How to Understand Confidence Scores

CargoLint assigns a confidence percentage to each extracted field, indicating how certain the AI is about that value’s accuracy. Understanding these scores helps you prioritize review work and trust the platform’s results.

Before you start

• You have documents that have been processed
• You understand basic document types (invoices, packing lists, etc.)
• You know how to access the review queue

Steps

1. Know the 70% confidence threshold

CargoLint flags any field with confidence below 70% for review. This threshold balances accuracy with efficiency—fields above 70% are typically reliable enough to use without review, while fields below require human verification.

2. Understand what confidence means

A confidence score reflects the AI’s mathematical certainty that a value is correct, based on patterns in its training data and how clearly the value appears in the document. A field with 95% confidence was extracted from clear, unambiguous text. A field with 55% confidence comes from faint, obscured, or ambiguous text that could be misread.

3. Recognize per-document-type thresholds

Different document types have different extraction difficulty:

• Invoices: Usually highest confidence because invoices follow standard formats
• Bills of lading: Medium confidence due to varied templates and handwritten sections
• Packing lists: Medium-high confidence with clear item descriptions
• Certificates of origin: Medium confidence because regulatory language varies by country

A 65% confidence on a bill of lading is normal; on an invoice, it signals something unusual.

4. Identify high confidence fields in the UI

In the review editor, high-confidence fields (80% and above) appear with a green badge or checkmark. You can usually approve these without detailed review, though spot-checking is always wise.

5. Identify medium confidence fields in the UI

Fields between 70% and 79% show a yellow or orange badge. Review these carefully, but they’re usually correct. Compare them against the original document image.

6. Identify low confidence fields in the UI

Fields below 70% display a red badge or warning icon. CargoLint automatically queues these for review. Always verify low-confidence fields against the original document before approving.

7. Investigate why confidence is low

Low confidence often occurs when:

• Text is handwritten or printed at an unusual angle
• The document is blurry, faded, or has poor contrast
• A field contains special characters or formatting
• Multiple interpretations exist (for example, date formats like “03/14” could be March 14 or the 14th of March depending on regional convention)

Improving the scan quality (see Scanning Tips) often resolves low-confidence extraction.

Tip: Don’t automatically distrust low-confidence fields. The AI is simply indicating uncertainty, not that the value is wrong. Always compare against the original document.

Tip: High confidence doesn’t mean 100% accuracy. Even 95% confident fields can occasionally be incorrect, especially with unusual document layouts. Spot-check a few high-confidence fields each day.

What’s next

• Reviewing Documents – Correct fields in the review queue
• Scanning Tips – Improve scan quality to boost confidence scores
• Troubleshooting – Resolve issues with consistently low confidence