> ## Documentation Index
> Fetch the complete documentation index at: https://parabola.io/docs/llms.txt
> Use this file to discover all available pages before exploring further.
# Usage & analytics
> How credits are calculated, who can see the usage reports, and where to find the flows driving most of your consumption.
What consumes a credit, where to see how many you've used, and how to spend fewer.
## Usage
Credits are calculated every time you run a Prowork flow or have AI complete complex tasks. The amount used in a given run depends on the number of unique steps and the complexity of the process being automated. If a step isn't successfully executed, or has zero input and output rows, it doesn't count.
* **What consumes credits?** Steps, visualizations, and artifacts.
* **When does it happen?** Every run, whether you clicked **Run flow** or a schedule, webhook, or inbound email fired it.
There's one credit pool per account. Steps draw from the same balance as everything else.
## Usage analytics
Users across all Prowork plans have access to embedded usage analytics and reports. These reports are meant purely for analytical purposes and do not impact any account-specific billing. Only org admins, billing admins, and reporting admins have access to them.
For accounts that have access, the reports are on the [Billing & Usage](https://parabola.io/app/billing) page of your Prowork account. All credit usage data is updated once per day; the previous day's credits are reflected at approximately 3am EST the following day.
### Who can see usage
| Admin type | What they see on the Billing & Usage page |
| :- | :- |
| **Org admin** | Everything, including subscription, payment method, and all usage sections |
| **Billing admin** | Subscription and payment settings, plus all usage sections |
| **Reporting admin** | **Credit usage**, **Usage overview**, and **Usage reports** only. No billing access. |
If you'd like to explore usage reports on your Prowork account and don't currently have access, please reach out to your Account Manager.
### Reports
There are currently four reports available.
**Summary**
* **Total flows.** Every flow in the account, whatever its last run date.
* **Flows created last 30 days.**
* **Users added last 30 days.** Counts users added whether or not they verified their account.
* **Total credits last 30 days.** Team-wide, regardless of the current billing period.
**Usage Summary**
Usage information across the entire team, focused on credits. Adjust the time frame with the **Starting Date** and **Ending Date** parameters.
* Total credits by day
* Top credit using users (by percentage)
* Total credits by flow
**User Report**
Usage information on a per-user level.
* Total flows owned, and how many of those include AI
* Created flows last 30 days
* Unique flows run last 30 days
* Total flow runs last 30 days
* Total credits last 30 days
**Flow Report**
Usage information on a per-flow level.
* Unique flow ID
* Flow name
* Owner email
* Includes AI
* Total runs last 30 days
* Total credits last 30 days
The AI fields on both reports look at the flow's most recent successful run, and are left blank for flows that haven't run. User Report and Flow Report data is a snapshot from the most recent date with usage, as of yesterday at midnight UTC.
### Downloading reports
All data in the usage reports is downloadable as a PDF, CSV, Excel, or JSON file. Either way, files download to the default download location on your device.
For a **PDF**:
* Select the cloud icon in the top right of the report
* Click **download as a PDF**
For **CSV, Excel, or JSON**:
* Hover over the three dots icon of the specific report you want
* Click **download results**
### Troubleshooting
You may see a field show null or blank. There are a few reasons a value can appear empty rather than zero:
* **No usage yet.** A brand-new flow or user with no runs in the reporting window has nothing to report, rather than a broken measurement.
* **The day's data hasn't landed.** Usage data updates once per day, reflected at approximately 3am EST for the previous day. Checking earlier the same day, today's numbers won't be there yet.
* **The field only applies once the flow has run.** **Includes AI** is left blank for flows that haven't run, because there's no most recent execution to check.
If a flow with confirmed recent runs still shows null credits, send the flow name and the report you're viewing to [help@parabola.io](mailto:help@parabola.io).
The reports show current names. If you rename a flow or change a user's email, the report updates to match. Match on **Unique Flow ID** if you're comparing snapshots over time.
## Maximize efficiency
A flow that's built well uses exactly the work it needs: no step running longer than necessary, no schedule running more often than the data actually changes. These are the habits that keep it that way as a flow grows and its usage scales with it.
**Filter early.** Every step downstream then works on a smaller table.
**Disconnect what you're done with.** A step that still processes rows consumes credits even when nothing downstream uses its output. Disconnecting it returns those credits to your pool.
**Match run frequency to how fast your data changes.** A source that refreshes nightly is well served by a nightly schedule. The same flow on a 10-minute interval runs about 143 more times a day for the same result.
**Split slow-changing work into its own flow.** Heavy cleaning on a static file or a weekly reference set consumes credits on every run of the flow it sits in, however often that flow runs. Move that work into a second flow, land the cleaned result in a [Parabola Table](/docs/product/integration/table), and have your primary flow read the table instead of redoing the cleaning. When you want a refresh on demand, the [Run another Prowork flow](/docs/product/integration/parabola-flows) step lets the primary flow trigger the upstream one.
**Measure a single flow's consumption.** The credit counter on the Billing page updates in real time, so note it, run the flow, and note it again. There's no per-run credit report yet; this before-and-after method is the current way to isolate one flow's usage.
**Review the Flow Report monthly.** Total Credits by Flow surfaces the handful of flows carrying most of the consumption. That's where optimizing pays off.
This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.