Skip to content
Performance Marketing Hub
Analytics, Tracking & Privacy

Google Analytics Cross-Channel Budgeting: Plan Spend by Ad Platform, Not Channel Group

Google Analytics cross-channel budgeting now supports source platform dimensions, so projection and scenario plans can compare Meta Ads, TikTok Ads or Microsoft Advertising side by side instead of broad channel groups. It plans spend but moves no budget, and its suggestions need a check against platform data and lift tests.

Maxim Baeten
By Freelance digital performance marketer
Updated 9 min read

Key takeaways

  • Google Analytics cross-channel budgeting supports source platform dimensions since November 19, 2026, so plans can compare individual ad platforms such as Meta Ads and TikTok Ads instead of channel groups.
  • Projection plans track in-flight pacing against a Budget, Conversions or Revenue target, while scenario plans model returns at different budget levels for future dates and stay static once created.
  • The DDA-based budgeting model needs at least one year of campaign data from at least two channels, Google and non-Google, plus one year of data for each web conversion used.
  • GA4 assigns non-Google traffic to a Source Platform only when the medium is paid, so untagged or organically tagged ad clicks end up as Unlabeled and drop out of the platform view.
  • Cross-channel budgeting is a planning tool that changes no budgets in connected accounts, and a suggested shift should pass a check against platform data and a holdout or lift test before money moves.

Google Analytics cross-channel budgeting now lets projection and scenario plans compare individual ad platforms, such as Meta Ads, TikTok Ads and Microsoft Advertising, instead of broad channel groups like Paid Social. It is a planning tool: it estimates pacing and returns per platform, but it moves no budget, so its suggestions still need a check before you act.

This explainer is part of the analytics, tracking and privacy hub and reflects the state as of November 2026. The main recommendation: clean up your paid medium tagging first, because a platform GA4 cannot recognize does not appear in the plan, and treat every suggested budget shift as a hypothesis to test, not an instruction.

What is Google Analytics cross-channel budgeting and what changed in November 2026?

Google Analytics cross-channel budgeting is a set of planning tools in the Advertising section of GA4 that estimates how paid spend turns into conversions and revenue across channels. Since November 19, 2026, it supports source platform dimensions, so a plan can show each ad platform on its own line instead of one channel group.

What is officially confirmed on Google's About cross-channel budgeting page, as of November 2026:

  • Two tools. Projections and Scenarios, both under Advertising, then Budgeting, on desktop only.
  • New dimensions. Source grouping and source platform dimensions are listed as "launching", and a source platform x campaign type dimension as alpha-only.
  • Planning only. Google states the tools "are for planning purposes only and won't affect budgets or spend in your connected accounts."

What is reported: PPC Land wrote on November 20, 2026 that the Google Analytics LinkedIn page announced the change on November 19, promising advertisers can "Move past broad channel groups to evaluate individual ad platforms side by side." The same post said you can shift budget "in real time", which PPC Land notes sits uneasily with the help page's weekly pacing. Google's "What's new" page had no entry for it as of November 20, 2026.

What do projection plans and scenario plans each answer?

A projection plan answers whether live campaigns are on pace for a target this period, and which platforms are over or under delivering. A scenario plan answers how to split a future budget, by modeling the return at different spend levels. Projections update as data arrives; scenarios freeze at the moment you create them.

Google's About cross-channel budgeting page describes the projection chart as a solid line for actual performance, a dotted line for projected performance and a shaded band for model confidence. Wide bands mean patterns changed and the model needs more data.

CriterionProjection planScenario plan
Question it answersAre live campaigns on pace?How should a future budget be split?
Target KPIBudget, Conversions or RevenueConversions or Revenue, one conversion at a time
DatesCurrent, in-flight periodFuture start dates only
OutputActual vs projected line, confidence band, most and least efficient channelsResponse curve, suggested allocation, projection without changes
UpdatesRefreshes with new data, reviewed weeklyStatic; create a new plan for new results
PathAdvertising, Budgeting, Projection, Create planAdvertising, Budgeting, Scenarios, Create plan
Projection plans vs scenario plans in Google Analytics cross-channel budgeting, as of November 2026 (Analytics Help).

A response curve is a chart that shows the estimated return for each extra unit of spend on a channel or platform. With source platform dimensions, each platform gets its own curve.

Which GA4 properties can use cross-channel budgeting?

A GA4 property can use cross-channel budgeting when the Advertising section is set up, Google Ads is linked, conversions are configured and cost data from at least one non-Google platform is imported. The property also needs enough history: the DDA-based model asks for a full year of campaign and conversion data before it builds a plan.

The requirements on Google's DDA-based budgeting page, as of November 2026:

  • Campaign data. At least one year of campaign data from at least two channels, and those must include Google and non-Google.
  • Conversions. At least one year of data for each conversion you plan on. Web conversions qualify; app conversions are not yet supported, and key events are not supported.
  • Model check. The evaluation of cost import and model quality "can take up to two days."
  • Access. Google says the feature "may not be available for every property" while access expands.

Cost data for other platforms comes in through campaign data import. Meta, Pinterest, Reddit, Snap and TikTok have direct connectors that run daily; everything else arrives by file or database upload. Each row needs Source (utm_source) and Medium (utm_medium), and a currency whenever cost is included.

Models fail when history is too short, third-party cost data is missing, or predictions diverge from observed data.

How do the DDA-based and Meridian-based budgeting models differ?

The DDA-based model is in beta for Google Analytics and Google Analytics 360 and uses data-driven attribution as its main input. The Meridian-based model is in alpha for select Google Analytics 360 customers only, uses marketing mix modeling instead of attribution and needs two years of data instead of one.

Data-driven attribution is an attribution model that splits conversion credit across touchpoints based on how each one changed the chance of converting. The attribution models guide covers what GA4 still offers. Because DDA works on click paths that GA4 can see, the budgeting help page says the models consider click-through conversions, YouTube engaged-view conversions and all cost data. Impressions from data imports are recommended for future updates, so view-based effects from other platforms are not yet a model input.

Marketing mix modeling (MMM) is a statistical method that estimates how spend in each channel affects a KPI from aggregate trends over time. Google's Meridian-based budgeting page calls Meridian "an open-source marketing mix modeling (MMM) framework developed by Google" and states "It does not use attribution as an input." It requires at least 2 years of campaign cost data and 2 years of conversion and revenue data. Google's May 13, 2026 announcement positions Meridian budgeting as an Analytics 360 feature for 2026.

How does GA4 decide which ad platform a session belongs to?

GA4 fills the Source Platform dimension from source and medium together. Google traffic is mapped by its inventory to values such as Google Ads or Display & Video 360. For other traffic, GA4 first checks whether the medium is paid, then maps the source to a platform; without a paid medium the session becomes Unlabeled.

That logic, from Google's About the Source Platform dimension page as of November 2026, makes UTM tagging a precondition for platform-level budgeting. A medium counts as paid when it contains cp (cpc, cpm), equals ppc, retargeting, display or appremarketing, or starts with paid. Values for other platforms include Meta Ads, TikTok Ads, Microsoft Advertising, Pinterest Ads and LinkedIn Ads.

A Meta ad tagged with utm_medium=social lands in Unlabeled, and its spend has no platform line to match. Fix the tags before you build a plan:

  1. Export source and medium for paid traffic. Use the last 13 months, so the one year training window is covered.
  2. List every paid medium value in use. Flag values such as social, influencer or video that GA4 does not treat as paid.
  3. Standardize the medium. Use cpc or a paid_ prefix per platform, following one UTM builder naming convention.
  4. Match source names to the import. The utm_source in ad URLs must equal the source in campaign data import rows.
  5. Check Source Platform after a week. Confirm each platform has its own value and Unlabeled paid spend is near zero.

Google does not document whether corrected tags relabel past sessions, so expect the model to improve only as clean months accumulate.

What does cross-channel budgeting not do?

Cross-channel budgeting does not move money, prove causality or settle disputes between platforms. It produces estimates from GA4's own data and a model Google builds, and every change still has to be made by hand in Google Ads, Meta Ads Manager or any other ad account. Its outputs are directional, not guarantees.

Four limits matter for budget decisions, as of November 2026:

  • No budget changes. Google's help page says the tools "won't affect budgets or spend in your connected accounts."
  • Web conversions only. App conversions are not yet supported in the DDA-based model, per the DDA-based budgeting page.
  • Static scenarios. A scenario plan reflects conditions at creation, so a new auction price or creative shift is invisible until you build a new plan.
  • Open questions. PPC Land found no list of recognized source platforms for budgeting, no rollout timeline and no note on whether existing plans gain the new dimensions.

There is also a structural point. The tool is built by Google and, in its DDA version, trained on attribution data GA4 collects, while it ranks Google Ads next to Meta Ads or TikTok Ads. That does not make the output wrong, but it makes an outside check part of the job. The GA4 conversion discrepancy guide explains why GA4 and each platform already count conversions differently.

Should you act on a suggested budget shift from GA4?

Act on a suggested budget shift only when it passes a short check: the platform is cleanly tagged, the model confidence is acceptable, the platform's own data points the same way, the shift is small enough to reverse and a holdout or lift test can confirm it. A suggestion that fails any of these is a hypothesis, not a plan.

The Budget Shift Gate is the author's five-question rule of thumb, not Google guidance. Answer each with yes or no:

  • Clean source? Paid spend for the platform shows its own Source Platform value, with little Unlabeled spend.
  • Narrow band? The projection's confidence band is not widening for that platform.
  • Same direction? The platform's own reporting shows the same trend in cost per conversion over the same weeks.
  • Reversible size? The move is a step you can undo within one planning period, not the full suggested allocation.
  • Testable? You can run a holdout or Google Ads conversion lift study before scaling.

Five yes answers on the Budget Shift Gate: move the step and measure. Any no: fix that item first.

Hypothetical example, invented numbers: a scenario plan suggests moving 20 percent of a $50,000 quarterly budget from Google Ads to TikTok Ads. Unlabeled paid spend is low and the band is narrow, but TikTok Ads Manager shows cost per purchase rising. That is one no on the Budget Shift Gate, so fix it or run a holdout before any money moves. Benchmarks such as Google Ads cost and Facebook ads cost show whether implied prices are realistic.

Common mistakes with GA4 cross-channel budgeting

The common mistakes with GA4 cross-channel budgeting come from treating a planning estimate as a measurement, or from feeding the model data it cannot read. Most of them are visible before you build a plan: messy mediums, partial cost imports and conversions that do not have a full year of history.

  • Tagging paid social as social. GA4 needs a paid medium to assign a Source Platform. Use cpc or a paid_ prefix for every paid platform.
  • Importing cost without matching UTMs. Cost rows whose source and medium do not match the ad URLs cannot be joined to sessions. Align both before the first plan.
  • Picking a new conversion. The DDA-based model needs a year of data per conversion. Plan on an older conversion until the new one qualifies.
  • Reading a scenario as live. Scenario plans are static. Rebuild the plan after a big change in prices, creative or seasonality.
  • Moving the full suggested allocation at once. A large move removes the baseline you need to judge it. Shift in steps and test each one.
  • Changing channel groupings mid-plan. Google's DDA-based budgeting page warns that edits to product links, campaign data import or channel grouping can affect models. Schedule those changes between planning periods.

FAQ

Does Google Analytics cross-channel budgeting change my budgets in Google Ads or Meta?

No. Google states that projections and scenarios are for planning only and do not affect budgets or spend in connected accounts. Any shift it suggests has to be made by hand in each ad platform.

Do I still need a full year of imported spend data to use it?

Yes, for the DDA-based model. Google requires at least one year of campaign data from at least two channels, Google and non-Google, and one year of data for each conversion. The Meridian-based model in alpha needs two years.

Should I use a projection plan or a scenario plan?

Use a projection plan to check whether live campaigns are on pace for a budget, conversion or revenue target this period. Use a scenario plan to decide how to split next quarter's budget. Many teams run both: a scenario before the period starts and a projection while it runs.

Why are my Meta or TikTok ads missing from the budgeting view?

The most common reason is the medium. GA4 maps non-Google traffic to a Source Platform only when the medium is paid, such as cpc or paid_social, and labels the rest Unlabeled. Missing cost data for that platform is the second reason.

Can I trust a budget model built by Google to judge Google Ads against other platforms?

Treat it as one input, not a verdict. The DDA-based model learns from attribution data that GA4 collects, and other platforms run their own models. Before moving budget, compare the suggestion with platform reporting and confirm it with a holdout or conversion lift test.

Sources and further reading

Maxim Baeten
Written byMaxim Baeten

Digital performance marketer from Belgium, specialised in lead generation for B2B (SaaS) software companies. Full-time freelancer since January 2019.