> ## Content Index
> Fetch the complete content index at: https://blog.jestor.com/llms.txt
> Use this file to discover other available public pages before exploring further.

# How to track AI usage in your operation
- URL: https://blog.jestor.com/track-ai-usage-and-ocr-pages/
- Published: 2026-09-16T18:50:25.000Z
- Updated: 2026-09-16T18:50:25.000Z
- Author: Bruno Bannach

To **track AI usage** meaningfully, you need a metric people can predict. In **Jestor**, document reading is measured in processed pages rather than in variable credits, so a three page **PDF** counts as three and an image counts as one.

## Why unpredictable metrics discourage adoption

When consumption is measured in credits that vary per document, nobody can estimate the cost of automating a process. A dense invoice might consume far more than a simple one, and the team has no way to plan.

That uncertainty is usually what stops a promising **AI** workflow from being rolled out beyond a pilot.

What a predictable metric enables:

- Estimating the cost of automating a document flow before building it
- Comparing the effort of two approaches honestly
- Explaining consumption to a finance team
- Scaling a pilot without a budget surprise

## How the metrics differ

| Metric          | What it measures          | Predictability              |
| --------------- | ------------------------- | --------------------------- |
| Processed pages | Document reading with OCR | One page counts as one      |
| AI credits      | Other AI features         | Varies with the task        |
| Usage metrics   | Consumption over time     | Visible inside the platform |

## Step by step to keep usage under control

1. **Open the usage metrics** inside the platform to see current consumption.
2. **Identify which processes consume most**, usually the ones reading documents.
3. **Count pages, not files**, when estimating a document flow.
4. **Pilot on a real sample** before enabling a flow across the whole operation.
5. **Add a validation step** so a person checks extraction quality early on.
6. **Review consumption monthly** and adjust which processes use reading.

## Why pages is the right unit for document reading

A page is something a person can count before building anything. Ask how many invoices arrive per month, multiply by their average length, and you have an estimate. Credits that vary per document cannot be estimated that way, which is exactly why the separation exists.

- **One page equals one:** a three page PDF counts as three
- **Images count as one:** a photo of a document is a single page
- **Separate from other AI usage:** document reading measured on its own
- **Visible in the platform:** consumption tracked without external reporting

## Frequently Asked Questions

### How is document reading measured?

In processed pages. A three page PDF counts as three and an image counts as one. See [jestor.com](https://www.jestor.com/?ref=blog.jestor.com).

### Why separate pages from AI credits?

Because reading cost varies per document, and pages give a number teams can predict.

### Where do I see my consumption?

In the usage metrics available inside the platform.

## Video Tutorial: Step by Step

With **Jestor**, you can automate workflows, connect teams, and build internal systems your way, all without code and powered by **AI**. Discover Jestor at [jestor.com](https://www.jestor.com/?ref=blog.jestor.com) and see how to take your company's operations to a new level of efficiency and control.