FoxyAI Significantly Improves Fello's AVM

The FoxyAI condition score enhanced a large regional iBuyer’s AVM, allowing for better Real Estate investment decisioning.

5%

Improvement in PPE10 (Percent Predicted Error within 10%)

~$3k

Increase in individual property values

~$57mm

Improvement in portfolio value

2%

Improvement in Median Absolute Percentage Error (MdAPE)

All Case Studies
fello logo white

fello is a real estate company that simplifies the rental process through an online platform connecting tenants with properties and providing a seamless user experience.

Real Estate Investment

Condition-Enhanced Automated Valuation Model

FoxyAI Computer Vision Models

Download Now

Executive Summary

In the original pilot using 19,105 properties and 400,000 photos, Fello-a large regional iBuyer-found that the FoxyAI-enhanced AVM “significantly improved” their original model, enabling them to offer a better, more accurate product during a time of rapidly changing valuations. In addition, Fello increased their brand footprint and industry recognition while adopting the latest in visual intelligence technology.

The Company

Fello is a large regional iBuyer (“Instant Buyer”)—a company that uses algorithms and technology to quickly buy and resell homes.

The FoxyAI score significantly improved our Automated Valuation Model with a 5% improvement in PPE10 offering us an edge over more traditional models allowing us the ability to make better investment decisions.

Head of Data Science,

at Fello

The Challenge

In 2021, Fello—after previously buying residential homes entirely online— decided to expand their offering to brokers and agents, providing them with the tools to offer their own iBuying solution.

In creating this new offering, Fello’s leadership identified the key components necessary to execute and become more profitable. One essential component was to source the technology to create a hyper-sophisticated AVM to give its team and new B2B audience the confidence to acquire properties instantly in a low-interest-rate environment.

Consistently improving a successful iBuyer’s proven AVM is a serious challenge. The iBuyer must have created an accurate model to become successful, and to improve upon it takes the latest in computer vision advancements and a discerning data scientist.

Fello faced an added challenge of entering the saturated B2B market. Now, in addition to improving a demonstrated AVM, Fello’s team had to attract an astute agent and broker audience and needed to find the right technology partner to help them do so.

The Solution

After vetting technology partners, Fello chose to run a pilot AVM test with FoxyAI’s models. Fello’s data science team incorporated the FoxyAI Condition Score Model into their current AVM to see if it would improve accuracy.

Fello found that FoxyAI’s models:

Accurately Analyzed Property Media

The FoxyAI Condition Score Model analyzes a company or government entity’s property media and provides a continuous condition score, using a 6-point scale ranging from Brand New to Heavy Damage/Not Livable. It is based on the scoring system from the Uniform Appraisal Dataset used by Fannie Mae, Freddie Mac, and others for underwriting.

Retrained & Recomputed AVM Models

By retraining the iBuyer’s AVM with the FoxyAI Enhanced AVM, the iBuyer was able to “significantly improve” their original model, reporting a “5% improvement in PPE10 and a 2% improvement in MdAPE.”

Improved Property Value

In dollar terms, this AVM enhancement on average brought each of these 19,105 properties ~$3,000 ($2965.52) closer to the sales price, accounting for a ~$57,000,000 improvement in property value.

Fello rolled out their AVM, which was enhanced with FoxyAI’s Condition Score. Their team successfully improved their demonstrated AVM in order to expand their product offering to a new market. In addition, Fello increased their brand footprint and industry recognition while adopting the latest in visual intelligence technology.