Property Valuation Pack

We built the Property Valuation Pack so you don't have to reconcile HCAD data, MLS comps, and rental rolls in a spreadsheet that breaks every time you add a row. Real estate valuation isn't just a math problem; it's a data engineering problem. You're pulling land values, improvements, school districts, and neighborhood boundaries from disparate sources, then trying to force them into a single defensible number. The result is usually a fragile model where a single typo in the square footage flips the cap rate, or where the AVM confidence score is just a hallucination with no audit trail.

The industry treats valuation as a black box. Agents guess, brokers eyeball, and algorithms spit out numbers without showing their work. We reject that. Valuation requires a deterministic workflow: comparable sales analysis, income capitalization, and market trend assessment, all validated against canonical formulas and regulatory standards. This pack gives you the schema, the scripts, and the validators to automate the heavy lifting. If you're also building investment analysis workflows, you know that inconsistent inputs destroy downstream ROI calculations. We fixed the input layer so your outputs are defensible.

The Data Reconciliation Nightmare

Valuation data is a zoo of incompatible formats. You have the County Appraiser's HCAD-style appraised value, which splits land and improvements but might lag the market by six months. You have the MLS, which gives you sales prices but often lacks the granular detail needed for income properties, like operating expense ratios or vacancy rates. You have CoStar or local market reports that give you cap rates and GRMs, but they're PDFs or charts, not structured data. Reconciling these into a single valuation requires manual extraction, normalization, and cross-checking.

The Sales Comparison Approach is the most common method, but it's deceptively complex. It estimates property value by reference to comparable sales, requiring four main steps: collecting data, verifying accuracy, comparing properties, and reconciling differences ^[2]. Each step is prone to error. Did you adjust for the garage? The lot size? The year built? The school district? A manual adjustment matrix is easy to mess up. One wrong adjustment factor and your value drifts 10%. The Market Approach, also known as the sales comparison approach, relies on recent sales of similar properties to estimate value ^[5]. Without automated validation, you're trusting your own spreadsheet formulas. We built this pack to enforce the schema. The avm-input-schema.json template requires fields like land_value, improvements_value, square_footage, year_built, and school_district. If the data is missing, the validator catches it before you commit to a listing.

Why Manual Valuations Bleed Deals

A 5% error on a $2M property isn't a rounding issue; it's a $100k mistake that kills deals. When you list a property based on a sloppy comp adjustment, and the appraisal comes in $50k lower, the buyer's lender flags the discrepancy. The deal stalls. The seller loses confidence. You look unprofessional. The cost of a bad valuation isn't just the hours you spent building the model; it's the opportunity cost of lost transactions and damaged trust.

The Income Capitalization Approach adds another layer of risk. It estimates value based on the income a property generates, converting future benefits into present worth ^[7]. The math is simple: NOI divided by Cap Rate. But getting the inputs right is hard. You need to derive the cap rate from comparable sales ^[1]. You need to calculate NOI accurately, accounting for vacancy, collection losses, and operating expenses. If your cap rate is off by 25 basis points on a $5M asset, you're off by $125k. The Income Approach assumes that value equals the present value of the future income stream ^[8]. Without a structured workflow, you're manually deriving rates and calculating NOI, which introduces human error. A bad valuation also impacts your commercial lease negotiations; if your market analysis is wrong, you might sign a lease at a rate that doesn't support your valuation thesis.

Compliance is another silent killer. Automated Valuation Models (AVMs) are computerized systems that use comparable or repeat sales indexes to determine value, generally without an onsite inspection ^[3]. But if you're using an AVM for a listing or transaction, you need to ensure it meets model validation standards and error tolerance guidelines. The CFPB requires audit trails and compliance with fair lending practices. If your valuation process is a black box, you're exposing yourself to regulatory risk. You need a system that logs every adjustment, every data source, and every calculation step.

A Mixed-Use Asset That Exposed the Math Gap

Imagine a portfolio manager evaluating a 12-unit multifamily asset in a growing submarket. The property has 4,500 square feet of commercial space on the ground floor and 12 residential units above. The manager has the rental roll: $4,000 per unit per month, 90% occupancy, and operating expenses of $15,000 per month. They need to calculate the NOI, find three comparable sales, and determine the market value.

Without the Pack, the manager opens Excel. They calculate NOI: ($4,000 12 0.9) - $15,000 = $28,800 annualized. They search for comps. Sale 1: $2.1M, 10 units, $280k NOI. Sale 2: $2.3M, 14 units, $310k NOI. Sale 3: $2.0M, 12 units, $260k NOI. They manually calculate cap rates: 1.33%, 1.35%, 1.30%. They average to 1.33%. They apply adjustments for the commercial space and the submarket trend. The math gets messy. They forget to adjust for the vacancy rate difference. They plug the numbers into a valuation formula and get a value. But did they check the GRM? Did they verify the land value against the county records? Did they ensure the AVM confidence score aligns with the manual analysis?

Now imagine running this through the Property Valuation Pack. The manager feeds the rental roll and property details into run-valuation.py. The script ingests the data, validates it against validate-property-data.sh, and runs the income capitalization math. It extracts the cap rate from the comps, applies the adjustment matrix, and calculates the value. It checks the GRM against the financial metrics framework. It flags any inconsistencies. The output is a structured JSON report with NOI, cap rate, GRM, ROI, and AVM confidence scores. The report is generated from valuation-report.yaml, with sections for comps, income approach, and market trends. The manager reviews the report, sees the audit trail, and confidently presents the valuation to the client. This level of rigor is what separates professionals from guessers. It also feeds directly into competitive research workflows, where you benchmark pricing against market positioning.

Automated Compliance and Structured Outputs

Once the Property Valuation Pack is installed, your workflow shifts from manual reconciliation to data validation and structured output. The skill.md orchestrator defines the end-to-end valuation workflow, referencing all templates, scripts, validators, references, and examples. You don't have to remember the steps; the skill enforces them.

The run-valuation.py script is the engine. It ingests validated property data, runs comparable sales and income capitalization math, applies market trend adjustments, and outputs structured valuation JSON. The financial-metrics.json framework defines the calculation keys and validation rules for NOI, cap rate, GRM, ROI, and AVM confidence scores. If your cap rate doesn't fall within the market range, the validator flags it. If your GRM is inconsistent with the cap rate, the script warns you. This is the kind of guardrail that prevents costly errors.

The iaao-avm-standards.md reference provides curated excerpts from the IAAO Standard on AVMs and CFPB compliance guidance. It covers model validation, error tolerance, and audit trails. You don't have to hunt for these standards; they're included in the pack. The valuation-methodologies.md reference contains canonical formulas and rules for comparable sales analysis, income capitalization, and market trend assessment with adjustment matrices. You get the math right, every time.

The valuation-report.yaml template generates compliant listing/transaction valuation reports. It includes sections for comps, income approach, and market trends. The output is structured, machine-readable, and ready for distribution. You can integrate this with real estate marketing workflows to optimize listing descriptions with accurate value propositions. You can also link it to financial modeling pipelines for scenario analysis and investor reporting. The result is a valuation process that is fast, accurate, and defensible.

What's in the Property Valuation Pack

• skill.md — Orchestrator defining the end-to-end valuation workflow, referencing all templates, scripts, validators, references, and examples for agent execution
• templates/avm-input-schema.json — Production JSON Schema enforcing required property fields (HCAD-style appraised value, neighborhood, school district, land value, square footage, year built) for AVM ingestion
• templates/valuation-report.yaml — Structured YAML template for generating compliant listing/transaction valuation reports with sections for comps, income approach, and market trends
• templates/financial-metrics.json — JSON framework defining calculation keys and validation rules for NOI, cap rate, GRM, ROI, and AVM confidence scores
• scripts/run-valuation.py — Executable Python workflow that ingests validated property data, runs comparable sales and income capitalization math, applies market trend adjustments, and outputs structured valuation JSON
• validators/validate-property-data.sh — Bash validator that checks input against schema constraints, verifies mathematical consistency, and exits non-zero on missing fields or invalid ratios
• references/iaao-avm-standards.md — Curated excerpts from IAAO Standard on AVMs and CFPB compliance guidance covering model validation, error tolerance, and audit trails
• references/valuation-methodologies.md — Canonical formulas and rules for comparable sales analysis, income capitalization (NOI/Cap Rate), and market trend assessment with adjustment matrices
• examples/worked-example.json — Realistic sample property dataset including land value, improvements, rental income, and three comparable sales for pipeline testing
• examples/expected-output.json — Pre-calculated expected results for the worked example, used by validators to confirm script accuracy and agent compliance

Ship Valuations with Confidence

Stop guessing property values. Stop wrestling with spreadsheets. Stop risking deals on manual math errors. The Property Valuation Pack gives you the tools to automate comparable sales analysis, income capitalization, and market trend assessment with enterprise-grade validation and compliance. Upgrade to Pro to install the skill and start shipping defensible valuations today.