Data and analytics

Beyond drive-by appraisals: Intelligent home valuations


There are two important measures lenders use to decide whether or not a homeowner can qualify for a home equity loan, second mortgage or line of credit. First, they consider the homeowner’s current financial picture. Second, they schedule an appraisal to determine if the home provides enough equity for the loan.

Lenders can quickly and efficiently evaluate a homeowner’s creditworthiness. However, there are still many opportunities to improve upon drive-by property appraisals, which can be used for non-GSE loans under $400,000, such as home equity loans, lines of credit and refinances.

The most notable drawback of drive-by valuations is they can make assumptions about the interior of a home, which can result in increased risk or reduced opportunity for both lenders and homeowners. For instance, a drive-by appraisal or evaluation may overvalue a home where the interior has signs of wear and tear beyond what would be expected for the home’s age. On the other hand, a home that has been meticulously maintained and upgraded by its owners can be undervalued. In either case, an exterior-only property condition report doesn’t always provide enough insight and can lead to a poor customer experience

New valuation technologies

Technology is having a transformative impact on valuations – powering solutions that don’t require an appraiser or data collector to enter the interior of a property or even visit the property at all. These new technologies can help expedite the valuation process, reduce origination costs, minimize risks and improve valuation accuracy.

Solutions are now available that can assess a property at scale using uniform appraisal data (UAD) standards. These tools can quickly and objectively identify the overall condition and quality of a property that are not always recognized in drive-by valuations.

Intelligent computer vision technology or smart image recognition are used to build these type of valuation solutions. With computer vision, the solution is trained to extract information from visual sources, and then take actions based on the information. This technology is already being used throughout the entire valuation process, including for data collection, report-building and quality control.

In the data collection process, computer vision capabilities proactively and automatically label and validate photos and extract property features from those photos. The technology can also identify damage or repairs to make sure they were addressed in an appraisal or valuation report.

For appraisal or valuation report creation, computer vision can identify a subject property’s design style, condition and quality to accelerate the comparables-selection process. Instead of having to sift through photos from a pool of comparable sales, computer vision automatically identifies features that can be sorted and filtered to significantly speed up the process, while maintaining objectivity and credibility in the analysis.

And in the QA process, computer vision double checks the property’s quality and condition, validates all the right photos were taken, assists in the reconciliation of relevant features and more


Validate, a new solution offered by ICE Mortgage Technology, leverages computer vision technology, along with other intelligent data and analytics features, to help create transparent, credible, objective and reliable valuations for many lending use cases, including home equity loans, lines of credit and refinances.

The borrower can access Validate through an email or text link sent by the lender or directly in the loan application. Following a series of simple prompts, the borrower takes specific photos of their home in a secure and controlled process, where only the camera on their device can collect the data. The collected data is tracked and recorded using the location services on the borrower’s mobile device. This entire process takes less time to complete than the borrower locating and uploading their financial documents for their loan application.

After assessing the property’s condition and quality, Validate compares it with similar homes to create a condition-adjusted value estimate. Home equity is calculated by deducting any outstanding mortgage liens, as identified in ICE’s public records data. In addition to submitting real-time property photos, the homeowner completes a short questionnaire that verifies public record data.

The homeowner is also able to report on any property-value-enhancing amenities and necessary repairs. Lenders receive results through an integrated API or a concise PDF report. The lender can choose to use Validate’s automated valuation model (AVM) as a standalone valuation or pass the data collection to their traditional valuation-services provider for additional analysis.

Removing subjectivity

Ten different appraisers could give a property 10 different valuations. By leveraging technology to do the job, we’re improving data consistency while helping to provide a credible, objective and reliable property value. And what currently takes days or longer using traditional methods, Validate can deliver in just minutes.

To learn more, watch our complimentary on-demand webinar.

John Holbrook, Vice President, Digital Valuation Solutions at ICE Fixed Income & Data Services has more than 28 years of experience in collateral risk and valuation, having held various roles as an appraiser, USPAP instructor and strategic roles at LPS, Fannie Mae, Collateral Analytics and Black Knight.

Watch webinar

Read more from The Connection Point

Data and analytics  

How Waterstone Mortgage became a mortgage data champion

Automation & technology  

Unlock scalability and innovation with mortgage automation

Data and analytics  

Unlocking the power of mortgage data: Proven use cases from industry leaders