Google Analytics released a Campaign data import validation report on August 10, 2026, giving advertisers a row-by-row readout of whether uploaded cost, click and impression figures actually attached to sessions the platform measured.

The report appeared in the Google Analytics Help Center release notes on Monday, August 10, alongside a dedicated support article describing its columns and calculations. According to Google Analytics documentation, the report exists to help operators assess non-Google campaign data and imported campaign data, and to pinpoint campaigns that lack useful performance data such as cost, clicks and impressions.

That framing matters more than it first appears. Campaign data import has been the mechanism through which spend from platforms outside Google enters Analytics, and the numbers it carries feed non-Google cost-per-click and return-on-ad-spend calculations. Until now, the only health signals available sat at the level of the import job itself, not the campaign. A file could report a healthy import percentage while individual campaigns inside it joined to nothing at all.

Three statuses, three different failures

The report lives under Reports, in the Data Import section, listed as Campaign data import validation report. Documentation notes that the report may not appear in a property's default view, and that editors and above can add it back to the left navigation.

Its central column is the import join status, described in the documentation as the validation status column. Every row resolves to one of three values, and each describes a materially different condition.

Joined indicates that imported campaign data attached successfully to Google Analytics data for the selected dimension. No campaign data covers rows where campaign data was not imported for that dimension, or was imported but not joined correctly. The third value, No Analytics data, describes the case where, according to Google's documentation, imported data "was not joined and will be reported individually without Analytics data."

The distinction between the second and third statuses is the analytically useful part. In one case the spend record is missing; in the other the spend record exists and the behavioural record does not. Both produce a gap, but the gap sits on opposite sides of the join. A campaign flagged No Analytics data will still contribute cost, clicks and impressions to reports, which means blended efficiency metrics continue to compute over a denominator that never received the corresponding sessions.

The status itself is derived rather than declared. According to the documentation, it is populated based on the Ads Clicks, Ads Cost or Ads Impressions value being greater than zero, either from linked integrations or from an explicit campaign data import.

Coverage rate flips the denominator

The second new metric is coverage rate, which the documentation defines as a measure of the completeness of imported campaign data, expressed as the ratio of ad campaign data that has associated campaign performance data.

The calculation is specified precisely: the sum of successfully joined status rows divided by the number of rows with collected Analytics data.

That formula is worth setting against the two health metrics that already existed on the data import details page. Percentage imported divides successfully imported rows by the total rows in the import file, and answers a question about file integrity. Match rate divides imported rows that joined by imported rows overall, restricted to the most recent two years of imported data, and answers a question about how much of an upload found a home.

Coverage rate reverses the reference point. Its denominator is Analytics traffic, not the upload. A property can therefore post a high match rate and a low coverage rate simultaneously: every uploaded row joined, but the upload covered only a fraction of the paid sessions Analytics observed. The reverse also holds. For teams reconciling media plans against measured outcomes, those are two separate diagnoses with two separate remedies, and the platform now exposes both.

Neither the release note nor the support article reconciles the two figures explicitly, states whether coverage rate inherits the same two-year window that constrains match rate, or specifies the retention period over which the validation report itself holds data.

Five dimensions, seven metrics

The validation collection ships with a fixed set of grouping dimensions: Source/Medium as the default, Campaign ID, Campaign name, Source platform, and Default channel group or primary channel group. All five are populated through traffic-source dimensions, manual tagging and auto-tagging, which is to say through the same UTM apparatus that governs the join.

Alongside import join status and coverage rate, the metric set includes event count, ad clicks, ad cost, ad impressions and key events. Ad clicks, ad cost and ad impressions populate automatically once Google Ads and Analytics are linked and campaign data is imported. Key events populate once an event has been marked as a key event, and can be selected individually to narrow the analysis.

A metric defined two ways in one document

One definitional inconsistency sits inside the support article. In the passage introducing the metric set, the documentation offers Event count as its example and describes it as the total number of times an event is triggered. In the metrics table further down the same page, the entry for Event count reads: "The average number of events per session."

A total and a per-session average are not the same quantity, and the difference is not cosmetic in a report whose entire purpose is diagnostic precision. Google has not published a clarification. Analysts comparing event volume across campaigns with sharply different session counts would arrive at different conclusions depending on which definition the report actually implements.

The default filter excludes Google traffic

An easily missed note carries operational weight. The report is session scoped, and it arrives with a filter applied by default that restricts the view to paid manual campaigns, "excluding all google sources and search mediums." The filter is removable, and removing it widens the report to all traffic in the property.

The default configuration therefore points the report at exactly the territory campaign data import was built for: spend that Google does not auto-tag. Auto-tagged Google Ads traffic sits outside the default view until an analyst deliberately clears the filter. The phrase "manual campaigns" narrows it further, to sessions carrying hand-applied or platform-applied UTM parameters.

Why joins fail in the first place

The mechanics behind a failed join are documented elsewhere in the same help centre, and they explain most of what the new report will surface.

Campaign data joins to Analytics data on three required keys: utm_source, utm_medium and date. Values in the imported file must match the values users actually logged, and the matching is literal. Google's documentation states plainly that "Facebook" and "facebook" are not exact matches. Dates must follow ISO 8601 format, must be later than 2013, and cannot sit in the future. Since a change on July 28, 2026, any import carrying cost figures must also specify a currency, either mapped from a column in ISO 4217 format or hardcoded as a static value applied to the whole dataset.

Tagging architecture determines how much of this is survivable. Documentation sets out three configurations. Using a distinct utm_source per platform, such as fb and insta, is described as best practice and produces the most granular reporting. Using one unified value across an entire platform family, such as meta for Facebook, Instagram, Messenger and Threads, is described as acceptable, with a caution that no channel be left blank in the import source configuration. The third pattern requires intervention: where a single platform emits multiple pairs, such as fb/cpc, Meta/cpc and facebook/paid at the same time, automated third-party integrations do not support multiple values for one platform, and the documentation directs those advertisers toward CSV, Google Sheets or BigQuery connectors instead.

Naming drift compounds the problem. Inconsistent campaign values such as SpringSale and Spring_Sale cause Analytics to treat one marketing effort as two. Dynamic UTMs are recommended at the ad platform level because they eliminate manual entry error, but the import interface itself does not accept dynamic parameter syntax during mapping, so those placeholders have to resolve before data reaches Analytics.

Where the report sits in the import stack

Campaign data import currently accepts data from eleven storage and database sources: Amazon Redshift, Amazon S3, BigQuery, CSV, Google Cloud Storage, Google Sheets, HTTPS, MySQL, PostgreSQL, SFTP and Snowflake. Five advertising platforms connect directly, running on an automatic daily schedule: Meta, Pinterest, Reddit, Snap and TikTok. Those direct connections attempt to pull twenty-four months of historical data on setup, and each requires that all paid traffic from the selected source carry a single unique utm_source and utm_medium value.

Because campaign data import uses a reporting and query time join rather than altering processed data, deleting an import source removes the imported campaign data from reports without further reprocessing. Uploaded data takes up to twenty-four hours to become available in reports, audiences and explorations, and users must engage with a campaign after upload for the metrics and campaign properties to associate with their activity.

The infrastructure has been assembled in stages. Google added nine new data sources in June 2025, connected Snap Adsin September 2025, and brought Meta and TikTok online on October 7, 2025. The feature was renamed from cost data import to campaign data import on November 19, 2025.

A validation layer for features that assume clean input

The August 10 report is the second data-quality release inside two weeks and the fourth this year aimed at the accuracy of traffic-source records rather than at new measurement capability.

On July 30, 2026, Google added a diagnostic that flags properties where GBRAID and gad_ parameters are absent from URLs carrying a Google click identifier, a condition that pushes paid sessions into organic or (not set) buckets. On June 11, 2026, the platform introduced Source Group, consolidating variant source strings such as facebook, fb and Meta-facebook into single reporting values. On May 13, 2026, an AI Assistant channel began classifying referrals from chatbots automatically. Each addressed a different way in which a session can end up filed under the wrong label.

The validation report targets the remaining case, where the session is labelled correctly and the money attached to it is not.

The dependency chain runs in one direction. Cross-channel budgeting, launched in beta on January 16, 2026 alongside improved web conversion management and a conversion attribution analysis report, generates projection and scenario plans from imported cost data. Projections built on a campaign set with a low coverage rate inherit that gap without displaying it. Task Assistant, released on April 29, 2026, walks property owners through the configuration steps those features rest on, and conversion reporting reached the Data API in alpha on May 4, 2026, exposing the same numbers to programmatic pipelines.

Each of those surfaces consumes campaign data import output. None of them previously offered a campaign-level view of how much of that output was real.

What the announcement leaves open

Several questions sit outside the published material. Google has not stated whether the validation report is available across all property tiers or restricted in any way, has not described any alerting or scheduled export behaviour, and has not indicated whether the report treats the five automated third-party connections differently from manual file uploads, despite those connections carrying different failure modes. No threshold is offered for what constitutes an acceptable coverage rate, and no historical backfill window is specified.

The support article points to separate guidance on improving join status or coverage rate, which suggests remediation documentation exists as its own resource rather than inside the report.

Why it matters

For media buyers running spend across platforms Google does not own, the practical significance is that a reconciliation task previously performed in spreadsheets now has a native surface. Discrepancies between platform-reported spend and Analytics-reported non-Google cost have historically been investigated by exporting both and comparing them by hand, a process that identifies the total mismatch but rarely isolates which campaign caused it.

For publishers and agencies reporting blended efficiency to clients, the coverage rate creates a disclosable figure. A return-on-ad-spend number calculated across campaigns with partial coverage is not wrong so much as incomplete, and the difference has been difficult to quantify without access to the join logic.

The wider point concerns the direction of Google's measurement releases through 2026. The platform has spent the year adding classification layers, standardisation fields and diagnostics rather than new metrics. That sequence suggests the constraint on cross-channel measurement inside Analytics is no longer the absence of data but the reliability of the joins connecting it, and the validation report is the first instrument Google has shipped that measures those joins at the level where budget decisions are actually made.

Timeline

Summary

Who: Google, through the Google Analytics Help Center, addressing property owners, media buyers, agencies and publishers who import campaign data from advertising platforms outside Google, including Meta, Pinterest, Reddit, Snap and TikTok.

What: A Campaign data import validation report adding an import join status column with three values - Joined, No campaign data and No Analytics data - and a coverage rate metric calculated as successfully joined rows divided by rows with collected Analytics data. The report ships with five dimensions and seven metrics, is session scoped, and applies a removable default filter limiting the view to paid manual campaigns while excluding Google sources and search mediums.

When: Published August 10, 2026, in the Google Analytics release notes, thirteen days after the July 28 currency requirement and eleven days after the July 30 aggregate identifier diagnostic.

Where: Inside Google Analytics under Reports, in the Data Import section. Properties that do not display the report in the left navigation can have it restored by users with editor access or above.

Why: Campaign data import feeds non-Google cost-per-click and return-on-ad-spend calculations and supplies the cost figures behind cross-channel budgeting projections, yet the existing health metrics measured the import job rather than the campaign. The validation report identifies which individual campaigns carry spend that never joined to measured behaviour, and quantifies how much of a property's paid traffic has associated performance data.