> ## 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 data source.

## Why is my AppsFlyer dashboard slower than other connectors?

AppsFlyer is a real-time connector: report data is fetched from AppsFlyer on demand on every dashboard query, with no local daily cache to read from. This keeps the numbers fresh, but it means each query waits on a live call to the AppsFlyer Master API instead of returning a pre-built snapshot, so dashboards built on this connector load slightly slower than cache-backed connectors.

<Note>
  Live queries are subject to AppsFlyer's API rate limit (approximately 5 calls per 4 seconds per API token). Persistent slowness or throttling usually means another tool is consuming the same token's request budget.
</Note>

To reduce load, keep the number of AppsFlyer widgets on a single dashboard reasonable and avoid very wide breakdown combinations (a maximum of 15 breakdowns can be applied to a single query).

## Why do the last few days of installs keep changing?

This is expected. Every query re-fetches the last **8 days** of attribution data, and AppsFlyer continues to revise recent numbers after the fact. The 8-day refresh window exists specifically to absorb those revisions, so the most recent days settle over time rather than being final on day one.

Recent figures move for several reasons:

* **Delayed postbacks** — attributions can land after the install itself.
* **Fraud filtering** — AppsFlyer removes installs flagged as fraudulent, which lowers earlier counts.
* **SKAdNetwork conversion-value updates** — iOS conversion values arrive on Apple's timer, not in real time.

Once a day falls outside the 8-day window, its numbers are considered stable.

## Why did my recent iOS installs arrive late?

SKAdNetwork (SKAN) installs and their conversion values are delivered on Apple's postback timer rather than in real time. Apple deliberately delays and batches these postbacks, so iOS installs from the last few days often appear lower at first and then fill in as the postbacks land.

This is a platform behavior, not a sync issue. The 8-day refresh window is what allows those late-arriving SKAN postbacks to be picked up on subsequent queries, so the recent days keep updating until they settle.

## Why can't I change my app's reporting currency without recreating the data source?

The data source's currency is read from the app's attributes at the moment the data source is created and is then locked to that data source. There is no in-place setting to change it later.

If the app's currency is changed in AppsFlyer afterward, the existing data source keeps the currency it was created with. To pick up the new currency, recreate the data source:

<Steps>
  <Step title="Confirm the new currency in AppsFlyer">
    Verify the app's currency is set correctly in the AppsFlyer account first, since the value is captured at creation time.
  </Step>

  <Step title="Create a new AppsFlyer data source">
    Add a fresh AppsFlyer data source for the same app so the current currency is captured. See [How to connect AppsFlyer to Adriel](/data-sources/a-f/appsflyer/how-to-connect).
  </Step>

  <Step title="Repoint widgets to the new data source">
    Update dashboards to use the new data source, then remove the old one once everything reads correctly.
  </Step>
</Steps>

## Where are my cohort retention and LTV curves?

Cohorted retention curves and lifetime-value (LTV) projections do not come from this connector. They are handled by a separate connector, [AppsFlyer Cohort](/data-sources/a-f/appsflyer-cohort/data-reference), which reads AppsFlyer's Data Locker exports and must be set up on its own.

The standard AppsFlyer connector does include a simple retention snapshot — day-1, day-7, and day-30 retention and retention rate — but these are single-point figures, not cohort curves aligned by conversion or event date. For period-by-period retention and LTV, use the AppsFlyer Cohort data source.

<Note>
  The two connectors share the same brand but are entirely separate: they use different credentials, different data sources, and different refresh strategies. Connecting one does not connect the other.
</Note>

## Related

* [AppsFlyer data reference](/data-sources/a-f/appsflyer/data-reference) — metrics, breakdowns, refresh cadence, and limits
* [How to connect AppsFlyer to Adriel](/data-sources/a-f/appsflyer/how-to-connect) — paired setup guide
* [AppsFlyer Cohort data reference](/data-sources/a-f/appsflyer-cohort/data-reference) — cohorted retention and LTV metrics
* [Adjust data reference](/data-sources/a-f/adjust/data-reference) — another mobile attribution (MMP) data source