Skip to main content

​
Introduction

Google Campaign Manager 360 (CM360, formerly DoubleClick Campaign Manager / DCM) is Google’s ad-management and measurement platform for advertisers and agencies. The Adriel CM360 connector does not pull a fixed set of metrics — instead it pulls a pre-configured CM360 report that the customer has already built in Campaign Manager 360. During setup the customer selects a CM360 user profile and then one of that profile’s reports; Adriel runs that report through the DFA Reporting API, downloads the resulting CSV, and caches the cleaned rows as the data source. Because the data comes from a customer-defined report, the available fields are a schema-on-read result: the metrics and breakdowns are determined entirely by the CM360 report configuration, not by Adriel. Any column present in the report becomes an available field in Adriel, and each column’s Adriel data type is inferred from the cached values. This page documents the report model and the field-type-mapping rules rather than a metric catalog. To connect this data source, see How to connect Google Campaign Manager 360 to Adriel.

​
Data refresh strategy

CM360 is a scheduled cache connector, not an on-demand one. Adriel re-runs the bound report on a fixed schedule, downloads the new CSV, and replaces the cached data. Widgets and dashboards read from this cache, so they reflect the most recent successful refresh rather than a live query.
  • Refresh cadence. The report is refreshed twice daily, at approximately 07:00 and 19:00 KST (equivalently 22:00 and 10:00 UTC). CM360 data is not real-time.
  • Full atomic replace. Each refresh replaces the previous snapshot in full (SimpleFileCache with an ATOMIC_REPLACE strategy). Data does not accumulate across refreshes — the previous version is discarded when the new report is fetched, so history is only as deep as the date range configured in the CM360 report itself.
  • Report-run mechanics. On each refresh the connector triggers the report asynchronously and polls for completion with exponential backoff (immediate, then roughly 3, 7, and 19 minutes), because large reports can take 10–20 minutes to generate. A recently completed report file is reused within a short window to avoid re-running the same report.
Add a Date breakdown for time-series widgetsBecause the connector stores whatever the report produces, time-based widgets only work if the report itself includes a Date breakdown. Configuring the CM360 report with a Date breakdown and a dynamic (rolling) date range — rather than a fixed date period — is recommended so the cached data stays current after each refresh.

​
Architecture levels

CM360 user profile → Report. Setup walks two levels:
  1. User profile — the customer selects one CM360 user profile they can access.
  2. Report — the customer selects one pre-configured report belonging to that profile. Up to 100 reports per profile are listed (most-recently-modified first); profiles with more than 100 reports have the oldest ones omitted from the list.
One Adriel data source binds to one CM360 report. Selecting multiple reports during setup creates one data source per report. Each report’s data is cached in its own table.

​
Report model

The CM360 report defines the columns. Adriel supports reports exported in CSV format only; non-CSV reports (for example Excel) are not selectable. A report must contain at least one metric to be connectable.
  • Report types. Standard, Floodlight, and Activities reports are all supported. Once connected they behave identically — the report type only affects which metric field is validated at setup time.
  • CSV cleanup. The raw CM360 CSV wraps the data rows in report metadata: a title, metadata rows, a reportFields marker, the header row, the data rows, and a trailing total row. Adriel strips the surrounding metadata, the header artifacts, and the total row, keeping only the data rows before caching.
  • What you see equals what you export. The data available in Adriel is equivalent to the report the customer can export manually from Campaign Manager 360 — the same columns and the same rows (subject to the row cap in Limitations).

​
Field type mapping

CM360 has no fixed metric or breakdown catalog. The available fields are the columns of the cached report, and each column’s Adriel data type is inferred from the cached values by sampling the first 100 rows. Numeric columns become metrics; non-numeric columns become breakdowns.
How to read the columnsThe Data type column uses the platform’s vocabulary: Number, Currency, Percentage, Ratio, Duration, Date, Text, URL, Array, Boolean. The Field role column indicates whether the field can serve as a metric, a breakdown, or both.
Monetary values are treated as USDThe connector reports currency as USD regardless of the currency configured in the source CM360 report. Monetary columns are still inferred as numeric (Number). If the underlying report uses another currency, values are not converted — they are labeled USD as-is.
Notes on inference:
  • Inference is based on the sampled values, not on any CM360 field metadata. A numeric column that happens to contain non-numeric values in the first 100 rows may be typed as text.
  • A Date column, when the report includes one, is what powers date breakdowns and time-series widgets. Whether a date column is exposed as a dedicated Date field or as Text depends on inference — see the reviewer note. (Inference — confirm.)
  • Internal bookkeeping columns are never exposed as fields.

​
Limitations

Based on the current connector:
  • Generic / free-format data source. CM360 data is viewed in a generic table and is not natively compatible with other connector types. To combine it with data from other connectors, the breakdowns must match, or the data must be aligned through Blend Data settings.
  • Deleted data is not provided. Elements deleted in Campaign Manager 360 are not retained — once an element is removed at the source, its data is no longer available in Adriel.
  • No accumulated history. Each refresh fully replaces the cached snapshot; historical depth is limited to the date range in the CM360 report configuration. Fixed date ranges will freeze the data unless updated in CM360.
  • 50,000-row cap per query. Each query is hard-capped at 50,000 rows regardless of any higher requested limit. Reports that produce more rows are truncated at read time.
  • Report configuration drives stability. Adding unnecessary breakdowns inflates the report’s row count, which can slow refreshes and processing. Configuring the report with only the needed breakdowns is recommended.
  • CSV format only. Reports not exported as CSV cannot be connected.
  • Up to 100 reports per profile listed. If a profile has more than 100 reports, the oldest are not shown in the selection list.
  • USD only. Currency is fixed to USD (see the Field type mapping note).
  • Floodlight/Activities validation caveat. A report whose metrics live only under Floodlight or Activities criteria may pass initial setup validation but fail the periodic permission check, which can disconnect the data source. (Inference from code docs — confirm customer-facing impact.)

​
API references

​
See also