# Document Capture Confidence Threshold Simulator

> Source: https://netray.co/tools/document-capture-confidence-threshold-simulator
> Publisher: Netray (netray.co), AI and integration engineering for ERP and manufacturing systems.
> Section: Free Tools

## Short answer

Simulate where to set the auto-accept confidence threshold by weighing review labor against the cost of errors that slip through.

## Category

ai-automation

## Audience

- AP manager
- Quality manager
- Automation lead

## Tool

**Kind:** calculator

**Currency:** $

**Inputs:**

#### Documents per month

**Key:** docs

**Type:** number

**Default:** 5000

**Min:** 0

**Step:** 100

**Unit:** docs

#### Auto-accept confidence threshold

**Key:** threshold

**Type:** slider

**Default:** 90

**Min:** 60

**Max:** 99

**Step:** 1

**Unit:** %

#### Share auto-accepted at a 90 threshold

**Key:** share90

**Type:** number

**Default:** 70

**Min:** 0

**Max:** 100

**Step:** 1

**Unit:** %

**Help:** Read this from your tool's confidence report.

#### Share lost per threshold point above 90

**Key:** slope

**Type:** number

**Default:** 2.5

**Min:** 0

**Step:** 0.1

**Unit:** pts

**Help:** Points of auto-accept share lost for each point of threshold raised (and gained when lowered).

#### Error rate in auto-accepted docs at 90

**Key:** err90

**Type:** number

**Default:** 2

**Min:** 0

**Step:** 0.1

**Unit:** %

#### Error reduction per threshold point

**Key:** decay

**Type:** number

**Default:** 12

**Min:** 0

**Max:** 50

**Step:** 1

**Unit:** %

**Help:** Relative fall in error rate for each point the threshold is raised.

#### Minutes per reviewed document

**Key:** review_min

**Type:** number

**Default:** 2.5

**Min:** 0

**Step:** 0.5

**Unit:** min

#### Loaded hourly cost of reviewer

**Key:** rate

**Type:** number

**Default:** 30

**Min:** 0

**Step:** 1

**Unit:** $/h

#### Cost of one escaped error

**Key:** err_cost

**Type:** number

**Default:** 45

**Min:** 0

**Step:** 5

**Unit:** $

**Help:** Rework, payment correction, vendor contact and so on.

**Outputs:**

#### Auto-accepted share

**Key:** auto_share

**Expr:** clamp(share90 - slope * (threshold - 90), 0, 100)

**Format:** percent

#### Error rate in auto-accepted documents

**Key:** err_rate

**Expr:** err90 * pow(1 - decay / 100, threshold - 90)

**Format:** decimal

#### Documents sent to review per month

**Key:** review_docs

**Expr:** docs * (1 - auto_share / 100)

**Format:** integer

#### Monthly review cost

**Key:** review_cost

**Expr:** review_docs * review_min / 60 * rate

**Format:** currency

#### Escaped errors per month

**Key:** escaped

**Expr:** docs * auto_share / 100 * err_rate / 100

**Format:** decimal

#### Monthly cost of escaped errors

**Key:** error_cost

**Expr:** escaped * err_cost

**Format:** currency

#### Total monthly cost

**Key:** total_cost

**Expr:** review_cost + error_cost

**Format:** currency

**Primary:** true

**Chart:**

#### Monthly cost vs confidence threshold

**Type:** line

**X:** threshold

**From:** 70

**To:** 99

**Steps:** 30

**Series:**

- review_cost
- error_cost
- total_cost

**Disclaimer:** A simple two-parameter model anchored at a 90 threshold. Replace the anchors and slopes with values read from your own confidence-versus-correctness report.

## Content

**Intro:** Every capture tool lets you decide how confident it must be before a document skips review. Set it too low and errors reach your ERP; set it too high and reviewers drown. This simulator makes the trade-off visible as two cost curves and a total, so the threshold becomes a decision rather than a default.

**Sections:**

#### How to use the simulator

**Body:** Take a calibration run from your tool, if it provides one: the share of documents that would auto-accept at several thresholds and the error rate among them, found by checking a sample. Enter the values at the 90 point, plus how fast the auto-accept share falls and the error rate improves as you raise the threshold. Add reviewer time and cost, and the cost of one error that escapes. The chart then sweeps the threshold from 70 to 99 and plots review cost, error cost and the total.

**Bullets:**

- Calibrate with real, checked samples, not the tool's own claims
- Estimate escaped error cost from actual rework cases
- Look for the bottom of the total curve, then check how flat it is

#### The model

**Body:** Auto-accept share is assumed to fall linearly as the threshold rises, because fewer documents reach the bar. The error rate among accepted documents is assumed to fall geometrically, since the lower confidence tail is what you are removing. Review cost is the review volume times minutes times rate. Error cost is accepted volume times error rate times the cost of a single error. The sum has a minimum where the marginal review cost equals the marginal errors avoided, which is the economic threshold.

#### Reading the curve

**Body:** A flat valley means the exact threshold is not critical and you can favor the lower review load. A steep rise on the left tells you errors are expensive. Different fields deserve different thresholds, so run it separately for money fields and for free text. Remember that a regulated or high-value document, such as a certificate used for release, may justify a threshold far above the economic optimum, because one escaped error has an unbounded cost.

**Bullets:**

- Run separate scenarios for high and low value document types
- Re-run monthly while the model is learning
- Never use the simulator to set thresholds for safety or compliance documents without a risk review

**Cta:** If you want thresholds set from a measured calibration on your own documents, and a review queue that feeds corrections back to the model, we can build that on your own infrastructure with human approval on every write-back.

**Faq:**

#### What does a confidence score actually mean?

**Answer:** It is the tool's internal estimate that a value is correct, and the scale differs between products. A score of 0.9 in one tool is not equivalent to 0.9 in another, so always calibrate the score against checked outcomes from your own documents.

#### Should I use one threshold for all fields?

**Answer:** Usually not. Amounts, tax IDs and PO numbers carry higher consequences than descriptions or dates, so they deserve stricter thresholds. Many tools allow per field thresholds, and the simulator can be run once per field class.

#### How do I get the anchor values?

**Answer:** Export confidence scores for a few hundred processed documents, have a person verify them, and bucket results by score. The share above each score and the error rate within that share give you the anchor and slopes.

#### Why is my total cost curve nearly flat?

**Answer:** A flat curve means review cost and error cost trade off almost evenly across the range. Choose the lower threshold to reduce review load, but watch for rare high-cost errors that the average cost per error may understate.

## Search terms

ocr confidence threshold, document capture confidence score threshold, auto accept threshold intelligent document processing, human in the loop threshold, idp review queue sizing, confidence threshold trade off, straight through processing threshold, invoice capture threshold tuning

---
Talk to an engineer: https://netray.co/contact?intent=erp_ai | Machine catalog: https://netray.co/agent/catalog.json