Extract by DocompareAI

Supplier catalog workflow

Turn supplier catalogs and price lists into an Excel-ready CSV

Define a comparison schema once, extract repeated product rows, and review the source evidence behind prices, pack sizes, minimum orders, and lead times.

Use public or non-confidential PDFs only. Live-job exports require paid access; the linked example output is a public demonstration download.

Who this is for

Procurement analysts, operations specialists, consultants, and spreadsheet users comparing public or non-confidential supplier information.

The problem

Suppliers describe equivalent products differently. Direct copy-and-paste preserves those differences but rarely produces a clean comparison table.

Typical input

Public catalogs, product sheets, and price-list PDFs that you are allowed to process. Use a separate job when currencies, tax treatment, or commercial terms cannot be normalized safely.

Workflow

A repeatable path from PDFs to reviewable rows

  1. 01

    Choose comparable catalogs

    Confirm publication date, currency, tax basis, and whether the documents cover the same product category.

  2. 02

    Define the comparison columns

    Specify supplier SKU, normalized product, pack size, unit price, minimum order, lead time, and any exclusions.

  3. 03

    Extract repeated product records

    Use relational output when each catalog contains several product lines rather than one summary row.

  4. 04

    Review evidence and normalize

    Check the available source evidence, resolve unit differences, and use the Excel-ready CSV for further spreadsheet analysis.

Fields

Define the columns before the run

Supplier SKU

Keep the supplier-provided identifier unchanged so each row remains traceable.

Normalized product

A reviewable comparison label, separate from the supplier wording.

Pack size

The number and unit included in the quoted price, such as 5 reams or 100 gloves.

Unit price and currency

The listed price with currency and tax basis stated explicitly.

Minimum order and lead time

Commercial constraints that can matter more than the lowest listed price.

Worked output

A small, concrete result

SupplierNormalized productPackPrice (AUD)Lead time
NorthstarA4 recycled copy paper5 reams24.903–5 days
MeridianA4 recycled copy paper5 reams25.402–4 days

Synthetic worked output. Prices and suppliers are fictional and are provided only to demonstrate the comparison structure.

Comparison

Structured extraction compared with a manual copy-and-paste workflow

CriterionExtract workflowManual workflow
NormalizationSupplier wording and normalized comparison labels can be kept as separate fields.The reviewer invents a convention while copying each row.
Repeated recordsRelational output can create one row per product line.Rows are added manually and may be skipped or duplicated.
TraceabilityAvailable evidence connects a value with its source page and quote.A spreadsheet cell may not retain its exact PDF source.

Limitations to plan for

  • • A listed price is not a total-cost comparison; freight, tax, rebates, and contract terms may sit elsewhere.
  • • Product names alone do not prove that specifications are equivalent.
  • • Scanned tables and multi-column layouts may produce approximate or unavailable highlights.
  • • Review every material price and unit before relying on the CSV for purchasing decisions.