> ## 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.

# FAQs

> Common questions, expected behaviors, and troubleshooting for the AppsFlyer Cohort data source.

## Why is my cohort data up to 12 hours behind?

This is expected. AppsFlyer Cohort reads daily cohort exports that AppsFlyer writes into the connected Amazon S3 bucket, and the connector serves dashboards from a cache that it refreshes on a fixed schedule rather than on every query. Because a new export has to land in S3 first and then be picked up by the next scheduled sync, the worst-case gap between an event occurring and it appearing on a dashboard is roughly 12 hours.

Newly-made changes are not immediate. They appear after AppsFlyer publishes the next export and the following sync processes it. For the exact sync schedule, see the [AppsFlyer Cohort data reference](/data-sources/a-f/appsflyer-cohort/data-reference).

## Should I use conversion-date or event-date alignment?

Each AppsFlyer Cohort data source is anchored to one of two cohort dates, chosen when the data source is set up.

* **Conversion-date alignment** anchors each row to the install or conversion date, and period 0 reflects activity on that same day. This is best for retention curves and lifetime-value-by-day analysis, and it is what most teams use.
* **Event-date alignment** anchors each row to the event date, and the period reflects how long after install the event occurred. This is best for time-of-event analysis.

The alignment is a per-data-source choice. To analyze the same app under both models, set up a separate data source for each alignment.

## Can I report on a cohort day that isn't a standard checkpoint?

No. Cohort metrics are reported at eleven fixed day-offset checkpoints: `D0`, `D1`, `D3`, `D7`, `D14`, `D30`, `D45`, `D60`, `D90`, `D180`, and `D360`. Days between checkpoints, such as `D31` through `D44`, are not reported on their own. Activity in those gaps accumulates into the cumulative `Total*` series and is counted at the next checkpoint.

The checkpoint set is fixed to keep the connector stable, so a metric like day-45 retention is available while an arbitrary day-45-through-day-59 breakdown is not. For a full list of the periods and the point-in-time versus cumulative metrics, see the [AppsFlyer Cohort data reference](/data-sources/a-f/appsflyer-cohort/data-reference).

## Why doesn't my new AppsFlyer event show up as a metric?

When the connector finds an event name in the export that isn't already on the data source's enabled events, it records the name but does not expose it as a metric yet. Discovered events stay gated until they are added explicitly, which keeps the metric list scoped to the events a team actually reports on.

At data source creation, a set of common AppsFlyer event names is available to choose from. App-specific and custom events surface after the first sync has discovered them. To turn a discovered event into metrics, edit the data source and add the event to its enabled events.

<Steps>
  <Step title="Confirm a sync has run">
    A newly-fired event only becomes selectable after the export containing it has landed and been processed. If the event is very recent, wait for the next sync.
  </Step>

  <Step title="Edit the data source">
    Open the AppsFlyer Cohort data source settings, where discovered event names appear alongside the default events.
  </Step>

  <Step title="Add the event to the enabled events">
    Select the event so its cohort metrics are generated across the supported periods, then save.
  </Step>
</Steps>

<Note>
  An event that was enabled but is not present in the app's data is flagged as invalid on the settings page after a sync. Removing it keeps the metric list clean.
</Note>

## Why is all my revenue shown in USD?

Revenue is normalized to USD in the AppsFlyer Data Locker export before the connector reads it, so AppsFlyer Cohort stores and reports revenue in USD only and does not convert it to the workspace currency. This applies to every revenue and lifetime-value metric on the connector.

For revenue in the app's local currency, and for cost and ROAS, use the standard [AppsFlyer data reference](/data-sources/a-f/appsflyer/data-reference). AppsFlyer Cohort carries cohorted retention and revenue only; installs totals, in-app event totals, cost, and ROAS come from the standard AppsFlyer connector.

## Should I create more than one data source for the same app?

Generally, no. One data source per app is recommended. A single refresh processes all of a connection's data sources together, so duplicating the same app under one connection multiplies processing cost without adding data.

The main reason to create a second data source for the same app is to analyze it under a different cohort alignment or report template, since those are fixed per data source. Outside of that, a single data source per app is the efficient setup.

## Related

* [AppsFlyer Cohort data reference](/data-sources/a-f/appsflyer-cohort/data-reference) — cohort periods, metrics, breakdowns, and limits
* [How to connect AppsFlyer Cohort to Adriel](/data-sources/a-f/appsflyer-cohort/how-to-connect) — paired setup guide
* [AppsFlyer data reference](/data-sources/a-f/appsflyer/data-reference) — the standard AppsFlyer connector for installs, in-app events, cost, and ROAS