> ## Documentation Index
> Fetch the complete documentation index at: https://docs.adriel.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Data reference

> Metrics, breakdowns, refresh cadence, and limits for the Across ADN data source in Adriel.

## Introduction

Across ADN is a Korean display advertising network. The Adriel connector fetches daily campaign- and ad-group-level reports from the Across ADN REST API, flattens the nested campaign-to-group response into one row per date, campaign, and ad group, and stores the result so widgets can read it. Metrics and currencies are standardized so Across ADN data aggregates alongside other sources without extra configuration, and Korean campaign and group names are preserved end-to-end.

To connect this data source, see [How to connect Across ADN to Adriel](/data-sources/a-f/across-adn/how-to-connect).

## Data refresh strategy

Across ADN is a file-cache connector. Each scheduled run pulls the current report from the API and upserts it into a per-asset monthly cache. There is no real-time or on-demand fetch; widgets always read cached data.

### Architecture data

Across ADN has no separate architecture sync. Campaign and ad group names arrive inside the report response and are surfaced as breakdowns rather than as independently fetched hierarchy entities.

### Reports data

Report data is cached, not queried live. The cache refreshes twice per day at **22:10 UTC** and **05:10 UTC** (07:10 and 14:10 KST). The 10-minute offset is intentional so the connector does not request each day's report before Across ADN has finished making it available. Rows are upserted on the composite key of campaign name, cache file, date, and ad group name, and the report date is normalized from `yyyyMMdd` to `yyyy-MM-dd` during ingestion.

## Architecture levels

Across ADN exposes group-level performance, with campaign and ad group surfaced as breakdowns:

Campaign → Ad group

## Metrics

<Note>
  **How to read the columns**
  **Data type** — how the metric is formatted (Number, Currency, Percentage, Ratio, Duration, Date, Text, URL, Array, Boolean). Widgets and formulas use this to render values correctly.
  **API Key** — code style like `impressions` is the literal Across ADN API field name. *Italic text* describes how a value is produced when it doesn't map cleanly to a single API field.
</Note>

### Reach & impressions

| Metric      | Description                     | Data type | API Key    |
| ----------- | ------------------------------- | --------- | ---------- |
| Impressions | Number of times ads were shown. | Number    | `view_cnt` |

### Click performance

| Metric      | Description                       | Data type | API Key       |
| ----------- | --------------------------------- | --------- | ------------- |
| Clicks      | Number of clicks on ads.          | Number    | `click_cnt`   |
| Click sales | Sales value attributed to clicks. | Currency  | `click_sales` |

### Conversion performance

| Metric                    | Description                                     | Data type | API Key               |
| ------------------------- | ----------------------------------------------- | --------- | --------------------- |
| Conversions               | Total attributed conversions.                   | Number    | `conv_cnt`            |
| Conversion value          | Monetary value of attributed conversions.       | Currency  | `conv_price`          |
| Direct conversions        | Conversions attributed to a direct interaction. | Number    | `direct_conv_cnt`     |
| Direct conversion value   | Monetary value of direct conversions.           | Currency  | `direct_conv_price`   |
| Indirect conversions      | Conversions attributed indirectly.              | Number    | `indirect_conv_cnt`   |
| Indirect conversion value | Monetary value of indirect conversions.         | Currency  | `indirect_conv_price` |

## Breakdowns

| Breakdown | Description                                                               | API Key         |
| --------- | ------------------------------------------------------------------------- | --------------- |
| Date      | Report date. Normalized from `yyyyMMdd` to `yyyy-MM-dd` during ingestion. | `wdate`         |
| Campaign  | Campaign name. Korean text is preserved end-to-end.                       | `campaign_name` |
| Ad group  | Ad group name. Korean text is preserved end-to-end.                       | `group_name`    |

## Limitations

* Only group-level rows are stored. Pre-aggregated date-level and campaign-level totals from the raw API response are discarded, so campaign and date totals are recomputed in Adriel from group-level rows using breakdowns.
* Campaigns without any ad group produce no rows. When a campaign's group array is null or undefined, that campaign is skipped entirely.
* Data is cache-only. It refreshes twice per day at 22:10 and 05:10 UTC, so recent changes appear only after the next scheduled run rather than immediately.
* Across ADN can be combined with other connectors in a single widget only when their breakdowns match exactly. Otherwise, use blended data to bring the sources together.
* Authentication uses a single API key sent as the `API-KEY` HTTP header. There is no OAuth flow or token refresh.

## API references

* Across ADN report API endpoint: `https://manage.acrosspf.com/api/api_report/adriel_across_adn_api_report.php`

## See also

* [How to connect Across ADN to Adriel](/data-sources/a-f/across-adn/how-to-connect)
* CrossTarget data reference
* Targeting Gates data reference