Applovin¶
AppLovin Corporation is an American mobile technology company headquartered in Palo Alto, California. AppLovin enables developers of all sizes to market, monetize, analyze and publish their apps through its mobile advertising, marketing, and analytics platforms MAX, AppDiscovery, and SparkLabs.
omniload allows ingesting data from AppDiscovery reporting API
URI Format¶
The URI format for Applovin is as follows:
applovin://?api_key=<your_api_key>
URI Parameters:
api_key. report key generated from your applovin account.
Setting up Applovin Integration¶
Generate a Report Key¶
You can generate a report key from your analytics dashboard.
Example: Loading Publisher Report¶
For this example, we’ll assume that:
api_keyisapi_key_0
We will run omniload to save this data to a duckdb database called report.db under the table name public.publisher_report.
omniload ingest \
--source-uri "applovin://?api_key=api_key_0
--source-table "publisher-report" \
--dest-uri "duckdb:///report.db" \
--dest-table "public.publisher_report"
Example: Incremental loading¶
We will extend the Loading Publisher Report example to demonstrate incremental loading.
First, we run the example with a start date of 2025-01-01 and an end date of 2025-01-05
omniload ingest \
--source-uri "applovin://?api_key=api_key_0
--source-table "publisher-report" \
--dest-uri "duckdb:///report.db" \
--dest-table "public.publisher_report" \
--interval-start "2025-01-01" \
--interval-end "2025-01-05"
We can query the database to see which dates the data was ingested for:
$ duckdb report.db 'select day from public.publisher_report group by 1'
┌────────────┐
│ day │
│ date │
├────────────┤
│ 2025-01-01 │
│ 2025-01-02 │
│ 2025-01-03 │
│ 2025-01-04 │
│ 2025-01-05 │
└────────────┘
Now, we will run omniload again, but we will omit start and end date to demonstrate an incremental load.
omniload ingest \
--source-uri "applovin://?api_key=api_key_0
--source-table "publisher-report" \
--dest-uri "duckdb:///report.db" \
--dest-table "public.publisher_report"
Now we can check the database again, and we will see that the data was loaded for the rest of the days automatically.
$ duckdb report.db 'select day from public.publisher_report group by 1'
┌────────────┐
│ day │
│ date │
├────────────┤
│ 2025-01-01 │
│ 2025-01-02 │
│ 2025-01-03 │
│ 2025-01-04 │
│ 2025-01-05 │
│ 2025-01-06 │
│ . │
│ . │
│ . │
│ 2025-01-28 │
│ 2025-01-29 │
│ 2025-01-30 │
│ 2025-01-31 │
├────────────┤
│ 35 rows │
└────────────┘
Tables¶
Name |
Merge Key |
Inc Key |
Inc Strategy |
Details |
|---|---|---|---|---|
|
day |
day |
merge |
Provides daily metrics from the |
|
day |
day |
merge |
Provides daily metrics from the |
|
day |
day |
merge |
Provides daily metrics from the |
|
day |
day |
merge |
Provides daily metrics from the |
Custom Reports¶
applovin source supports custom reports. You can pass a custom report definition to --source-table and it will dynamically create a report for you.
The format of a custom report looks like the following:
custom:{endpoint}:{report_type}:{columns}
Where:
{endpoint}is the API endpoint for applovin reports (one ofreport,probabilisticReportorskaReport){report_type}is the report type (one ofpublisheroradvertiser){columns}are the columns of the given report type.
Custom Report Example¶
For this example, we will ingest data from report end point with the report type publisher.
We want to obtain the following columns:
ad_type
clicks
country
To achieve this, we can pass the custom report defintion in --source-table
omniload ingest \
--source-uri "applovin://?api_key=api_key_0
--source-table "custom:report:publisher:ad_type,clicks,country" \
--dest-uri "duckdb:///report.db" \
--dest-table "public.custom_report"
[!NOTE] The
daycolumn is automatically added to any custom report if it is not specified in the custom report definition.