# Sales Comparison Approach (CMA)

> Component slug: `valuation-sales-comparison`. Researched: `2026-05-18T02:00:00Z`. Sources cited: `12`. Token usage: `~38000`.

## 1. Summary (≤3 sentences)

The Sales Comparison Approach (SCA) — known in residential practice as a Comparative Market Analysis (CMA) — estimates market value by comparing the subject property with recent arm's-length sales of sufficiently similar properties, then systematically adjusting each comparable's price for differences via the IVS/API adjustment hierarchy: property rights → financing terms → conditions of sale → market conditions (time) → location → physical characteristics [ref: #1][ref: #4]. In Australia, the approach falls under the IVS "market approach" and the API "comparable transactions method," which are the primary methods for owner-occupier residential, vacant land, and standardised commercial property; the API and IVSC mandate professional reconciliation (not simple averaging) of adjusted comp prices into a single opinion of value [ref: #3][ref: #4]. Automated Valuation Models (AVMs) and hedonic regression models serve as statistical adjuncts to the SCA, using large datasets to estimate quality-adjusted price indexes and provide confidence intervals around traditional appraiser-derived estimates [ref: #5][ref: #7].

---

## 2. Key Concepts

- **Sales Comparison Approach (SCA) / Market Approach** [ref: #1] — Valuation method that estimates market value from observable sales of similar properties by systematic comparison and adjustment. Called "comparable transactions method" in IVS [ref: #4].
- **Comparative Market Analysis (CMA)** [ref: #2] — The residential and agent-level application of SCA; in Australia, only a licensed real estate agent or registered valuer may prepare a CMA [ref: #9].
- **Comparable / Comp / Subject Property** [ref: #1] — A recently sold property used as a benchmark for comparison; the subject property is the one being valued.
- **Principle of Substitution** [ref: #1] — A rational buyer will not pay more than the cost of acquiring an equally desirable substitute property; the SCA's theoretical basis.
- **Arm's-Length Transaction** [ref: #4] — A sale between unrelated, uncoerced parties acting in their own interests; the required transaction type for a valid comparable.
- **IVS/API Adjustment Hierarchy** [ref: #1][ref: #3][ref: #4] — The canonical sequence for applying adjustments to comparable prices: (1) property rights conveyed, (2) financing terms/cash equivalency, (3) conditions of sale, (4) market conditions/time, (5) location, (6) physical characteristics.
- **Transactional Adjustments** [ref: #1] — Adjustments applied before property adjustments; cover items 1–3 of the hierarchy (rights, financing, conditions of sale); these convert the comp's recorded price to an equivalent cash, arm's-length, market-conditions price.
- **Property Adjustments** [ref: #1] — Adjustments for items 4–6 of the hierarchy (time, location, physical) applied after transactional adjustments.
- **Gross Adjustment / Net Adjustment** [ref: #11] — Gross adjustment = sum of absolute values of all adjustments (measures how different the comp is); net adjustment = algebraic sum (positive + negative). Rule of thumb: gross > 25% or net > 15% of sale price signals a weaker comp.
- **Reconciliation** [ref: #3][ref: #11] — The professional process of weighting and synthesising adjusted comp prices into a final opinion of value; not simple averaging; API mandates professional judgement supported by explanation.
- **Paired Sales Analysis** [ref: #1] — Technique for quantifying a single adjustment: find two sales differing in only one characteristic, isolate the value of that characteristic. Unreliable in thin markets.
- **Hedonic Regression** [ref: #7][ref: #8] — Econometric method decomposing sale price into implicit marginal values of individual property attributes; standard form is OLS in semi-log specification.
- **Repeat Sales Index (RSI)** [ref: #8] — Price index restricting data to properties sold twice or more; controls for quality mix by holding property constant; the Case-Shiller approach.
- **Automated Valuation Model (AVM)** [ref: #5] — Software system providing property value estimates using mathematical modelling (hedonic models, comparable-based algorithms, or hybrid). In Australia, deployed by CoreLogic/Cotality, PropTrack, and bank risk systems [ref: #5].

---

## 3. Methodology / Formulas / Process

### 3.1 Step-by-Step SCA Process

**Step 1 — Define the subject property**
Identify and document all value-influencing characteristics: property type, land size, GFA, bedrooms, bathrooms, parking, condition, age, and special features (pool, granny flat, corner block, views, road noise). [ref: #10]

**Step 2 — Compile the comparable sales database**
Selection criteria (AU/Vic context):
- **Recency**: Ideal ≤ 3 months; acceptable ≤ 6 months. Beyond 6 months a time adjustment is mandatory or newer comps must be found [ref: #10].
- **Proximity**: Within 1–2 km; same suburb is ideal. Location change requires an explicit location adjustment [ref: #10].
- **Property type**: Must match (house vs. house, unit vs. unit — never cross types) [ref: #10].
- **Configuration**: Same bedroom/bathroom count ±1; similar land size and GFA.
- **Sale type**: Arm's-length only; exclude mortgagee-in-possession, related-party, or estate sales without adjustment for non-market conditions [ref: #1].
- **Minimum count**: At minimum 3 comps; 4–5 preferred [ref: #6][ref: #9]. In Victoria, estate agents are legally required to identify 3 comparable sales when advertising a property [ref: #9].

**Step 3 — Apply the IVS/API adjustment hierarchy**
Adjustments are applied sequentially in order, because percentage adjustments applied later are calculated on an already-adjusted base price [ref: #4]:

| Order | Category | What it corrects |
|---|---|---|
| 1 | Property rights conveyed | Fee simple vs. leased fee interest; easements; partial interests [ref: #1][ref: #3] |
| 2 | Financing terms / cash equivalency | Below-market seller financing inflates nominal price; adjust downward to cash-equivalent value [ref: #1][ref: #4] |
| 3 | Conditions of sale | Forced/distressed sales, related-party sales; adjust to arm's-length equivalent [ref: #1][ref: #4] |
| 4 | Market conditions (time) | Adjust for price movements since date of comp sale; derived from index (e.g., CoreLogic monthly change, RPData quarterly) [ref: #10][ref: #1] |
| 5 | Location | Differences in infrastructure access, school zones, view, noise, flood/bushfire risk [ref: #2][ref: #10] |
| 6 | Physical characteristics | Age, condition, GFA, land size, bedrooms, bathrooms, parking, improvements, pool, etc. [ref: #1][ref: #6] |

For commercial properties, two additional categories may follow: (7) economic characteristics (NOI differences, lease terms, vacancy) and (8) non-realty components of value (business goodwill, FF&E). [ref: #1]

**Adjustment direction rule**: If the comparable has MORE of a feature than the subject, SUBTRACT from the comparable's price; if it has LESS, ADD to the comparable's price. The goal is to make the comparable look like the subject. [ref: #6]

**Adjustment quantification methods**:
- Paired sales analysis (direct isolation of one variable) [ref: #1]
- Statistical regression / market-derived coefficients [ref: #7]
- Cost approach for physical items (replacement cost minus depreciation)
- Appraiser/valuer professional judgement anchored in market data [ref: #3]

**Step 4 — Develop adjusted sale prices**
Adjusted Sale Price (comp_i) = Sale Price + Σ(adjustments)

Example (from PropertyMetrics) for a 24,000 sqft subject office property: [ref: #1]
- Comp 1 sold $2,675,000 5 months ago → time adjustment +1.5% → physical area adj → Adjusted: ~$2,760,000
- Comp 2 sold $4,200,000 8 months ago → time +2% → Adjusted: ~$4,290,000
- Comp 3 sold $1,950,000 18 months ago → time +4.5% → Adjusted: ~$2,030,000

**Step 5 — Reconcile to a value conclusion**
The API and IVS prohibit simple averaging [ref: #3]. Reconciliation principles:
- Weight comps by similarity (fewest gross adjustments = most weight) [ref: #11]
- Weight comps by recency and data reliability
- Identify the range of adjusted values; the opinion of value falls within or at the most-supported end of that range
- Explain in writing why each weight was assigned [ref: #3]
- For residential properties, the final value is often a range (e.g., $780,000–$810,000) with a point estimate drawn from the reconciled analysis

### 3.2 Victoria / AU Regulatory Requirements

Under the **Estate Agents Act 1980 (Vic)** and Consumer Affairs Victoria underquoting rules: [ref: #9]
- Licensed agents must document at least 3 recent comparable sales when providing a price estimate
- Advertised price ranges must be within a 10% spread
- The comparable evidence must be "matching" — recent, proximate, and genuinely similar to the subject
- Failure to comply exposes agents to underquoting penalties

### 3.3 Adjustment Sequence Matters (Percentage Adjustments)

If adjustments are dollar amounts, sequence does not affect the result. If any adjustments are percentage-based, apply in hierarchy order because each percentage applies to the already-adjusted base [ref: #4]. Best practice: apply dollar adjustments for finite features (extra bathroom, pool) and percentage adjustments for continuous variables (market time, location premium).

---

## 4. Tools & Libraries

| Name | Type | Link | Notes | [ref] |
|---|---|---|---|---|
| CoreLogic / Cotality AVM (AU) | Commercial AVM | https://www.corelogic.com.au | 200M+ property attribute records; AI/ML hybrid hedonic + comparables model; used by Australian banks | [#5] |
| RP Data (CoreLogic AU) | Data platform | https://www.corelogic.com.au/our-data | Backend for most Australian bank valuations; API access to comparable sales history | [#5] |
| Hedonic OLS regression | Statistical method | — | Standard form: ln(price) = β₀ + β₁(land_size) + β₂(bedrooms) + ... + ε; OLS in semi-log format most common [ref: #7][ref: #8] |
| Repeat Sales Index (Case-Shiller / RBA variant) | Statistical method | https://www.rba.gov.au/publications/rdp/2006/pdf/rdp2006-03.pdf | Restricts to repeat transactions to control quality mix; lacks single-sale data; used for time adjustment derivation [ref: #8] |
| Hybrid Hedonic-RSI model | Statistical method | — | Combines hedonic and repeat-sales to capture both quality adjustment and property-specific price path | [#8] |

---

## 5. Case Studies / Examples

### 5.1 Standard Residential CMA (AU Suburban Context)

A 4-bedroom, 2-bathroom, 600 sqm house in a Melbourne outer suburb is the subject. Three recent sales are identified:

- **Comp A**: 4br/2ba, 580 sqm, sold 4 months ago at $780,000. Minor location difference (main road vs. quiet street): –$15,000 adjustment. Physical (smaller land): +$4,000. **Adjusted: $769,000**. Low gross adjustment — highest weight.
- **Comp B**: 3br/2ba, 620 sqm, sold 2 months ago at $745,000. Physical (one fewer bedroom): +$20,000 estimate from paired sales. **Adjusted: $765,000**. Medium weight.
- **Comp C**: 4br/2ba, 550 sqm, sold 8 months ago at $760,000. Market conditions +3% (market risen 0.375%/month): +$22,800. Physical (smaller land): +$5,000. **Adjusted: $787,800**. Older sale, lower weight.

Reconciled range: $765,000–$787,800. Professional opinion: **$770,000–$780,000** (weighted toward most recent and most similar comps A and B).

### 5.2 Time Adjustment Using Market Index

If CoreLogic/RPData data shows the local suburb index rose 4.2% over the past 12 months (0.35%/month), a comparable sold 9 months ago at $850,000 would receive a time adjustment of: $850,000 × (1 + 0.035 × 9/12) = approximately $850,000 × 1.0262 = **$872,270** [ref: #10]. The 3.15% time adjustment converts the older price to a "today" equivalent before any property adjustments are applied.

### 5.3 Commercial Office (PropertyMetrics Example)

For a 24,000 sqft office building with steady 1%/quarter market growth [ref: #1]:
- Comp 1 (26,500 sqft, sold 5 months ago, $2,675,000): Time adj +1.67%; Physical (larger): –$75,000. Adjusted price per sqft: **~$97/sqft**.
- Comp 2 (46,200 sqft, $4,200,000, 8 months ago): Time adj +2.67%. Adjusted value per sqft: **~$91/sqft**.
- Comp 3 (22,300 sqft, $1,950,000, 18 months ago): Time adj +4.5%. Adjusted value per sqft: **~$90/sqft**.
- Reconciled range: $90–$97/sqft → Subject value: **~$2.16M–$2.33M**.

---

## 6. Pitfalls & Gotchas

- **Insufficient or stale comparables**: The SCA breaks down when there are fewer than 3 genuine comps or when all comps are older than 6 months. In thin or illiquid markets (rural, unique properties), appraisers should recommend a formal valuation rather than an agent CMA [ref: #6].
- **Subjective, unsupported adjustments**: Adjustments must be grounded in market evidence (paired sales, regression data, or market-published coefficients). Arbitrary "round number" adjustments (e.g., always $5,000/bedroom) undermine credibility and can constitute basis for dispute [ref: #1][ref: #11].
- **Ignoring transactional adjustments**: Many informal CMAs skip property rights and financing adjustments, inadvertently using inflated comp prices (e.g., a sale with seller-paid closing costs or below-market vendor financing). This leads to systematic overvaluation [ref: #1].
- **Sequence error with percentage adjustments**: Applying percentage-based location or market-conditions adjustments before dollar transactional adjustments inflates or deflates the base, compounding error through the hierarchy [ref: #4].
- **Paired sales analysis in thin markets**: In low-volume markets or for unique features, paired sales may yield wildly different implied values depending on which pair is used. Quoting a single paired-sales result as if it is definitive is a common error [ref: #1].
- **AVM overreliance**: AVM confidence scores drop sharply for atypical or thinly traded properties, new builds (lack of comp history), and large apartments (aspect premium invisible to ML models) [ref: #5]. AVMs are adjuncts, not substitutes, for professional valuation.
- **Victoria underquoting risk**: Using insufficiently comparable sales to justify a lower-end price guide violates Consumer Affairs Victoria underquoting rules. The 3-comp requirement must be met with genuine matches, not cherry-picked lower outliers [ref: #9].
- **Cross-suburb comparisons**: Moving to a different suburb to find comps requires an explicit location adjustment. Appraisers with limited market knowledge in that suburb cannot reliably quantify this adjustment — the comp should be avoided [ref: #6].
- **Reconciliation by simple averaging**: Expressly prohibited under API Valuation Protocol. Averaging treats all comps equally regardless of quality; professional reconciliation must be weighted and justified [ref: #3].
- **Hedonic model instability**: OLS hedonic coefficients can be unstable if the dataset is small, if collinearity exists between attributes (e.g., bedrooms and GFA), or if the market has shifted structurally since model calibration [ref: #8].

---

## 7. Open Questions / Knowledge Gaps

- **IVS 2025 edition specifics**: The API Valuation Protocol references the IVS 2025 edition for definitive adjustment hierarchy wording; full text of the January 2025 IVS update was not accessible during this research (PDF is access-controlled). The 2022 IVS wording confirmed the comparable transactions method but specific paragraph numbering for the 2025 edition requires direct API/IVSC membership access.
- **APRA AVM standards**: The Australian Prudential Regulation Authority's specific requirements for AVM use in mortgage lending (minimum confidence scores, acceptable LVR thresholds for AVM-only valuations) were not captured; this is a gap relevant to lender-instruction valuations.
- **Regional and rural AU**: Evidence on comp selection methodology and time adjustment calibration in rural Victoria (very low transaction volumes) is limited; the approaches likely deviate significantly from metropolitan practice.
- **Overlap with income approach for dual-income properties**: Properties with granny flats or dual-income configurations blur the boundary between SCA and income capitalisation. The approach for reconciling these in a CMA context is not formally documented in AU-specific guidance reviewed here.
- **Emerging ML-based adjuncts**: While XGBoost models showed superior accuracy in Melbourne metropolitan data, their interpretability (explainability of individual comp adjustments) remains poor. Regulatory acceptance of pure ML valuations for bank-grade purposes (APRA-supervised lenders) is an open question.

---

## 8. Sources

1. PropertyMetrics — "The Sales Comparison Approach to Real Estate Valuation" — https://propertymetrics.com/blog/sales-comparison-approach/ — accessed 2026-05-18 — credibility: medium (professional real estate education resource, detailed methodology coverage)
2. Corporate Finance Institute — "Sales Comparison Approach (Real Estate)" — https://corporatefinanceinstitute.com/resources/valuation/sales-comparison-approach-real-estate/ — accessed 2026-05-18 — credibility: medium (finance education platform, well-cited)
3. Australian Property Institute — "Valuation Protocol: Valuation Approaches and Methods" — https://www.api.org.au/professional-development/knowledge-hub/valuation-protocol-valuation-approaches-and-methods/ — accessed 2026-05-18 — credibility: high (peak AU valuation industry body; official protocol document)
4. IVSC — "IVS 105: Valuation Approaches and Methods" — https://www.ivsc.org/wp-content/uploads/2021/10/IVS105ValuationApproaches.pdf — accessed 2026-05-18 — credibility: high (international standards setter for valuation; primary standard)
5. Wikipedia — "Automated Valuation Model" — https://en.wikipedia.org/wiki/Automated_valuation_model — accessed 2026-05-18 — credibility: low (not primary source; useful for taxonomy of AVM types and industry context)
6. YouTube — "Sales Comparison Approach to Value" — https://www.youtube.com/watch?v=M7HusRc9Km8 — accessed 2026-05-18 — credibility: low (educational real estate exam video; useful for practitioner-level process illustration)
7. Hansen, J. (2009) — "Australian House Prices: A Comparison of Hedonic and Repeat‐Sales Measures" — Economic Record, Wiley — https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1475-4932.2009.00544.x — accessed 2026-05-18 — credibility: high (peer-reviewed, AU-specific academic study in leading AU economics journal)
8. Reserve Bank of Australia — "RDP 2006-03: A Comparison of Hedonic and Repeat-sales Measures" — https://www.rba.gov.au/publications/rdp/2006/pdf/rdp2006-03.pdf — accessed 2026-05-18 — credibility: high (RBA primary research; foundational Australian study on house price index methodology)
9. Consumer Affairs Victoria — "Underquoting information for real estate agents" — https://www.consumer.vic.gov.au/licensing-and-registration/estate-agents/running-your-business/underquoting-information-for-real-estate-agents — accessed 2026-05-18 — credibility: high (Victorian state government regulatory body; authoritative for AU/Vic compliance context)
10. Picki.com.au — "How to Estimate a Property's True Market Value Using Comparable Sales Analysis in Australia" — https://picki.com.au/blog/post/how-to-estimate-a-property-s-true-market-value-using-comparable-sales-analysis-in-australia — accessed 2026-05-18 — credibility: medium (AU property investment platform; 2026 publication with AU-specific selection criteria)
11. McKissock Learning — "Weighted Mean: A Simple Appraisal Reconciliation Technique" — https://www.mckissock.com/blog/appraisal/weighted-mean-a-simple-appraisal-reconciliation-technique/ — accessed 2026-05-18 — credibility: medium (accredited US appraisal education; reconciliation methodology directly applicable to AU practice)
12. Real Estate Business AU — "New Consumer Affairs Victoria Legislation - Matching Comparable Sales" — https://www.realestatebusiness.com.au/sales/11800-new-consumer-affairs-victoria-legislation-matching-comparable-sales — accessed 2026-05-18 — credibility: medium (AU real estate industry news; primary reporting on Victorian CMA legislation)