AI Ads strategist + execution layer
NotFair gives your AI live Google Ads and Meta Ads access via MCP to diagnose issues, recommend fixes, and execute campaign changes only after you approve.
Why did leads get more expensive, and what should I fix?
Connect NotFair to Claude, OpenClaw, Codex, Cursor, or Hermes and operate Google Ads or Meta Ads from a chat - diagnose, draft fixes, and approve every write.
Name
NotFair-GoogleAdsRemote MCP server URL
https://notfair.co/api/mcp/google_adsAdd, Claude opens a sign-in flow. Sign in with NotFair and you are all set. Try: Audit my Google Ads account and rank fixes by impact.Your ads operator should feel native inside Claude: live account context, sharp diagnosis, and no writes until you approve the exact campaign change.
Claude + NotFair
Partner-ready ads operations
Built for Claude's connector workflow, not a generic dashboard bolted onto chat.
Approval-gated writes keep every campaign change reviewable before it touches Google Ads or Meta Ads.
The same MCP layer works across Claude, Cursor, OpenClaw, Codex, and Hermes.
Why it matters
Claude can inspect spend, search terms, structure, and recent changes through NotFair, then ask for approval before pausing waste or shifting budget.
Diagnostics, recommended fixes, writes, and undo - all driven by one MCP connection per ad account.
CPL up 42% WoW. Three campaigns drove it:
Brand-Search+$1,820 spend - 0 conversionsPMax-US+$940 - CPA $38 -> $89LeadGen-Q4search terms drifted to "free"Want me to draft 14 negatives + pause Brand-Search?
Your agent reads the same numbers you would open in the platform UI: cost, conversions, impression share, learning phase, frequency, quality score. No CSV exports, no stale dashboards.
Checked spend, search terms, quality scores, and assets. Top fixes by spend at risk:
Total recoverable: $4,530/wk. Open the fix list?
One prompt fans out the reads, scores findings by spend at risk, and turns the result into prioritized actions your agent can draft for approval.
Found 6 ad sets matching. Diff before I apply:
Not a passive reporting tool. Every entity an account manager touches is exposed as a write tool, with a diff-and-approve gate before anything hits the ad platform.
Found change chg_8f21 - paused Brand-Search at 14:32 today.
Reverted. Anything else to roll back?
Reads are unrestricted. Writes show you a diff before they land, and NotFair logs every change to its own history - one call rolls anything back.
MCP Servers
NotFair runs four remote MCP servers. Connect them to Claude, Codex, or Cursor in minutes and your agent can correlate paid spend, organic queries, and GA4 conversions in one conversation.
Spend, search terms, quality scores, and change history — plus keyword, budget, and campaign writes that wait for your approval.
Explore the server
Insights, ad sets, audience overlap, and creative fatigue — pauses and budget shifts are approval-gated.
Explore the server
Top queries, page CTR, average position, index coverage, and sitemaps for every verified property.
Explore the server
GA4 reports over traffic, conversions, and revenue — plus realtime activity from the last 30 minutes.
Explore the server
Affiliate Program
Refer a customer to NotFair. Get half of everything they pay us for their first 12 months. Monthly payouts, no caps.
Pricing
Built for people who want AI to find issues and turn approved recommendations into campaign changes.
Unlimited access for 7 days. 300 free operations every month after.
No credit card required
For operators making real campaign changes.
Prices in USD. Billed via Stripe. Cancel anytime from the billing portal - your access continues until the end of the current period.
Spending $50K+/mo? Let's talk
No. NotFair can read your account freely, but every write tool surfaces a diff first. You confirm before anything hits Google Ads or Meta Ads - and any change is reversible with a single tool call.
Closing
Free to connect - By connecting, you agree to our Privacy Policy and Terms.