner’s philosophy towards resident experience, allowing us to construct our ORA Score model to take Apartments.com ratings at greater face value. The newest update will be reflected in the May 2026 ORA Score.
Projected ORA Score Change for Properties with 10+ Reviews on Apartments.com
| Apts.Com Rating Change | Avg ORA Change |
| Down -3.0 or More | -3.72 |
| Down -2.0 to -2.9 | -2.20 |
| Down -1.0 to -1.9 | -1.20 |
| Down -0.5 to -0.9 | -0.36 |
| Down -0.1 to -0.4 | 0.26 |
| Stayed the Same | 0.92 |
| Up 0.1 to 0.4 | 1.29 |
| Up 0.5 to 0.9 | 2.49 |
| Up 1.0 or More | 3.37 |
| Average Change | 0.15 |
The scale of Apartments.com ratings shifted more negatively and the site now has substantially less bias, which is reflected in the way the new ORA model analyzes properties' data. To provide a reasonable projection for your ORA Score as a result of this change, J Turner Research compared properties' ORA Scores on the previous ORA Model using the old Apartments.com methodology and the new ORA Model using the new Apartments.com methodology. The analysis was limited to properties with at least ten reviews on Apartments.com to remove as many external factors as possible. With any change of this kind, the goal is keep ORA as similar as possible, wholistically. If your property saw less of a drop off than the national average (-0.75), it is reasonable to expect this to have a neutral or positive effect, although all changes will be property dependent. The average property with ten or more reviews went up by 0.15.
If you’d like to know how you may have been impacted by the change, you can send us an email at ORA@jturnerresearch.com.