> For the complete documentation index, see [llms.txt](https://help.rankability.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://help.rankability.com/account-and-settings/control-your-usage.md).

# Control your usage

Rankability shows a credit estimate before supported metered work begins. Use the final pre-run amount—not a base rate by itself—when deciding whether to continue. Keywords, platforms, pages, competitors, locations, and Local Pack grid points can multiply the total.

For current base rates, use the [Credit costs reference](/account-and-settings/credit-costs-reference.md). This article focuses on the decisions that keep a workflow within budget.

## Before starting metered work

1. Confirm the selected workspace and project. Work created in the wrong workspace consumes the same usage and does not automatically move its history or Serena context.
2. Read the action name, quantity, and displayed estimate in the confirmation step.
3. Remove platforms, URLs, competitors, locations, or keywords you do not need.
4. Check whether a matching project or run already exists. When Rankability warns about recent or in-progress work, open that result before choosing to create another paid run.
5. Approve the action only after the credit estimate matches the intended scope.

Cached results can be free when the product explicitly labels them that way. Do not assume that repeating a request will use a cache.

## Control Track usage

Tracking is calculated per keyword, platform, scan, and applicable location configuration.

* Select only the traditional and AI platforms needed for the reporting goal.
* Use a benchmark when you need a one-time measurement instead of recurring tracking.
* Turn automatic tracking off when a keyword no longer needs scheduled scans.
* Review added locations and Local Pack grid size. Each grid point is a separate check, so moving from a `3x3` to a `7x7` grid materially changes the total.
* Avoid starting a manual scan while another run is pending or running.

See [Setting up keyword tracking](/track/setting-up-keyword-tracking.md) for project configuration and [Supported tracking platforms](/track/supported-tracking-platforms.md) for platform and grid examples.

## Control Copywriter usage

* Open the avatar menu and choose **Usage & billing**, or use **Settings → Billing**, to review the shared credit balance.
* Choose **Review the brief first** when you want to approve the brief before deciding whether Serena should draft.
* Choose **Write it myself** when you want research and editor guidance without generating an initial AI draft.
* The current in-app Create workflow does not offer a new Quick mode toggle. Existing research-only Quick mode projects remain accessible, but they should not be treated as a selectable way to reduce a new project's cost.
* For multiple content assets, review every row and the combined credit estimate before starting the batch.
* If an Autopilot start times out, check for a recent active project before retrying. The server may have accepted the first run even when the browser did not receive the response.

Research-platform breadth changes the evidence gathered, not Copywriter's fixed completed-outcome credit rate. Autopilot reaches its charge point after it produces a usable draft; a later optimization, fact-checking, or delivery step can still need attention. Images remain user-selected in the editor. Read [Creating a content project](/copywriter/creating-a-content-project.md) before choosing a workflow.

## Control Researcher, Audit, Promote, and Serena usage

* Analyze only the saved keywords, competitors, pages, URLs, or prospects you intend to act on.
* Use the lighter research mode when it answers the question; deeper modes cost more because they collect broader evidence.
* Import only useful client sources. URL imports charge per page, while pasted text and supported file behavior are shown in the source dialog.
* Serena chat is free, but deep research, URL imports, and Serena-started tool actions follow their displayed rates.
* Do not rerun an audit merely to refresh the page. Check the existing run status and timestamp first.

## Audit usage after the run

Use **Settings → Billing** to review the current credit balance and available breakdowns by feature and client.

A reservation is not always final usage. If an operation fails, refresh Billing after its status settles and check the operation's rule in [Understanding billing and credits](/account-and-settings/understanding-billing-and-credits.md).

If a completed transaction still does not match the client, action, quantity, or estimate you approved, contact support with the project or run ID, transaction timestamp, and displayed amount. Never include an API key or private source content in the support message.