> For the complete documentation index, see [llms.txt](https://docs.luganodes.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.luganodes.com/rpc/platform-guide/statistics-and-ai-insights.md).

# Statistics and AI Insights

See what your apps are doing, and understand what changed.

The Statistics page shows requests, compute units, latency and errors for your organization. The home dashboard summarizes the same data at a glance: monthly CU usage against your plan, CU used today, peak requests per second against your plan limit, and error rate over the last 24 hours.

### Filters and scope <a href="#filters-and-scope" id="filters-and-scope"></a>

Everything on the page follows the app switcher in the top bar: pick **All Apps** for the whole organization or a single app. Three filters at the top right refine every chart and table at once:

<table><thead><tr><th width="122.140625">Filter</th><th>Options</th></tr></thead><tbody><tr><td>Network</td><td>All chains, or one of Ethereum, Bitcoin, Tron</td></tr><tr><td>Method</td><td>All methods, or a specific method</td></tr><tr><td>Time range</td><td>This billing cycle, last 7, 30 or 90 days, or a custom date range from the calendar</td></tr></tbody></table>

The **Get AI insights** button sits beside the filters, so whatever scope you set is the scope it analyzes.

### What the page shows <a href="#what-the-page-shows" id="what-the-page-shows"></a>

<table><thead><tr><th width="187.7734375">Section</th><th>What it tells you</th></tr></thead><tbody><tr><td>Usage overview</td><td>Total requests, success rate, error count and throttled requests for the selected period</td></tr><tr><td>Monthly CU cap</td><td>Compute units used against your plan allowance, with cycle dates and percent used</td></tr><tr><td>Daily request volume</td><td>Requests per day over time, with a toggle between requests and CU</td></tr><tr><td>Requests by chain</td><td>Per chain volume, share percent and totals, with a requests vs CU toggle</td></tr><tr><td>Requests per second</td><td>RPS over time with your plan limit drawn as a reference line</td></tr><tr><td>Usage by app</td><td>Per app CU consumed, limit and percent used for the selected period</td></tr><tr><td>Top methods</td><td>Methods ranked by requests, with share, CU consumed and success and failure percent</td></tr><tr><td>Latency distribution</td><td>Histograms by network and by method across buckets from 0-10 ms up to above 1000 ms</td></tr></tbody></table>

### Request logs <a href="#request-logs" id="request-logs"></a>

Recent requests with timestamp, method, chain, latency and status. Three filters narrow the list: search by method, status code (200, 400, 401, 429, 500, 503) and latency band (under 50 ms, 50-200 ms, above 200 ms). Logs cover the last 30 days.

### AI Insights <a href="#ai-insights" id="ai-insights"></a>

Select **Get AI insights** on the Statistics page to generate a written explanation of your usage, scoped to whatever filters are applied. It describes what changed in the selected period and why, for example which app, chain or method drove a spike in compute units or errors. Set the filters you care about first, then generate.

### Data retention <a href="#data-retention" id="data-retention"></a>

* Statistics can be queried for up to 6 months back. The date picker does not allow earlier dates.
* Request logs are available for the last 30 days.
* Members with the Viewer role see only the apps they have access to, in every chart, table and log. See [teams, members and roles](/rpc/platform-guide/teams-members-and-roles.md).

{% hint style="info" %}
To be notified when usage crosses a threshold instead of checking the dashboard, set up [alerts](/rpc/platform-guide/alerts.md).
{% endhint %}