How does the Automated Valuation Model work in the real estate industry?

If you were to invest in real estate, what would be the most important factor you’d take into account? Would it be the age of the building, its location, or maybe how many owners it previously had?

While all of the above will certainly be important to you (though to varying extents), there’s one universal factor that will either be a deal-breaker or make you want to go all in.

You’ve guessed it — the price.

We all have a rough overview of what influences a property’s market value, right? Now, what about how it’s actually calculated?

In this day and age, does each property valuation still require the “human factor” — so, an appraiser who inspects the property in real life?

In the following post, we’ll take a look at what automation valuation model is, how it’s calculated, and which industries apply AVMs in their daily work.

What is an Automated Valuation Model?

In the simplest terms, an  (also referred to as ) is a term used to describe a service that leverages a mathematical model to provide a real estate property value. It minimizes the need to personally inspect and scrutinize each and every property on the market — the way it was in the (not so distant) olden days.


The real estate industry is a huge collection of data. AVM allows to make decisions faster and gives a better assessment of the situation.

AVM leverages all types of data, such as property age and condition (which we discuss further in this post), and generates a report within seconds.

Now, let’s take a closer look at what AVMs take into account and what they can and cannot do.

Types of Automated Valuation Models

There are roughly 4 types of AVMs currently used on the market. These are called: hedonic, tax assessor, price index, and combined valuation.

For the sake of this article, let’s discuss the latter (combined model) as it’s most comprehensive and incorporates all of the models above and more.

Complex valuation — how it’s calculated

When a  is employed to provide a property’s market price, the software inspects:

  1. Recent deed transactions (how many times a property has been sold)
  2. The most recent price paid for the property (so, for instance, what the current owner paid for acquiring the real estate vs how much the property could have gained or lost given all the factors that arose since the purchase)
  3. Prices of similar properties (the hedonic model, i.e. a comparative analysis of similar properties sold)
  4. Any previous valuations — most prominently, a tax assessor indication of value, if available.

As you can see, an Automated Valuation Model of this type provides a holistic overview of all objective data that can be processed by software.

Now, what about the potential limitations or discrepancies of such reports?

Automated Valuation Model — limitations

As you’ve likely guessed it, despite all the technological advancements in recent decades, there’s still a fair share of what software or AI cannot do. An AVM can’t take into account what would oftentimes be apparent to a human valuator.

For instance, an AVM will likely price a property highly or poorly based on its location and previous ownership, but it won’t in itself know about any potential nuances such as poor quality of materials used for interior design.

In short, what looks good on paper (or screen) won’t necessarily be the same to the human eye.

Now, understanding aesthetics isn’t the only area where an AVM might fail to deliver.

An automatic report might be poor to reflect the price of the property if it can’t perform a comparative analysis — that is, if a property had never been sold before.

Equally so, an AVM won’t be able to reflect an accurate price if a property is unique and nothing similar had been found in records.

That’s where a professional valuator will still need to pay the property a visit or do his/her own investigation to provide a custom pricing.

Now, with all of the above in mind, let’s discuss which industries and professions can benefit from AVMs the most and why.

Investment professionals

One extremely prominent group are investment companies or investment advisors, who use AVMs at scale. Primarily, they apply AVM reports to quickly evaluate whether a property available on the market is a bargain or simply a good investment, given other factors like ROI on similar estates.

Investment advisors, on the other hand, can use AVM to generate a list of properties their clients might be interested in. Without the need to engage a real estate expert at this stage, they can offer a much bigger list of investment venues than if each had to be personally inspected.

Now, as you’ve likely guessed, AVMs are equally attractive for those who sell properties for a living.

Real estate agencies/realtors

When you turn to a real estate agency to help you find your dream home, what do you expect to receive? A decent list of potential places, naturally!

In order to stay competitive, real estate brokers need a way to quickly valuate places that are up for sale. While they might still employ a valuator later in the process, this minimizes the initial effort prior to adding a place to a listing. Not to mention, it gives those interested in selling a place via a real estate agency a genuine valuation of how much they can realistically expect to sell for.

Next group?

Lending institutions

Similarly to real estate agencies, these institutions need AVM calculations for when someone wants to take a mortgage against a property of their own. This is key to effective risk management — lending institutions such as banks need a way to quickly verify if their potential customers have the means to back up their loan.

Last, but not least…


While AVMs partially take away traditional tasks from valuators, nowadays they can also make their work easier and more effective. As mentioned previously, appraisers are still engaged by lending institutions, investors, and property owners to provide a realistic price that incorporates what’s hidden from even the most advanced technology.

Final thoughts

AVMs are an immensely valuable technology that transitioned more industries than just those directly related to dealing real estate. Given the time savings and cost optimization it has brought to property-related markets, it is to be expected that AVMs will remain indispensable for many years to come.


Views: 578

Tags: AI, AVM, automated, big, data, deep, dsc_fintech, dsc_tagged, estate, learning, More…machine, model, real, valuation


You need to be a member of Data Science Central to add comments!

Join Data Science Central

© 2021   TechTarget, Inc.   Powered by

Badges  |  Report an Issue  |  Privacy Policy  |  Terms of Service