On-demand course: B2B ad campaigns with Claude Code and n8n
Ship ICP-driven ad campaigns and automate monitoring using GitHub, Claude, and n8n
Use GitHub & Claude Code to turn ICP research into scored, visualized ad variants and automate daily campaign monitoring through n8n.
- Set up a GitHub repo with ICP data, copy variants, and performance logs as a working system
- Extract hooks, objections, and proof points per segment using Nick Christensen ICP visualization approach
- Score copy against ICP criteria inside Claude before uploading to your ad platforms
- Pull daily spend, CPA, CTR, and CPL from Google and Meta via API into a Slack digest
- Log every campaign decision to GitHub as a traceable, searchable record
On demand course included when you join.
Black Friday offer: 25% off, already applied. A year of CXL, January cohort included.
AppSumo and CXL instructors: Scaled growth to $90M, scaled ad spend to 500k per month and produced industry recognized content. +25k students trained from the likes of Unilever, Red Bull and Heineken.
From CXL, trusted by 90,000+ marketers at companies like these:
One marketer with the right setup covers what used to need a copywriter, an analyst, and a media buyer.
Most demand gen teams still write copy without a systematic connection to ICP research, and check campaign performance by logging into dashboards manually.
The Google Ads and Meta Ads APIs combined with n8n changed what’s possible.
You can receive live campaign data and surface decisions without a CSV export or a dashboard build. Teams not adapting are monitoring campaigns slower and writing copy with less signal than their competitors.
This course ships both ends: ICP-driven copy and automated daily monitoring.
Black Friday offer: 25% off, already applied. A year of CXL, January cohort included.
Part 01: Build an ad copy scoring dashboard from ICP research
Build a GitHub-based ad operations workflow and a Claude Code dashboard where ICP research becomes scored copy variants for Google RSA and Meta.
- Set up the GitHub repo (/icp-data, /copy-variants, /performance-logs) and load ICP research from surveys or public signals
- Iterate on the ICP profile and extract language relevant to paid ad copy using Claude Code
- Build a Claude Code artifact dashboard: surface hooks, objection lines, and proof points per segment
- Generate copy variations per segment inside the dashboard, structured for Google RSA and Meta
- Score all variations against ICP criteria within the dashboard before exporting to the platform
Outcome:
- Claude Code artifact: ICP visualization dashboard for ad copy scoring and variant generation
- GitHub repo starter with /ICP-data, /copy-variants, /performance-logs and README
- Claude Code prompt: ICP data to segment copy brief, parameterized for Google RSA and Meta
- Scored ad copy variants, generated and scored inside the dashboard, ready to upload
Prerequsite tools:
- Claude Pro, GitHub (free), n8n (free trial account)
Part 02: Automate daily spend decisions: Google Ads and Meta via n8n
Connect Google Ads and Meta Ads to n8n for daily performance monitoring and Slack-delivered optimization suggestions.
- Connect Google Ads and Meta Ads to n8n via APIs: pull spend, conversions, CPA, CTR, CPL, and frequency
- Build a daily performance summary that surfaces decisions (scale, pause, test) in plain language
- Run ICP data from part 1 through the daily digest: flag when performance drops below ICP-aligned benchmarks
- Automate the full flow: trigger at 8am, pull from both platforms, run analysis, send to Slack
- Commit daily decision summaries to GitHub as a traceable campaign decision log
Outcome:
- n8n workflow JSON: daily Google Ads and Meta Ads performance digest via API, with optimization suggestions and Slack delivery
- Daily spend decision framework: scale / pause / test with reasoning, connected to ICP benchmarks from part 1
Prerequisite tools:
- Claude Pro, GitHub (free), n8n (free trial account)
Marketing Engineer: eight-week cohort, starts 11 January
Live Tuesdays, 3 PM UTC, to 9 March. With Diane Wiredu, Ann Smarty, Tyler Durman and Nick Christensen. Included in the annual plan.
Meet your instructors
Scaled AppSumo’s paid growth engine from $7M to $90M. Helps you fix tracking, lower CAC, and grow paid acquisition.
This is what marketers said after our live workshops and courses
B2B eCommerce platform
Sales signed off the target in writing before the brief passed Gate 1. Five review rounds took the drafts to 87% and 96%.
Campaign engine
Co-Founder, Feral MKT
“That alone used to take me a whole day working in AI-assisted mode (more days without AI), since I had to write all the copy from scratch because I didn’t like what Claude used to give me. This time it took me about an hour.”
Campaign engine
Fractional CMO & GTM Strategist
“I fixed the site and Crunchbase last week and reran Build 1 this morning with the same questions, tools and judge. Accuracy went from 53% to 71%, and wrong answers went from three to none.”
AI visibility tracker
Social club for people 60+, pre-launch
A paid first dinner, to test whether older adults will pay for it themselves. All four channels reached ship after 9 fixes.
Campaign engine
Senior copywriter, elective healthcare agency
“I was really impressed with how it handled the copy, including adding a medical compliance check due to the industry I work in … I loved all of the quality checks and scoring, since I usually do that type of QA work manually.”
Campaign engine
Clinic management software
A campaign to win 50 demo or trial requests in 7 weeks: five channels, all drafted and through Gate 2.
Campaign engine
E-commerce & Brand Strategist
“I built a visual plugin on top of your Marketing Brain, two actually: Visual Brain for the brand’s visual direction and Product Visual for product image storyboards … I put my own brand through it and it already gives me something I can work with.”
Own plugins on Marketing Brain
AI-Forward Marketing Executive
“I was in your workshop 12 hours ago. I already have a workflow live… I feel like a freaking WIZARD!”
Last cohort
Fractional CMO practice, launching
Tested a new fractional CMO offer on her own website, using public data only.
Marketing Brain
Marketing Communications Manager, TXI
“Connecting my systems made the biggest difference. Pulling deals straight from our HubSpot CRM, plus Semrush data on our content, gave the ICP and the messaging real evidence instead of guesses.”
Marketing Brain
Director of Messaging
“Being able to identify valuable work to be done, build a workflow/agent to do it effectively and consistently, and deploy it across my team will be a skill that sets my team and I apart within the org.”
Workflows
Marketing agency for health clinics
Ran the brand book through it: every claim about the buyer came back tagged hypothesis or blank.
Marketing Brain
Content Strategist
“This course got me to take writing instructions seriously and tailor them to how I work. Before, I was mostly just chatting with AI. Now I’m trying to build actual workflows.”
Workflows
Founder, Mesh Creative Studios
“Biggest learning: the review forced me to stop claiming things I hadn’t proven yet.”
Campaign engine
Here is what other marketers said
Need some more convincing?
Listen to this agency owner explain why he trains his team at CXL
The newsletter for the marketing engineer.
Free, weekly, on AI and B2B marketing.
AI changes how marketing works every week. We help you keep up and grow into the marketing engineer role: practical AI tactics, B2B playbooks, and the workflows teams are running right now.