A comparative market analysis (CMA) is the report a real estate agent builds to estimate what a property is really worth in today's market. You compare the home to similar properties that recently sold, that are currently for sale, and that failed to sell, then adjust for the differences to land on a defensible price range. Done well, a CMA wins listings, sets the right asking price and helps buyers bid with confidence. Done badly, it loses you the listing or leaves money on the table. This guide walks through the seven steps, with a worked example.
A CMA is a market estimate, not a certified valuation. It reflects an agent's read of supply, demand and recent transactions. An appraisal, by contrast, is an independent valuation produced by a licensed appraiser, usually for a lender deciding whether to fund a mortgage. The two often land close together, but they answer different questions. The CMA answers "what should we list at, and what is a smart offer?" The appraisal answers "is the lender's security sound?" Keep that distinction clear with clients so nobody treats your CMA as a guarantee.
Step 1: Define the subject property precisely
Everything keys off an accurate picture of the home you are pricing. Record the living area in square metres or feet, plot size, number of bedrooms and bathrooms, build year, energy efficiency, parking, outdoor space, and condition. Note anything unusual: a recent renovation, a non-standard layout, a busy road, a protected view. These are the attributes you will adjust comparable homes against, so the more precise you are here, the tighter your final range.
Step 2: Pull the right comparable sales
Comps are the backbone of the CMA. The strongest comparables are sold properties that are close on three axes: location (same neighbourhood or street where possible), type and size (similar living area, bedroom count and property type), and recency (ideally sold within the last three to six months). Aim for three to six solid sold comps. Then add a couple of active listings to see your current competition, and one or two expired or withdrawn listings to learn what the market rejected. If you work without an MLS, public sold-price registries, land registry records and portal listing histories give you the same raw material.
Step 3: Adjust each comp for differences
No two homes are identical, so you adjust each comp's sale price up or down to reflect how it differs from the subject. If a comp has an extra bathroom your subject lacks, you subtract the local value of that bathroom from the comp's price; if the subject has a renovated kitchen the comp didn't, you add it. The goal is to answer: "what would this comp have sold for if it were as similar to the subject as possible?" Be consistent and conservative. Over-adjusting is the fastest way to talk yourself into a number the market won't support.
Step 4: Build the adjusted price range
With each comp adjusted, you now have a cluster of equivalent values. Discard outliers you can't explain, weight the closest matches most heavily, and derive a range rather than a single number. A range is more honest and gives you negotiating room. Cross-check it against a price per square metre figure for the area as a sanity test. If your range and the per square metre benchmark disagree sharply, find out why before you present anything.
A worked example
Say you are pricing a three-bedroom house of 110 m². You find three sold comps:
Comp A sold for €430,000, also 110 m², but with a renovated kitchen the subject lacks. Adjust down €15,000 → €415,000.
Comp B sold for €395,000, 100 m², no garden. Add €12,000 for the extra 10 m² and €8,000 for the subject's garden → €415,000.
Comp C sold for €440,000, 120 m², similar condition. Adjust down €18,000 for the size difference → €422,000.
The adjusted values cluster at €415,000 to €422,000. You would list around €415,000 to attract competition, with a defensible ceiling near €422,000. That tight, evidence-backed range is what wins the listing conversation.
Step 5: Add the market context
Numbers alone don't tell the story. Layer in how fast homes are selling, the ratio of sale price to asking price in the area, inventory levels and the direction of recent price movement. In a hot market where homes sell over asking in days, you price to spark a bidding war. In a slow market, you price to sit just inside the range buyers are searching. The same comps support very different strategies depending on context.
Step 6: Present it as a clear, branded report
The analysis is only half the job. Clients buy your judgement, but they need to see the evidence. A clean, branded report with the subject, the comps, the adjustments, the range and the recommended strategy makes you look like the professional you are. This is where dedicated CMA software earns its keep: instead of formatting a spreadsheet for an hour, you generate a polished, client-ready document and spend your time on the conversation.
Biedradar is built for exactly this step. You enter an address and the platform pulls comparable sales, valuation and market signals, then produces an automated, branded property analysis report you can hand straight to a seller or buyer. It turns hours of manual comp-pulling and formatting into minutes, so the agent's time goes into advice, not admin.
Step 7: Refresh it before the conversation
A CMA is a snapshot. Markets move, new comps close and listings expire. Before any pricing meeting or offer review, refresh the data so you are never advising on a stale picture. If a major comp closed last week at a surprising number, you want to walk in already knowing it, not get blindsided by the other side's agent.
Common CMA mistakes to avoid
Cherry-picking comps that flatter the number you or the client want. The market will correct you.
Using stale comps. A sale from a year ago in a moving market is closer to fiction than fact.
Ignoring active and expired listings. Solds tell you where the market was; actives and expireds tell you where it is now.
Presenting one number instead of a range. Precision you can't defend reads as guesswork.
A strong CMA is part data, part judgement and part presentation. Get the comps and adjustments right, frame them in current market context, and deliver them in a report that earns trust. Biedradar handles the heavy lifting on the data and the report so you can focus on the part only you can do: advising the client. To see how the buyer side reads the same evidence, our guides on buying a house step by step and how much a buyer can borrow show how pricing connects to what offers a property can realistically attract.
Frequently asked questions
What is a comparative market analysis (CMA)?
A CMA is a report a real estate agent prepares to estimate a property's likely market value by comparing it to recently sold, active and expired listings that are similar in location, size, condition and features. Unlike a formal appraisal, it is not a legally certified valuation, but it is the everyday tool agents use to price listings and advise buyers on offers.
How is a CMA different from an appraisal?
An appraisal is a certified valuation produced by a licensed appraiser, often required by a lender before a mortgage is approved. A CMA is an agent's market estimate used for listing and offer strategy. The CMA is faster and free to the client; the appraisal is independent, regulated and carries legal weight.
How many comparable sales should a CMA use?
Most agents anchor a CMA on three to six strong comparable sales (comps), supported by a few active and expired listings for context. Quality matters more than quantity: three genuinely similar homes that sold in the last few months beat ten loose matches.
How long does it take to create a CMA?
Done by hand, a thorough CMA takes one to three hours: pulling comps, adjusting for differences, building the price range and formatting the report. CMA software and automated valuation data can cut that to minutes, leaving you time to focus on the narrative and the client conversation.
Can you create a CMA without an MLS?
Yes. In markets without a multiple listing service you build the same analysis from public sold-price registries, land registry data, portal listing history and automated valuation models. The method is identical; only the data source changes.
