Case Study

Mynd Goals 

Mynd’s primary goal with utilizing DwellsyIQ was to bolster our rent underwriting capabilities. We felt the quality and scope of data provided by other vendors was not meeting our needs, so we looked at DwellsyIQ as a potential improvement in three key areas:

• Density of data within our active markets 
• Accuracy and recency of data records 
• Image access for properties
  
  

After review, we determined DwellsyIQ provided us an opportunity to implement a product we felt offered us a higher quality of data and unique listings that our other listing sources were not providing. 

“Across multiple key property features, DwellsyIQ possessed a high degree of accuracy”

Results and Outcomes 

To evaluate Dwellsy, we compared our results from DwellsyIQ to those we had from alternative listing sources in our Global Property Record (GPR). To validate data quality, we benchmarked Dwellsy’s listing information against our managed properties since we possess source of truth data on those properties. We found that, across multiple key property features, DwellsyIQ possessed a high degree of accuracy, particularly for square footage and bedrooms, full bathrooms, and half bathrooms. DwellsyIQ matched our first-party data on over 80% of these records. Square footage error was approximately normal with a mean difference of 0 and sharp fall-off. Most notably, DwellsyIQ matched records for our first-party data at a substantially higher rate than for our other third-party data. This suggested to us that DwellsyIQ data was likely more accurate than our existing third-party sources.

“DwellsyIQ data was likely more accurate than our existing third-party sources.”

In terms of listing volume, we found that DwellsyIQ did not typically provide as many new listings per week as some of our other third-party sources. We compared the new listings available weekly in DwellsyIQ compared to two other third-party vendors in MSAs shared across all three vendors. DwellsyIQ had an average of 12.1 new listings per week in our analysis period per MSA, compared to 89.9 and 63.2 for our two other vendors in the analysis. However, we believe the quality of those listings was higher: the average of a listing in our DwellsyIQ sample was 41.1 days, compared to 152.2 and 76.9 for respectively for the other two third-party vendors we compared. Similarly, DwellsyIQ also did not service as many MSAs as our other vendor sources (roughly 100 fewer MSAs had listings in the window), but they covered more unique MSAs than the other vendors in our comparison, who shared all but 10 MSAs. In comparison, DwellsyIQ provided data on 96 MSAs not found in either of the other two third-party datasets considered.

“We could effectively use DwellsyIQ to augment our current records with novel information we would otherwise not have had access to.”

We also found that in specific MSAs, DwellsyIQ provided us records on properties that were unique to the DwellsyIQ dataset. We sampled 5 distinct MSAs to evaluate Dwellsy property records, and found that up to approximately 12% of properties DwellsyIQ provided in some MSAs had no corresponding record in any of our other third-party sources. This gave us confidence we could effectively use DwellsyIQ to augment our current records with novel information we would otherwise not have had access to.

Contact us to learn more about what DwellsyIQ data can do for you.