Automated Valuation Model

Insightful data to enhance your valuation needs.

Manage opportunity and risk across the lending lifecycle

The Teranet Automated Valuation Model (AVM) produces current market estimates for residential properties across Canada using advanced mathematical models and Teranet’s national property information database.

Our AVM is an invaluable tool to assist financial institutions to manage opportunity and risk across the entire spectrum of the lending lifecycle: from prospecting and acquisition, to funding, to portfolio management, and finally to client retention and growth.

The Teranet Advantage

Improve Customer Acquisition

Gain deeper insights into target markets through real time valuation of customer’s real property assets. Immediate loan-to-value determination for faster approvals or valuation service escalation, enabling lower costs to acquire and faster turn-around-times.

Smarter Valuations

Utilize AVM range to quickly determine listing prices, generate qualified leads and improve customer acquisition strategies.

Optimize Portfolio Management

Ongoing assessment and reporting of overall property values in a portfolio. Prioritize and intelligently manage recoveries knowing consumers real estate ownership and equity position.

Optimize Risk Mitigation

Estimate the potential loan-to-value as part of your collateral risk assessment. Access additional dimensions of property value, risk profiles and equity position to determine viability of loan.

The Teranet AVM Methodology

The Teranet Repeat Sales Model estimates residential resale property values based on property sales pairs. This module uses all properties in the neighborhood of interest that have sold at least twice in the past and thus establish a trend for the rate-of-change of similar and proximal property values over time.

The model then uses that rate-of-change to provide an estimation of the current market value of a specified residential property and a statistically reliable price range, based on the sales activity of similar properties in the immediate neighbourhoods.

Teranet AVM produces its valuations from a proprietary national database compiled with data from public, commercial and proprietary data sources across Canada. Our national database is constantly updated with current sales and property information, which ensures that our AVM consistently reflects current market conditions.

Teranet AVM leverages a cascade of two proprietary statistics models to provide the best valuation of properties based on available neighbourhood information.

The Teranet Machine Learning model is a boosting-based machine learning model. Boosting is a machine learning ensemble meta-algorithm used primarily to reduce bias and variance in supervised learning algorithms. This model leverages a rich dataset including land registry data (i.e. title data, sales data and parcel data), property data (i.e. tax assessment data and structural data), demographic data and neighbourhood specific data, and other derived data from internal data processing.

A machine learning model is able to learn from these transactions and the contribution of each attribute on a property valuation in a short period of time – a feat not easily achieved by human statisticians.

Teranet Home Value Range

Home Value Range leverages the Teranet AVM and provides real time valuation ranges on Canadian residential properties. It enables lending institutions to generate qualified leads, enable digital strategies and improve customer acquisition strategies.

Home Value Range offers real estate sales professionals a lead generation tool that encourages prospective home sellers to reach out once they have sourced a home value estimation for their property.

Via an online application portal or company’s website, the tool provides property owners with insight into the current market value of their home. It also offers prospective customers key insights regarding their property of interest and neighbourhood values.

Performance Metrics

Teranet reports AVM accuracy by comparing the actual sale price against the AVM prediction immediately prior to each transaction. Detailed performance reports are available upon request.