Trusted by teams turning documents into production data
Real Estate
Closing disclosures detail a mortgage's final terms: interest rate, monthly payment, closing costs, and cash due. The TRID format packs hundreds of data points into five pages. Use Sensible to get structured closing disclosure data for post-closing QC and compliance.
Validated JSON
Schema-enforced output; every field matches your contract
Source coordinates
Every value links back to page + bounding box for audit
Per-document pricing
Predictable cost. No token-volatility surprises
Trusted by teams turning documents into production data
TRID-regulated fields, dense fee tables, and buyer/seller allocations leave no margin for error. We extract every figure and validate it against the rules before it reaches your system.
01
TRID-regulated fields carry tolerance requirements. An error in the wrong fee category creates compliance liability. Sensible anchors extraction to the official form structure so regulated fields map correctly every time.
02
Sections A through H. Borrower-paid, seller-paid, lender columns. Each fee line is extracted with its correct party attribution, eliminating manual reconciliation of the closing cost tables.
03
The same document contains separate sections for buyer and seller charges. Fields like "Total Closing Costs" appear twice with different values. Sensible preserves which side each fee belongs to, so your QC process can validate buyer and seller figures independently.
Managed services
Solutions engineers handle plan, build, deploy, and adjust on your behalf. You see clean JSON in your API response. Same engine as self-serve, just with the configuration work outsourced.
What's included
01Plan.Engineers review your samples and pick the right method
02Build.SenseML configs written from your samples
03Deploy.Same engine as self-serve, ready for production
04Adjust.We update configs when formats shift or new edge cases appear
05Integrate.Help with custom integration into your downstream systems
Every closing disclosure is different, so we build the config around your exact schema rather than a fixed list. These are the fields teams request most; we map whatever else your post-closing QC workflow needs.
01
Loan terms
Loan amount, interest rate, monthly P&I, prepayment penalty, balloon payment, loan product, loan term, loan purpose
02
Closing costs
Origination charges, points, appraisal fee, title insurance, recording fees, transfer taxes, prepaids, initial escrow, total closing costs
03
Cash to close
Total closing costs, closing costs paid before closing, deposit, funds from borrower, seller credits, adjustments, cash to close
config.json
SenseML
{ /* SenseML: closing disclosure extraction */
"fields": [
{
"method": {
"id": "queryGroup",
"queries": [
{ "id": "loan_amount", "description": "loan amount, amount financed" },
{ "id": "interest_rate", "description": "interest rate, your interest rate, note rate" },
{ "id": "monthly_pi", "description": "monthly principal and interest, P&I payment, monthly payment" },
{ "id": "total_closing_costs", "description": "total closing costs, closing costs, total costs" }
// + more fields, mapped to your schema
]
}
}
]
}Pre-built configurations cover TRID-era closing disclosures and HUD-1 settlement statements. Sensible's financial services template library handles both, with hybrid extraction for lender-specific layout variations.
CFPB Closing Disclosure (standard 5-page format), Closing Disclosure alternatives, closing cost worksheets
HUD-1 Settlement Statement, HUD-1A, revised HUD-1 (still used for commercial, HELOC, and reverse mortgages)
Answers about TRID compliance fields, fee table extraction, and tolerance checks.
Yes. Both documents output structured JSON with matching field names. You can compare fees, rates, and terms between the two to check for tolerance violations programmatically.
Sensible extracts each fee line from Sections A through H including origination charges, third-party services, taxes, prepaids, and escrow items with both borrower-paid and seller-paid amounts.
By default Sensible captures loan amount, interest rate, monthly P&I payment, prepayment penalty, balloon payment indicator, loan product type, loan term, and projected payment schedule. The field set is defined in configuration, so you extract exactly what your downstream system needs and add or drop terms as that changes.
Yes. Sensible extracts all five pages of the CFPB Closing Disclosure form including loan terms, projected payments, closing costs, cash to close, and loan calculations.
Yes. Sensible sends extraction results to your webhook endpoint when processing completes. You can also poll the API for status.
Yes. Sensible flags extractions with low confidence for human review. You can configure review thresholds and workflows.
Sensible is SOC 2 Type II certified and HIPAA compliant. Data is encrypted in transit and at rest.
Document data is stored indefinitely by default. Custom retention policies are available and can be configured for same-day deletion if needed.
Yes. Sensible offers a 14-day free trial on the Growth plan. No credit card required to start.
Sensible uses per-document pricing for predictable costs. No token-based billing or usage surprises. Volume discounts are available for higher throughput.
Sensible provides REST APIs and SDKs for Python and Node.js. Most integrations take a few hours. Webhooks, Zapier, and direct API calls are all supported.
Sensible processes PDFs (native or scanned), Microsoft Word (DOC, DOCX), spreadsheets (XLSX, XLS, CSV), single-page images (JPEG, PNG), multi-page images (TIFF), and email bodies with attachments.
Accuracy depends on document quality and configuration. Most production deployments achieve 95%+ accuracy with proper validation rules and confidence signals.
Processing speed depends on document size, page count, OCR requirements, and which extraction methods are used. Simple single-page documents process in seconds. Larger or more complex documents that use LLM-based extraction take longer.
Start free
Free 14-day trial. No credit card required. Start extracting in minutes.