What is MCP for paid media?
MCP is the Model Context Protocol — an open standard that lets an AI client call external tools and read external data. Without it, an assistant has no route to any of your ad accounts.
GoMarble MCP is an MCP server that connects supported AI clients to marketing data and workflows in GoMarble. For paid media that means reaching Meta, Google, TikTok, LinkedIn and Bing through a single connection, alongside GA4, Search Console, Shopify and Klaviyo.
The reason this matters is that almost no paid media question is answerable inside one platform. Blended CAC moves without any single channel looking broken. Cutting Meta prospecting raises brand-search CPA on Google a fortnight later. Budget sitting in a saturated campaign on one channel is only obviously wrong next to a constrained campaign on another. GoMarble MCP puts the channels in one context so the assistant can reason across them instead of one at a time.
How to connect paid media to GoMarble MCP
Connecting your paid channels to GoMarble MCP takes a few steps. Connect each platform once, and every one of them becomes reachable from the same conversation.
- Connect each ad platform you run — Meta, Google, TikTok, LinkedIn, Bing — to GoMarble.
- Connect GoMarble MCP to your preferred supported AI client.
- Authorize the access GoMarble requests.
- Choose the GoMarble workspace and accounts the client can use.
- Start asking cross-channel questions or running supported workflows.
Connecting to Claude
The current steps, from the GoMarble MCP page:
- Open Settings, then Connectors.
- Choose Add custom connector.
- Enter https://apps.gomarble.ai/mcp-api/sse, then select Authorize.
- Choose the GoMarble workspace and accounts Claude can use.
Authorization uses GoMarble’s consent and workspace permissions, completed in the browser window your client opens.
Other supported clients
GoMarble MCP is a remote MCP server at https://apps.gomarble.ai/mcp-api/sse, currently supported in Claude, ChatGPT, Cursor, n8n, Hermes and OpenClaw, plus other clients that support compatible remote or local MCP servers. Setup differs slightly per client, so follow the canonical instructions rather than adapting the Claude steps.
View the complete GoMarble MCP setup guide →
What the free plan includes
- Up to $50,000 in cumulative ad spend
- One account per data source
- No edit actions — read and analysis only
What can you do with GoMarble MCP for paid media?
Grouped by the job rather than by the tool that does it.
Analyze
- Analyse performance across every connected channel in one request
- Compare channels on spend, CAC, ROAS and conversion volume
- Compare periods across the whole paid portfolio
- Break down blended performance by channel contribution
- Read each platform's own metrics without switching context
Diagnose
- Investigate a blended CAC or ROAS move and attribute it to channels
- Separate a real efficiency change from a budget mix shift
- Find which channel absorbed a budget increase without returning on it
- Read change history across platforms to correlate edits with shifts
- Identify channel interactions — where a cut on one raised cost on another
Creative intelligence
- Compare creative performance across paid social channels
- Identify which angles travel between platforms and which do not
- Analyse creative fatigue signals per channel
- Research competitor advertising through the ads library
Research
- Search brands and find competitors
- Pull competitor ads and brand-level analytics
- Discover keyword ideas and demand signals on the search side
Recommend
- Recommend where the next increment of budget should go
- Identify constrained versus saturated campaigns across channels
- Surface channels worth testing more or scaling back
- Flag pacing problems against targets
Act
- Propose budget and bid changes on supported platforms
- Propose pausing or scaling campaigns
- Propose new campaign structures
- Every write is a proposal that routes through your workspace's approval gates
GoMarble MCP capabilities for paid media
| Capability | What GoMarble MCP can do |
|---|---|
| Cross-channel analysis | Analyse Meta, Google, TikTok, LinkedIn and Bing together in one request |
| Channel comparison | Compare channels on spend, CAC, ROAS and conversion volume |
| Blended diagnosis | Attribute a blended CAC or ROAS move to the channels responsible |
| Mix-shift detection | Separate genuine efficiency change from budget moving between channels |
| Budget allocation | Identify constrained versus saturated campaigns across the portfolio |
| Creative comparison | Compare creative performance across paid social platforms |
| Competitor research | Search the ads library and analyse competitor advertising |
| Pacing | Track spend against targets across channels |
| Actions | Propose budget, bid and status changes, subject to approval gates |
What data can GoMarble MCP access?
The data available for paid media, grouped by type.
Meta Ads
Campaigns, ad sets, ads · Creatives and copy · Audiences and targeting · Spend, ROAS, CPA, CPM, CTR, frequency · Change history
Google Ads
Campaigns, ad groups, keywords · Search terms · Assets and PMax asset groups · Spend, conversions, CPC, impression share · Change log
TikTok Ads
Campaigns, ad groups, ads · Creative reports · Audience reports · Spend and performance
LinkedIn Ads
Campaign groups and campaigns · Ad analytics · Targeting entities
Bing Ads
Campaigns, ad groups, ads, keywords · Budgets · Performance · Change history
Downstream context
GA4 traffic and behaviour · Search Console queries · Shopify orders and revenue · Klaviyo campaigns and flows
Competitor data
Brand search · Competitor discovery · Competitor ads · Brand analytics
GoMarble MCP workflows for paid media
Jobs marketers actually want done, rather than features. Each one is a single natural-language request.
Explain a blended CAC move
“Our blended CAC went up this month. Which channel caused it?”
What GoMarble MCP does: Pulls spend and conversion data from every connected channel for both periods, recomputes blended CAC, then holds the mix constant to separate genuine efficiency change from budget shifting between channels.
What you get: Whether the move was real efficiency loss or a mix effect, and which channels contributed how much.
Decide where the next budget goes
“I have another $20k this month. Where should it go?”
What GoMarble MCP does: Reads delivery and performance across channels to find campaigns limited by budget rather than by audience, and checks frequency and conversion trends for saturation signals.
What you get: The campaigns with genuine headroom, rather than simply the ones with the best average return to date.
Compare channels fairly
“Compare Meta and Google performance this quarter and tell me which is working harder.”
What GoMarble MCP does: Pulls both channels on a consistent window, normalises for their different funnel roles, and reads GA4 and store data alongside where connected.
What you get: A like-for-like comparison with the attribution caveat stated rather than hidden.
Catch a channel interaction
“We cut Meta prospecting two weeks ago. What happened to the rest of the account?”
What GoMarble MCP does: Reads the Meta change history for the cut, then tracks brand search volume, Google CPA and direct traffic in the period after it.
What you get: Whether the saving was real or simply moved cost downstream — the effect single-channel analysis always misses.
Find portfolio-wide waste
“Where am I spending without getting anything back, across all channels?”
What GoMarble MCP does: Scans every connected platform for line items with meaningful spend and weak conversion contribution, then sizes each in money.
What you get: A ranked list across channels rather than five separate audits.
Check pacing against target
“Are we on track to hit this month's spend and CAC targets?”
What GoMarble MCP does: Reads spend to date per channel, projects against the remaining days, and compares CAC against the target.
What you get: Where pacing is off and which channel is driving the variance.
Compare creative across paid social
“Which creative angles are working on Meta but not on TikTok?”
What GoMarble MCP does: Reads creative-level performance on both platforms and compares by angle, format and hook where the data supports it.
What you get: Which concepts travel between platforms and which are platform-specific.
Produce a cross-channel summary
“Summarise all paid performance this week.”
What GoMarble MCP does: Pulls every connected channel for the period, compares against the prior one, and surfaces the notable movements with their causes.
What you get: One summary covering the portfolio instead of a stack of per-platform exports.
Example GoMarble MCP prompts for paid media
Things real marketers ask, grouped by intent. Paste any of them once connected.
Cross-channel performance
- “How did paid perform across all channels this month?”
- “Compare Meta and Google on CAC this quarter.”
- “Which channel is delivering the best return right now?”
- “Show me spend by channel for the last 30 days.”
Diagnosis
- “Why did blended CAC increase this month?”
- “Is that a real efficiency change or just a mix shift?”
- “Which channel absorbed the budget increase?”
- “Did anything change across the accounts before performance moved?”
Budget allocation
- “Where should the next $20k go?”
- “Which campaigns are budget-constrained across all channels?”
- “Which campaigns look saturated?”
- “Where is my budget allocation inefficient?”
Waste
- “Where am I spending without conversions across all channels?”
- “Which channel has the most recoverable waste?”
Creative
- “Which creative angles work on Meta but not TikTok?”
- “Compare creative performance across paid social.”
Competitor research
- “Who else is advertising in my category?”
- “What has my main competitor been running longest?”
Pacing and reporting
- “Are we on track against this month's spend target?”
- “Summarise all paid performance this week.”
- “Build me a channel-level breakdown for the board.”
Actions
- “Propose moving budget from the saturated campaigns to the constrained ones.”
- “Propose pausing the worst-performing line items across channels.”
Why use GoMarble MCP for paid media?
An API gives AI access to data. GoMarble MCP gives it the marketing context and workflows needed to work with that data.
One connection instead of five
Wiring Meta, Google, TikTok, LinkedIn and Bing individually means five auth flows, five schemas and five sets of quirks. GoMarble MCP is a single connection that reaches all of them.
The channels arrive comparable
Each platform reports differently. Getting to a genuine comparison is most of the work, and it is work a raw API leaves entirely to you.
Interactions become visible
Channel effects on each other — the thing that most often explains a blended move — are invisible when each account is queried in isolation.
Downstream context is included
GA4, Search Console, Shopify and Klaviyo sit on the same connection, so paid results can be read against site behaviour and actual revenue.
Writes are governed
Changes are proposed and routed through your workspace's approval gates rather than issued directly against five different APIs.
GoMarble MCP for cross-platform marketing analysis
Marketers rarely decide anything from one platform. GoMarble MCP can give AI a view of the wider marketing system rather than forcing it to reason from one disconnected source at a time.
| Combination | What becomes answerable |
|---|---|
| Meta + Google | The core comparison for most advertisers, including the interaction where cutting social prospecting raises brand-search cost. |
| Paid + GA4 | Join ad delivery to on-site behaviour, which separates a traffic-quality problem from a landing-page problem. |
| Paid + Shopify | Put channel spend against store revenue and orders rather than platform-reported conversions that double-count. |
| Paid + Search Console | See where paid and organic overlap on the same queries, and where you are paying for terms you already rank on. |
| Paid + Klaviyo | Check whether the customers a channel acquires actually stay and repeat. |
How does GoMarble MCP work?
| Step | What happens |
|---|---|
| 1 | You ask a question in natural language |
| 2 | The AI client works out what information it needs |
| 3 | GoMarble MCP retrieves the relevant marketing data and tools |
| 4 | GoMarble supplies the marketing context around it |
| 5 | The AI analyses the result |
| 6 | You get an answer, a recommendation, a report — or a proposed action |
Write access is plan-dependent. On the read-only plan you can connect a client and analyse supported accounts without making changes. Where write access is enabled, changes are proposed rather than applied directly — every write tool in the server is a propose call, and consequential actions route through the permissions and approval gates configured for your workspace.
Who is GoMarble MCP for?
Performance leads running several channels
Answer a portfolio question in one place instead of assembling five exports.
Growth teams managing blended targets
Explain a blended CAC or ROAS move to the people who only see the blended number.
Media buyers splitting budget
Find genuine headroom rather than moving money toward whatever had the best average return.
Agencies running multi-channel accounts
Standardise the same cross-channel investigation across clients.
Founders and CMOs
Understand what paid is doing overall without learning five reporting interfaces.
GoMarble MCP vs traditional paid media workflows
| Traditional workflow | With GoMarble MCP |
|---|---|
| Open five platforms and export each | Ask once, across all of them |
| Reconcile channels in a spreadsheet | Compare them in the conversation |
| Guess at channel interactions | Ask what happened downstream after a change |
| Allocate budget on average return | Ask which campaigns have real headroom |
| Rebuild the weekly cross-channel report | Re-run the same prompt |
| Apply changes platform by platform | Propose them together, then approve |
Frequently asked questions
What is MCP for paid media?
Which ad platforms does GoMarble MCP support?
Can GoMarble MCP compare channels?
Can it explain a blended CAC change?
Can it tell me where to put more budget?
Why don't my channel numbers add up?
Can GoMarble MCP make changes across channels?
What are the free plan limits?
Which AI assistants work with GoMarble MCP?
How do I connect my channels?
Do I need to connect every channel?
How is this different from a reporting dashboard?
Why GoMarble MCP?
The work you stop doing:
- Stop exporting five platforms to reconcile them in a spreadsheet
- Stop explaining a blended number you cannot yet attribute
- Stop allocating budget on average return instead of headroom
- Stop missing the channel interactions that explain the move
- Stop rebuilding the same cross-channel report every week
Connect your marketing stack once. Give AI the context it needs. Ask in plain English, and let it analyse, explain, recommend and — where enabled — propose the change.
Your paid media data is already there. Now give AI access to it.
Connect GoMarble MCP to your marketing stack and start analysing paid media from the AI tool you already use.