The ETL export integration is only available on our Grow plan.
RevenueCat can automatically send data deliveries of all of your apps' transaction data to various cloud storage providers. These are in the form of .csv files delivered daily.
Setup Instructions
Transaction Format
All dates and times are in UTC.
Header | Can be null | Comments |
---|---|---|
| Can be used as a unique user identifier to find all of a user's transactions. | |
| Can be used together with | |
| β | Last seen country of the subscriber. |
| The product identifier that was purchased. | |
| Purchase time of transaction. | |
| β | Expiration time of subscription. Null when |
| The source of the transaction. Can be | |
|
| |
|
| |
|
| |
|
| |
| β | The gross revenue generated from the transaction. All prices are converted to USD. Can be null if product prices haven't been collected from the user's device. |
| 0.7 or 0.85. Use this to calculate the proceeds of a transaction. | |
| orderId or transaction_identifier. βCan be used as unique id. | |
| orderId of first purchase or | |
| β | When a refund was detected, |
| β | When we detected an unsubscribe (opt-out of auto renew). |
| β | When we detected billing issues, |
| β | The currency that was used for the transaction. |
| β | The product's price in the currency that was used for the transaction. |
| β | An array of entitlements that the transaction unlocked or |
| Always starts at 1. Trial conversions are counted as renewals. | |
| If | |
| β | The offering presented to users. Can be used to filter Experiment transactions. |
| β | The reserved subscriber attributes set for the subscriber. Keys begin with |
| β | The custom attributes set for the subscriber. |
| β | Last seen platform of the subscriber. |
Re-enable integration to update to latest version
If your exports don't contain all of the columns above, you may be on an older export version. To update to the latest version disconnect, and re-connect the integration from the RevenueCat dashboard.
A note on transaction data
All transaction data is based on the store receipts that RevenueCat has received. Receipts often have inconsistencies and quirks which may need to be considered. For example:
- The expiration date of a purchase can be before the purchase date. This is Google's way of invalidating a transaction, for example when Google is unable to bill a user some time after a subscription renews. This doesnβt occur on iOS.
- Google returns only 90 days of recent transaction history, so some old transactions maybe missing if Google fetch tokens were imported.
- Apple and Google do not provide the transaction price directly, so we must rely on historical data for the products that we have. This isnβt 100% accurate in cases where the prices were changed or receipts were imported.
- Renewal numbers start at 1, even for trials. Trial conversions increase the renewal number.
- Data is pulled from a snapshot of the current receipt state, this means that the same transaction can be different from one delivery to another if something changed, e.g.refunds, and billing issues. You should recompute metrics for past time periods periodically to take these changes into account.
We try to normalize or at least annotate these quirks as much as possible, but by and large we consider receipts as the sources of truth, so any inconsistencies in the transaction data can always be traced back to the receipt
Sample Queries
The following sample queries are in Postgresql.
-- Active trials
SELECT
COUNT(*)
FROM
transactions
WHERE
end_time > NOW()
AND is_trial_period = TRUE
AND is_sandbox = FALSE
AND refunded_at IS NULL;
-- Active subscriptions
SELECT
COUNT(*)
FROM
transactions
WHERE
end_time > NOW()
AND is_trial_period = FALSE
AND is_sandbox = FALSE
AND refunded_at IS NULL;
-- The RevenueCat charts exclude promotional transactions.
-- you can include the following filter to exclude promotional
-- transactions from your queries as well
product_identifier NOT ILIKE 'rc_promo%'
-- Revenue past 28 days (USD)
SELECT
SUM(price)
FROM
transactions
WHERE
start_time > (CURRENT_DATE - INTERVAL '28 days')
AND is_sandbox = false;
Updated 4 months ago