What gets extracted from a quote

Updated July 11, 2026

Once a carrier quote PDF is read, see How ProClose reads a PDF, the parser walks the text looking for a fixed set of fields. Here's the full list, organized the way it shows up on the deck.

#The basics

  • Named Insured, the client's legal name
  • Mailing Address
  • Policy Period
  • Program, the type of package the quote is written on
  • Transaction Type, new business or renewal
  • Quote Expires, the date the quote goes stale
  • Retail Producer, the producer of record on the quote

#Carrier and rating

  • The carrier name, matched from patterns like "X Insurance Company," "X Group," "X Exchange," or an "Underwritten by:" line. This name never reaches the client, see Carrier masking for why and how it's hidden.
  • AM Best rating

#Premium lines

Every premium line item the carrier lists gets pulled individually, not just a single total: general liability premium, commercial property premium, equipment breakdown enhancement, policy fee, surplus lines tax, state transaction fee, and stamping fee, wherever the carrier includes each. The carrier's own printed total is captured too, separately, so it can be checked against the sum of the line items.

#General liability limits

General aggregate, products and completed operations aggregate, personal and advertising injury, each occurrence, damage to premises rented to you, medical expense, and the deductible.

#Property

Building value, valuation method, coinsurance, cause of loss form, wind/hail deductible, all-other-perils deductible, business income, business income form, and whether equipment breakdown is included.

#Other quote details

  • TRIA, the additional premium for terrorism coverage, if the carrier quoted it
  • Flood, whether the quote mentions flood coverage anywhere
  • Excess liability, the schedule of excess limits and their premiums, when the quote includes an excess liability section
  • Subjectivities, the carrier's numbered list of binding requirements. This one is read carefully: the parser follows the list across page breaks and treats unnumbered continuation lines as part of the item above them, rather than cutting the item off at the page boundary. Missing a binding subjectivity is a real E&O exposure, so this field gets extra attention in the parsing logic.

#What gets read from a current policy

When you also drop the client's current policy for a side-by-side comparison, see Add the current policy for a comparison, the parser reads a narrower, comparison-focused set of fields: carrier, total premium, building value, valuation method, wind/hail deductible, business income (explicitly flagged if the current policy excludes it), equipment breakdown (included or excluded), the general liability each-occurrence limit, and the policy period.

The valuation field on a current policy is deliberately conservative. A loose pattern match can grab garbage off a real document, and on a client-facing comparison a wrong value does more damage than a missing one. So this field only fills in when the parser is confident it found a real valuation method, not just something that looks like one.

#Every field cites its page

Every one of these fields, on both the quote and the current policy, carries the page number it was read from. That's what makes the extraction checkable rather than a black box. See Page citations for where those citations live and how to use them.