Activity Exports

Create and download asynchronous impression, request, and bid activity exports with the CloudX CLI.

Activity Exports

Use the cloudx export group to create asynchronous activity exports for completed UTC dates:

• cloudx export impressions for impression-level revenue data
• cloudx export requests for request-level auction activity
• cloudx export bids for bid and non-bid activity

Each command prints an export ID and its current status. Add --wait to wait for a terminal status, or use --output <path> to wait and download the gzip-compressed CSV file.

cloudx export bids

Creates a full-day bid activity export as a gzip-compressed CSV file. By default, the file contains bid responses together with no-bid, timeout, and error rows, which supports analysis from bidder request through bid outcome. The export runs asynchronously so large result sets do not have to fit within a normal API response.

Winner rows contain the actual bid price. For losing bid rows, bid_price_cpm is always 0 and bid_price_disclosure is redacted.

Usage

cloudx export bids --date YYYY-MM-DD [filters] [--wait] [--output <path>]

--date is required and must identify a completed UTC day within the previous 45 days. The current UTC day cannot be exported because it is not complete.

Options

Only two bid activity exports may be processing for an account at the same time. Submitting the same date and filters again reuses the matching queued, running, or completed export instead of creating duplicate work.

When an export is ready, CloudX returns a download URL that expires after one hour. Re-running the same command can issue a fresh URL while the generated file is still retained. Generated files are retained for up to seven days.

Examples

Queue an export and print its export ID and current status:

cloudx export bids --date 2026-08-10

Wait for completion and download all bidder outcomes:

cloudx export bids \
  --date 2026-08-10 \
  --output ./bids-2026-08-10.csv.gz

Download only bid responses for one app and bidder:

cloudx export bids \
  --date 2026-08-10 \
  --app com.example.game \
  --bidder example-bidder \
  --bids-only \
  --output ./example-bidder-2026-08-10.csv.gz

API workflow

API clients can create the same export with POST /export/bids. The response contains an export_id and a processing, ready, or failed status. Poll GET /export/bids/{export_id} until the status is terminal; a ready response includes a short-lived url and expires_at.

{
  "date": "2026-08-10",
  "app_bundle": "com.example.game",
  "bidder": "example-bidder",
  "include_non_bids": true,
  "test_mode": "production"
}

CSV Schema

The CSV contains one row per matching bidder outcome. Row order is not guaranteed.

nonbid_reason follows the IAB Tech Lab Seat Non-Bid status-code ranges. CloudX passes through bidder-supplied OpenRTB nbr values from 1 through 99. When CloudX determines the outcome, it currently emits these values:

cloudx export requests

Creates a full-day request activity export as a gzip-compressed CSV file, with one row per CloudX auction request. Use it to analyze request volume, fill decisions, bidder participation, and auction latency. The export runs asynchronously so large result sets do not have to fit within a normal API response.

request_status is filled when the auction selected a winner, unfilled when it did not, and error when request processing failed. A filled request does not guarantee that the client ultimately rendered an impression; join auction_id to the impression export for impression delivery and revenue.

Usage

cloudx export requests --date YYYY-MM-DD [filters] [--wait] [--output <path>]

--date is required and must identify a completed UTC day within the previous 45 days. The current UTC day cannot be exported because it is not complete.

Options

Only two request activity exports may be processing for an account at the same time. Submitting the same date and filters again reuses the matching queued, running, or completed export instead of creating duplicate work.

When an export is ready, CloudX returns a download URL that expires after one hour. Re-running the same command can issue a fresh URL while the generated file is still retained. Generated files are retained for up to seven days.

Examples

Wait for completion and download all request outcomes:

cloudx export requests \
  --date 2026-08-10 \
  --output ./requests-2026-08-10.csv.gz

Download unfilled production requests for one app and country:

cloudx export requests \
  --date 2026-08-10 \
  --app com.example.game \
  --country US \
  --request-status unfilled \
  --output ./unfilled-requests-2026-08-10.csv.gz

API workflow

API clients can create the same export with POST /export/requests. The response contains an export_id and a processing, ready, or failed status. Poll GET /export/requests/{export_id} until the status is terminal; a ready response includes a short-lived url and expires_at.

{
  "date": "2026-08-10",
  "app_bundle": "com.example.game",
  "request_status": "unfilled",
  "test_mode": "production"
}

CSV Schema

The CSV contains one row per matching CloudX auction request. Row order is not guaranteed.

cloudx export impressions

Creates a full-day, impression-level revenue export as a gzip-compressed CSV file. The export runs asynchronously so large result sets do not have to fit within a normal API request. When a publisher supplies a hashed user ID and applicable privacy signals allow it, the export includes that value for user-level revenue analysis.

Usage

cloudx export impressions --date YYYY-MM-DD [filters] [--wait] [--output <path>]

--date is required and must identify a completed UTC day within the previous 45 days. The current UTC day cannot be exported because it is not complete.

Options

Only two impression exports may be processing for an account at the same time. Submitting the same date and filters again reuses the matching queued, running, or completed export instead of creating duplicate work.

When an export is ready, CloudX returns a download URL that expires after one hour. Re-running the same command can issue a fresh URL while the generated file is still retained. Generated files are retained for up to seven days.

Examples

Queue an export and print its export ID and current status:

cloudx export impressions --date 2026-07-16

Wait for completion and download the result:

cloudx export impressions \
  --date 2026-07-16 \
  --output ./impressions-2026-07-16.csv.gz

Export production impressions for one iOS app in the United States:

cloudx export impressions \
  --date 2026-07-16 \
  --app com.example.game \
  --country US \
  --device-os iOS \
  --test-mode production \
  --output ./example-game-us-ios.csv.gz

Check the downloaded archive and preview its header and first rows:

gzip -t ./impressions-2026-07-16.csv.gz
gzip -cd ./impressions-2026-07-16.csv.gz | head

API workflow

API clients can create the same export with POST /export/impressions. The response contains an export_id and a processing, ready, or failed status. Poll GET /export/impressions/{export_id} until the status is terminal; a ready response includes a short-lived url and expires_at.

CSV Schema

The CSV contains one row per matching impression. Row order is not guaranteed.

Related

• CloudX CLI Overview
• Reporting
• Authentication Commands
• Examples