INVOICE PROCESSING

Detect Price Creep in PDF Invoices Dropped into Google Drive

Triggers when a vendor invoice PDF is added to a Google Drive folder, extracts amounts, compares them to prior invoices.

CategoryInvoice Processing
EngineSim + Paperclip
Difficultyintermediate
Triggerevent
Steps5
Setup~15 min

How it runs

The automated pipeline, trigger to output.

  • TriggerNew invoice PDF added to Google Drive folderGoogle DriveGoogle Drive
  • ActionDownload PDF and extract vendor and amountsOpenAI
  • ActionLook up vendor's prior invoice amountsPostgreSQLPostgres
  • LogicCompute drift and assign pass or flagged verdict
  • OutputLog invoice and verdict to Notion databaseNotionNotion

What it does

For teams that collect invoices as PDFs rather than structured data, this workflow turns a shared Drive folder into a drift monitor. Each time a PDF lands, it reads the document, pulls out the vendor and line-item amounts, and checks them against what that vendor billed previously. Whether or not drift is detected, it records a full entry in a Notion database, building a searchable audit trail with a clear pass or flagged verdict on every invoice.

When to use it

Choose this when invoices arrive as PDF attachments saved to Drive and you want both detection and a durable, browsable history in Notion. Good for operations teams without a dedicated AP platform.

How it works

  1. 1A new PDF appears in the watched Google Drive folder and triggers the run.
  2. 2The workflow downloads the file and extracts vendor, totals, and line items.
  3. 3It queries Postgres for the vendor's previous invoice amounts.
  4. 4A logic step calculates drift and assigns a pass or flagged verdict.
  5. 5The new amounts are written back to Postgres for future comparisons.
  6. 6A Notion database row is created with the verdict, amounts, drift percentage, and a link to the PDF.

Set it up

What you configure once, before turning it on.

  1. 1
    Connect Google DriveDocs, sheets, slides, files.
  2. 2
    Connect PostgresAny Postgres URL — query, write, migrate.
  3. 3
    Connect NotionPages, databases, comments.
  4. 4
    Connect OpenAIModels, embeddings, files.
  5. 5
    Set each agent's modelWe leave models unset so you pick the tier — fast + cheap, or top-quality.
  6. 6
    Tune it to your dataEdit the prompts, filters, and field mappings so it matches how your team works.
  7. 7
    Test, then turn it onRun once against a sample, confirm the output, then enable the trigger.

Run this workflow in your colony.

14-day trial. No DevOps. No Sales call. Provisioned in under a minute.