Skip to content

Google Authorized Buyers still requires five separate CSV reports in 2026

Atomic fact: Google Authorized Buyers does not allow you to get bid requests and creative-level detail in a single export.

This is not a documentation gap. It is a deliberate schema constraint that has existed for years and remains in force in 2026.

Why five reports are mandatory

Google Authorized Buyers has field incompatibilities that prevent combining everything you need for real optimization into one file:

  • Creative-level performance metrics remove the "Bid requests" column.
  • Bid request / pipeline fields remove creative IDs and some performance detail.
  • Publisher data can sometimes ride with bid requests, but not with creative-level rows.
  • Quality signals (viewability, fraud) arrive in their own shape.
  • Bid filtering / rejection reasons live in a fifth report.

Cat-Scan therefore ingests five distinct daily CSV exports and joins them into a usable model.

Atomic fact: Cat-Scan imports exactly these five report types and maps them to three core tables: rtb_daily, rtb_bidstream, and rtb_bid_filtering.

The five reports (exact naming and purpose)

All reports follow the naming convention catscan-{type}-{account_id}-{period}-UTC.

# Report type Target table Primary purpose Key limitation
1 bidsinauction rtb_daily Creative-level bids, wins, impressions, spend No raw bid requests
2 quality rtb_daily Viewability and measurable impressions No bid-request volume
3 pipeline-geo rtb_bidstream Bid requests and funnel by country + hour No creative ID
4 pipeline rtb_bidstream Bid requests and funnel by publisher No creative ID
5 bid-filtering rtb_bid_filtering Why bids were rejected by Google Separate from performance

Atomic fact (June 2026): Data imported before 2026-01-14 is marked data_quality='legacy' because earlier reports used inconsistent timezones. All current reports must be UTC.

How the joins actually work in practice

The importers (see the Cat-Scan platform repo) use a combination of date + buyer account + creative ID (where present) and publisher or geo dimensions to reconstruct the full picture.

You cannot simply union the files. You must deduplicate on import (Cat-Scan uses a row_hash unique constraint) and then aggregate across the five sources.

This is why a purpose-built control plane is required. Downloading the five CSVs and opening them in a spreadsheet does not give you the QPS funnel by config, the waste by size, or safe pretargeting recommendations.

Why this matters for agencies

Most agencies that finally obtain a Google Authorized Buyers seat discover the reporting problem only after the first month of spend. The native UI and the emailed CSVs are intentionally limited.

The five-report reality is one of the strongest signals that you are dealing with a real seat operator rather than someone who has only read the Authorized Buyers documentation.

Last updated: June 2026
Part of the RTB.cat / Cat-Scan technical explainers.
Source: production operation of real Google Authorized Buyers (Google ADX / DoubleClick) seats + the open-source Cat-Scan platform.

Built from docs c4d2c4f (platform N/A) — last updated June 12, 2026 20:59:27.