One of the main responsibilities of a listing agent is advising a client on the value of a home. Oftentimes agents are even pitching their ability to price a home before they are hired during listing presentations. Agents use multiple tools to assist in this process and answer:
How do I generate an accurate price assessment for a client?
How do I show clients similar homes that sold in their market?
How do I convince a client to work with me by showing I can maximize the sales price?
A CMA provides context for the seller on how to price the home. For buyers, it shows what they should expect to pay for features and locations. There is also an investor application. An agent can show the appreciation of a property if it’s purchased and improvements are made.
The Comparative Market Analysis or CMA
What is a Comparative Market Analysis
A Comparative Market Analysis is often called a CMA. An agent finds recently sold properties that are similar to the client’s home. Usually these properties are nearby homes with similar features. Comparison homes are referred to as comps. This data is compiled and analyzed to present an estimate for the selling price of the client’s home. The agent may also put together additional CMAs. These extra CMAs can be based on active listings, homes with upgrades, or other components. Pricing a home is both art than science. One agent may say that larger square footage will yield a higher sale price. Another may say that an existing modern kitchen will be more important.
CMA software and features
The majority of the CMA process is research and presentation. Agents search sold and active listings for relevant information. They then compile results into an attractive package. CMA software does three things: automate research, update results, and present data.
Automate research: CMA software often suggests similar listings. This is based on the same criteria an agent would consider (sold date, location, features). Agents use this starting list to create a CMA quickly. This replaces multiple manual searches. Homesnap’s Rapid CMA is an example of this service.
Update results: without technology, an agent can run a CMA and present it with stale data. Maybe two properties sold for $30,000 more than they were listed for just days ago. CMA software often automatically updates listing information and notifies the agent. For example, MoxiPresent provides live data from the MLS. This ensures pricing and listing information is up to date.
Presenting data: MLSs tend to present search results in an unattractive way. Creating beautiful presentations is often a core feature for CMA software. Tools such as Cloud CMA create printable CMA presentations from MLS listings. It then allows agents to present data on iPads and websites.
Popular CMA software
There are three different types of CMA technology.
Built-in MLS functionality: many MLSs include a CMA tool. These services typically allow agents to export search results to a CMA presentation for print or email. But the interface from the MLS can be clunky and the output if often unattractive.
Standalone CMA software like MoxiPresent, Cloud CMA, and DashCMA: several CMA software options have emerged specifically to make researching and presenting CMA data better. Often, these software providers partner directly with the MLS. Other times, the software is paid for by a brokerage or agent. Either way, these products typically rely on MLS data to update property listing information. The output is usually faster and more attractive than built-in MLS options.
Integrated CMA software like Remine and Homesnap: Many search and collaboration tools include CMA software. For example, Remine has a full CMA report tool. Homesnap has a lightweight CMA option that allows agents to create mobile CMAs on the go. The feature sets on these types of tools might be less robust than the standalone options.
Some tech-enabled brokerages rely on CMA software instead of building it themselves. For example, Side, a tech-enabled real estate brokerage, offers MoxiPresent to its agents.
Spreadsheets are a competitor to CMA software
Oftentimes a CMA is not that complicated. An experienced professional using Microsoft Excel or Google Sheets can assemble a compelling CMA. Spreadsheets offer more customization than many CMA tools. But they are less presentation-ready and won't update automatically.
Sample CMA presentations
To get a sense of what a CMA looks like, below are links to four examples.
Virtually every real estate portal offers valuations via automated valuation models
Zillow has been a driving force for consumer-facing automated valuation models or AVMs. An AVM simply uses a formula to assess data on and related to a subject home to estimate the value of the property. When Zillow launched its Zestimate AVM in 2006, consumers could now see valuation data previously available only through agents. In the past fifteen years, the quality of these estimates have improved. Redfin and Trullia both have consumer-facing AVMs. Realtor.com launched its AVM in 2020. Companies like CoreLogic, Attom Data Solutions, and HouseCanary also offer AVM services on-demand for companies requiring many AVMs.
Agents often complain that AVMs are not very accurate. Redfin argues that its estimates are within 1.73% of the sales price half the time. Zillow claims that its AVM is within 2% of the selling price of homes half the time. Ultimately, this is a nuanced argument. To provide some tangible numbers for this, I’ll use a $500K home example. That means to be within 1% accuracy, the home sells for $495-505K. For 2% it’s $490-510K, 5% it’s $475-525K, 10% it’s $450-550K, 20% it’s $400-600K. AVMs get better not only because of improved algorithms but also from three other factors:
AVM accuracy claims are protected by median transaction data: as flagged above, the 2% price range for a $500K AVM is $20K. That is a large dollar range. Saying that half of the homes sell for $10K below or above that of recent homes sold may not be that bold of a claim. And to a homeowner $10K matters a lot. But the data will also be “within 2% accuracy 50% of the time” if 50 out of 100 homes in a neighborhood sell for $495K. The other 50 could sell for $800K with no change to that claim.
Agents do the heavy lifting with on-market homes: While AVMs are data-driven, much of this data is generated by agents. Essentially, most for-sale homes are represented by agents. When those homes are listed, an agent is performing a CMA to find the best price to list it at. AVMs will then use this data to estimate the sales price of a nearby home. This creates an interesting cycle. AVMs in use the collective research of agents who listed homes for sale.
AVMs influence sales prices: Agents and sellers may use the AVM price of a homeas a data point to set the sales price of their home. This becomes self-fulfilling. A Zestimate may actually raise or decrease the ultimate listing price. This is because an agent or client builds the Zestimate into their own best judgement for a price knowing prospective buyers will be influenced by the Zestimate.
To be clear, the technology to generate AVMs is incredibly sophisticated. Getting to the accuracythey have is an amazing feat. But they should also be taken with a grain of salt
Many iBuyers present instant offers to home sellers based on internal AVMs
iBuyers are companies that buy homes directly from sellers. Virtually all iBuyers ask sellers to enter their home address to receive an offer. Once a seller start this process, the iBuyer will ask some questions and then generate a cash offer. These offers are based on internal AVMs. Essentially the iBuyer has to understand what they think a home will sell for in order to make an offer. They don’t have an incentive to provide the best estimate. Instead, they need to provide an attractive estimate. But one that still means the iBuyer makes money after service fees and selling the home for more than the purchased price.
How agents differentiate themselves from real estate portal and iBuyer AVMs
Agents have a few advantages of portal AVMs and iBuyers. Most notably, their incentives are aligned with the seller in a way that portals and iBuyers are not. Major portals simply want to generate more traffic. They don’t lose money by generating inaccurate AVMs. Conversely, iBuyers want to pay the least possible for a home. This way they can sell the home for the greatest profit over the offer price. Agents, on the other hand, get paid more when a home sells for more. They also don’t want to price a home so high that it doesn’t sell. These all contribute to how agents differentiate themselves.
Local market knowledge: AVMs can look at recently sold homes and active data. But it’s very hard to account for a new employer driving up demand or a tax law that will make a property more attractive. Agents with local market specialty can offer these types of insights.
Impact of marketing: an AVM can’t account for agent expertise on marketing. If an agent knows how to market a property well, they might be able to sell it for more than an AVM would indicate.
Value of improvements: Listing agents often help prepare a home for sale. An agent is able to influence the value of a home by encouraging the owner to spend more on improvements. This means that an AVM may be based on the home having an outdated kitchen. But the agent may specifically push for a remodel. This effectively improves the home value but makes the AVM wrong.
Aligned incentives: Agents don’t all hate AVMs and iBuyers. In fact, services like Cloud CMA now offer agents to showcase automated cash offers from investors and iBuyers. Many agents like that iBuyers will price homes below the agent's valuation. Agents can use this to show an example of an actual cash offer today that is lower than a real estate portal's AVM. This reinforces how agent incentives are better aligned. Agents want to get sellers the best price that is realistic. The iBuyer may give a low but acceptable bid to get a house for cheap. The portals may be unrealistically high or not customized simply to drive traffic.