> ## 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 Adjust data source in Adriel.

## Introduction

Adjust is a mobile measurement partner (MMP) focused on attribution, fraud prevention, and analytics across mobile ad networks. It supports iOS, Android, and connected TV through the Adjust SDK, and Adriel reads aggregated attribution and performance data through the Adjust Reporting Service API (Adjust's Automate reporting service, delivered as streamed CSV reports).

Each data source maps to one Adjust app, selected from the account's app list when the connection is created. The connector surfaces four platform metric categories — Conversion, Ad spend, Revenue, and SKAdNetwork — plus cohort-based retention and revenue metrics that Adriel derives on demand. App Tracking Transparency (ATT) status splits are exposed as separate install variants.

As a mobile attribution connector, attribution events, breakdowns, and currencies are standardized so widgets can aggregate across sources without additional configuration.

To connect this data source, see [How to connect Adjust to Adriel](/data-sources/a-f/adjust/how-to-connect).

## Data refresh strategy

### Architecture data

Architecture covers the app inventory tied to the connection. The app is chosen from the Adjust app list at connection time, which locks the data source to a single app for the life of the connection.

### Reports data

**Daily cache.** Report data is cached at the day level. Already-cached days are skipped when a query is re-run, and settled (older) days are re-fetched only when explicitly invalidated. Reports are requested at daily granularity using the UTC offset selected at setup.

**Hardcoded attribution settings.** Adjust reports are requested with fixed attribution settings — attribution types of click, impression, and engaged ad, and a dynamic attribution source. These are not configurable from the connector.

## Architecture levels

App → Network → Campaign → Ad group → Creative

## Metrics

<Note>
  **How to read the columns**
  **Data type** uses this vocabulary: Number, Currency, Percentage, Ratio, Duration, Date, Text, URL, Array, Boolean.
  **API Key** in code style like `installs` is the literal Adjust reporting field name. *Italic text* describes how a value is produced when it is derived by Adriel rather than returned as a single Adjust field.
</Note>

### Installs & app events

<Note>
  iOS install metrics are also broken out by App Tracking Transparency (ATT) status — authorized, denied, restricted, and not determined — as separate `installs_att_*` variants. Use these to understand consent rates per network or campaign.
</Note>

| Metric                 | Description                                                                               | Data type | API Key          |
| ---------------------- | ----------------------------------------------------------------------------------------- | --------- | ---------------- |
| Installs               | Attributed installs.                                                                      | Number    | `installs`       |
| Reattributions         | Re-attribution events (returning users re-attributed to a new source).                    | Number    | `reattributions` |
| SKAN installs          | SKAdNetwork-reported installs.                                                            | Number    | `skad_installs`  |
| Installs by ATT status | Installs split by ATT consent status (authorized / denied / restricted / not determined). | Number    | `installs_att_*` |

### Clicks & impressions

| Metric      | Description          | Data type | API Key       |
| ----------- | -------------------- | --------- | ------------- |
| Clicks      | Tracked clicks.      | Number    | `clicks`      |
| Impressions | Tracked impressions. | Number    | `impressions` |

### Attribution & touchpoints

<Note>
  Apple's privacy-preserving SKAdNetwork returns fine-grained conversion values (0–63) only on the first postback for some users; postbacks 2 and 3 return coarse Low/Medium/High signals ([Adjust SKAdNetwork reporting](https://dev.adjust.com/en/api/rs-api/reports/)). Both the fine total and the coarse per-postback values are exposed, and are best read together for the fullest SKAN picture.
</Note>

| Metric                           | Description                                                     | Data type | API Key                                            |
| -------------------------------- | --------------------------------------------------------------- | --------- | -------------------------------------------------- |
| SKAN conversion value total      | Sum of fine-grained conversion values from the first postback.  | Number    | `skad_conversion_value_total`                      |
| SKAN coarse value (1st postback) | Apple's Low/Medium/High signal from postback 1.                 | Text      | `skad_coarse_cv_1`                                 |
| SKAN coarse value (2nd postback) | Coarse signal from postback 2.                                  | Text      | `skad_coarse_cv_2`                                 |
| SKAN coarse value (3rd postback) | Coarse signal from postback 3.                                  | Text      | `skad_coarse_cv_3`                                 |
| SKAN invalid payloads            | SKAN postbacks Adjust could not parse.                          | Number    | `skad_invalid`                                     |
| Reattributions per user          | Reattributions divided by cohort size over the selected window. | Ratio     | *Adriel-derived (reattributions ÷ cohort size)*    |
| Deattributions per user          | Deattributions divided by cohort size over the selected window. | Ratio     | *Adriel-derived (deattributions ÷ cohort size)*    |
| Reattributions per deattribution | Reattributions divided by deattributions.                       | Ratio     | *Adriel-derived (reattributions ÷ deattributions)* |

### In-app revenue

| Metric                         | Description                                                  | Data type  | API Key                                                     |
| ------------------------------ | ------------------------------------------------------------ | ---------- | ----------------------------------------------------------- |
| Revenue                        | In-app purchase revenue.                                     | Currency   | `revenue`                                                   |
| Ad revenue                     | Ad-monetization revenue.                                     | Currency   | `ad_revenue`                                                |
| All revenue                    | Combined in-app and ad revenue.                              | Currency   | *Adriel-derived (revenue + ad revenue)*                     |
| Revenue events                 | Count of revenue-generating events.                          | Number     | `revenue_events`                                            |
| Revenue per user               | Revenue divided by cohort size over the selected window.     | Currency   | *Adriel-derived (revenue ÷ cohort size)*                    |
| All revenue per user           | All revenue divided by cohort size over the selected window. | Currency   | *Adriel-derived (all revenue ÷ cohort size)*                |
| Revenue events per user        | Revenue events divided by cohort size.                       | Ratio      | *Adriel-derived (revenue events ÷ cohort size)*             |
| Revenue events per paying user | Revenue events divided by paying users.                      | Ratio      | *Adriel-derived (revenue events ÷ paying users)*            |
| Paying users rate              | Paying users divided by cohort size.                         | Percentage | *Adriel-derived (paying users ÷ cohort size)*               |
| First-time paying users rate   | First-time paying users divided by retained users.           | Percentage | *Adriel-derived (first-time paying users ÷ retained users)* |

### Cost & ROAS

| Metric                  | Description                                                           | Data type | API Key                                           |
| ----------------------- | --------------------------------------------------------------------- | --------- | ------------------------------------------------- |
| Ad spend                | Reported network spend.                                               | Currency  | `cost`                                            |
| Network-attributed cost | Cost variants attributed per network.                                 | Currency  | `network_cost_*`                                  |
| Click cost              | Cost per click.                                                       | Currency  | *Adriel-derived (cost ÷ clicks)*                  |
| Impression cost         | Cost per thousand impressions (eCPM-style).                           | Currency  | *Adriel-derived (cost ÷ impressions × 1000)*      |
| Install cost            | Cost per install (eCPI).                                              | Currency  | *Adriel-derived (cost ÷ installs)*                |
| Cost per paying user    | Ad spend divided by first-time paying users over the selected window. | Currency  | *Adriel-derived (cost ÷ first-time paying users)* |
| Gross profit            | Revenue minus cost.                                                   | Currency  | *Adriel-derived (revenue − cost)*                 |

### Retention & engagement

<Note>
  Retention and engagement metrics are cohort-based: the window (for example 7 days or 30 days) is chosen when the metric is added, and values are computed on demand from the underlying cohort data.
</Note>

| Metric                        | Description                                                      | Data type  | API Key                                              |
| ----------------------------- | ---------------------------------------------------------------- | ---------- | ---------------------------------------------------- |
| Sessions                      | App sessions.                                                    | Number     | `sessions`                                           |
| Retention rate (all users)    | Retained users divided by cohort size over the selected window.  | Percentage | *Adriel-derived (retained users ÷ cohort size)*      |
| Retention rate (paying users) | Paying users divided by retained users over the selected window. | Percentage | *Adriel-derived (paying users ÷ retained users)*     |
| Sessions per user             | Sessions divided by cohort size.                                 | Ratio      | *Adriel-derived (sessions ÷ cohort size)*            |
| Time spent per user           | Total time spent divided by cohort size.                         | Duration   | *Adriel-derived (time spent ÷ cohort size)*          |
| Time spent per active user    | Time spent divided by retained users.                            | Duration   | *Adriel-derived (time spent ÷ retained users)*       |
| Time spent per session        | Time spent divided by non-install sessions.                      | Duration   | *Adriel-derived (time spent ÷ non-install sessions)* |

### Categories not currently surfaced

Three additional Adjust metric categories exist in the connector source but are not currently exposed: **Subscription** (activations, cancellations, renewals, trial starts), **Fraud** (rejected installs and reattributions by type), and **Assist** (assisted installs, assisting clicks and impressions). Workspaces that need any of these should contact support.

## Breakdowns

### Hierarchy

| Breakdown | Description                  | API Key    |
| --------- | ---------------------------- | ---------- |
| Network   | Group by ad network.         | `network`  |
| Campaign  | Group by campaign.           | `campaign` |
| Ad group  | Group by ad group or ad set. | `adgroup`  |
| Creative  | Group by creative or ad.     | `creative` |

### Audience & geography

| Breakdown | Description            | API Key   |
| --------- | ---------------------- | --------- |
| Country   | Group by country code. | `country` |

### Device & platform

| Breakdown   | Description                                | API Key       |
| ----------- | ------------------------------------------ | ------------- |
| OS          | Group by operating system (iOS / Android). | `os_name`     |
| Device type | Group by device type (phone / tablet).     | `device_type` |

### Date & time grouping

<Note>
  Adjust returns daily records; Adriel groups them into these views for reporting, so all time-grouping breakdowns are Adriel-side aggregations of the same underlying daily data.
</Note>

| Breakdown           | Description                                                  |
| ------------------- | ------------------------------------------------------------ |
| Auto Time Breakdown | Automatically picks the best granularity for the date range. |
| Daily               | Split reports by day.                                        |
| Weekly (Start: Mon) | Split reports by ISO week, starting Monday.                  |
| Weekly (Start: Sun) | Split reports by week, starting Sunday.                      |
| Monthly             | Split reports by calendar month.                             |
| Quarterly           | Split reports by calendar quarter.                           |
| Yearly              | Split reports by calendar year.                              |

## Limitations

* **Attribution settings are fixed** — reports always request click, impression, and engaged-ad attribution types with a dynamic attribution source. Different attribution semantics are not configurable from the connector; contact support if you need them.
* **SKAN values are partly coarse** — for privacy reasons, Apple returns fine-grained conversion values only on the first postback for some users; later postbacks carry coarse Low/Medium/High signals.
* **Cohort windows are capped** — cohort-based retention and revenue metrics are limited by a maximum month-cohort window of 12 months, with a shorter default day-cohort window. Some workspaces have a raised day-cohort cap.
* **Subscription, Fraud, and Assist not surfaced** — these Adjust categories exist in source but are not currently exposed.
* **One app per data source** — the app is fixed at connection time; tracking another app requires a separate data source.
* **Timezone fixed at setup** — reports use the UTC offset chosen when the data source is created.

## API references

* [Adjust Reporting Service API (Automate)](https://dev.adjust.com/en/api/rs-api/)
* [Adjust SKAdNetwork reporting](https://dev.adjust.com/en/api/rs-api/reports/)
* [Adjust developer documentation](https://dev.adjust.com/en/)

## See also

* [How to connect Adjust to Adriel](/data-sources/a-f/adjust/how-to-connect) (paired how-to)
* [AppsFlyer data reference](/data-sources/a-f/appsflyer/data-reference) — alternative MMP with similar coverage
* [Singular data reference](/data-sources/o-z/singular/data-reference) — alternative MMP with SKAN and unified attribution