Automate ICP research and ad reporting with n8n and Claude Code

The two most useful AI workflows in paid marketing aren’t about writing better ad copy. They’re about building your ICP from real customer data instead of a guess, and automating campaign reporting so insights reach you the same morning instead of the following Monday.

We hosted the second live workshop in our AI Native Marketer program with Nick Christensen, head of growth at AppSumo, on exactly this: building agents to improve ad performance and monitor campaigns.

We opened by asking the room one question: which part of paid do you most wish just ran itself? Experimentation, budget optimization, campaign setup, creative, and targeting all came up. But the biggest bucket by far was reporting and insights: pulling the data, aggregating it, and turning it into something you can act on.

Nick shared two workflows he runs every day that address that directly. This post walks through both, with the exact inputs, tools, and steps.

TL;DR

• Two AI workflows can remove significant manual work from paid marketing: building an ideal customer profile (ICP) from real customer data and automating daily campaign reporting.
• Use Claude Code to turn customer exports, reviews, sales transcripts, and support tickets into an ICP dossier, then generate Google Ads and Meta campaign assets grounded in buyer language.
• Use n8n to pull Google Ads and Meta performance data, merge the results, evaluate campaigns against predefined targets, and send a daily digest to Slack or email.
• The value isn’t just faster reporting. It’s getting actionable campaign insights without waiting for the next manual report, while leaving budget and optimization decisions with the marketer.

Workflow 1: Build your ICP from real customer data

Most campaigns start with a channel and then go looking for customers to fill it. Nick argues for the opposite order: find your best customers first, and let the channels fall out of who they are.

The mechanism behind this is the Pareto principle. In many businesses, a small group of buyers accounts for a disproportionate share of revenue. The exact concentration is something you need to establish from your own customer data. It’s the pattern your highest-value customers share, found in your actual data.

The solution is a Claude Code prompt that reads a folder of real customer data and outputs an “ICP dossier”: a one-page profile of your highest-value customer segment, written in their own words, plus campaign assets ready to paste into ad platforms. That includes Google Ads headlines and descriptions, image specs, and Meta audience suggestions and Google Ads Performance Max search themes.

What data should you feed Claude Code to build an ICP dossier?

Start with purchase history to identify your highest-value customers, then add reviews, sales conversations, and support tickets to understand what they need and how they describe it. Claude Code can synthesize that evidence into an ICP dossier and use it to generate campaign assets grounded in actual customer behavior.

The workflow is only as good as what goes into the folder:

• A customer export (from Stripe or your payment processor) with spend, order counts, and any available profile data
• Customer reviews
• Sales or customer call transcripts (Nick uses Granola, which connects to Claude Code via MCP)
• Support tickets

Start with 10 to 20 customers to spot possible patterns, then validate those findings against the wider customer base. If you don’t have this data yet, the fix is old-fashioned: talk to a few of your best customers, record the calls, and drop the notes in the folder.

Why does grounding AI in customer data produce better ad copy?

Grounding gives Claude Code actual evidence of who your best customers are, what they’ve purchased, and how they describe their problems. Without that context, AI can generate plausible ad copy that has little connection to your buyers. The customer data is what makes the output specific rather than generic.

This workflow captures the finding that stood out most in our AI maturity benchmark: across every category we measured, how often marketers use AI far outpaced how well they used it. Everyone is prompting. Few are grounding.

That’s the difference here. Ask a chatbot for ad copy with no inputs and you get the same generic persona and headlines as everyone else, because the model is inventing a customer rather than describing a real one. Generic-sounding AI output is a well-documented failure mode in B2B content specifically.

This system synthesizes evidence instead of guessing: real spend data, real reviews, real call transcripts, so the campaign comes out in the words your buyers actually use. The skill isn’t the prompt. It’s choosing what to feed the system, and judging what comes out of it.

Titles are becoming obsolete. Shipping is what matters.

This exact output used to require a copywriter’s brief, a researcher’s survey project, and weeks of back-and-forth between teams. Now one marketer ships it in minutes, with the AI handling synthesis and the marketer handling judgment.

Workflow 2: Automate reporting so insights come to you 

A workshop poll matched what we’d have guessed: most attendees spend 1 to 4 hours per week assembling ad reports across GA4, ad platforms, and BI tools. Nick used to spend every Monday morning doing the same thing.

The replacement is an n8n workflow that runs automatically every day at 9am:

1. Pulls campaign data from Google Ads (via API) and Meta (via Meta’s MCP)
2. Merges the sources into one table
3. Analyzes the merged data against your goals: which campaigns meet target, which need attention, each with a grade
4. Posts the digest to Slack, email, or both

To reproduce the daily 9 a.m. run, configure n8n’s Schedule Trigger with the correct workflow timezone and publish the workflow. Before grading campaign performance, make sure the reporting periods, conversion definitions, and attribution settings are comparable across sources. Otherwise, the automation can flag differences caused by measurement rather than campaign performance.

The analyze step is where your goals live. The automation only knows what “good” looks like if you’ve defined it.

If you’re new to connecting these systems, our n8n Week breakdown covers the workflow architecture and the points where human oversight still matters.

How does automated reporting improve paid campaign decisions?

Automated reporting puts performance changes and campaign alerts in front of marketers every day instead of making them wait for a weekly report. The benefit isn’t just reducing the hours spent collecting data. It’s identifying opportunities and problems early enough to act on them.

The pressure to build these systems is showing up in hiring data. In our 2026 analysis of US marketing job descriptions, 35% of listings mentioned automation, up from 18% in our Q1 sample of 653 jobs. Employers are increasingly looking for marketers who can connect tools and build repeatable workflows, not just use AI to generate content.

That pressure is exactly why becoming AI-native matters here. And the win isn’t only the hours saved assembling the report. It’s speed to action: when the insight lands in Slack at 9am already graded, you scale or pause a campaign the same morning, not the following Monday.

	Manual reporting	Automated daily digest
Data pull	Manual export from each platform	API pull, scheduled daily
Merge	Manual, in a spreadsheet	Automatic, into one table
Judgment	Applied once a week, after the pull	Applied daily, against a pre-set goal
Time to action	Days (often the following Monday)	Same morning

How should paid marketers start using these AI workflows?

Start with one repeatable task that costs time or slows down decisions: researching your best customers or assembling campaign reports. Use AI to handle the data collection and synthesis, then keep targeting, testing, and budget decisions with the marketer. The goal is a workflow that runs consistently, not another one-off AI output.

The pattern across both workflows is the same: AI does the synthesis and the assembly, you make the decisions. Scaling, pausing, and testing are still judgment, taste, and experience. They just no longer come bundled with hours of busywork.

A few tips if you want to move in this direction:

• Expect the friction in the wiring, not the concepts. API credentials, webhooks, and app passwords are where people get stuck. When you hit an error, paste it into Claude Code and let it troubleshoot.
• Start with your customer data, not your channels. Export your best customers from Stripe or your payment processor, add reviews and call notes, and let Claude Code or Codex synthesize the ICP. Even if you think you already know your ICP, run it against real data.
• Define success before you automate reporting. Pick one blended metric and set a target per channel. The automation is only as useful as the goal it grades against.
• Merge your data sources before judging a campaign. In-platform numbers alone will mislead you; combine them with GA4 or your own database.

Frequently asked questions

What is an ICP dossier in paid marketing?

An ICP dossier is a structured profile of your highest-value customer segment, built from real customer information rather than assumptions. It brings together purchasing behavior, common problems, objections, customer language, and other characteristics relevant to targeting. For paid marketing, that information can inform audience selection, ad messaging, and campaign assets. An ICP defines the customers or accounts you want to reach; buyer personas describe individual decision-makers within those accounts.

How much customer data do you need to build an ICP using AI?

You can start with 10 to 20 customers to identify potential patterns, particularly if you’re using detailed interviews, reviews, and purchase histories. But that isn’t a validated minimum sample size or a guarantee of representative findings. Compare the initial patterns with your broader customer data, especially profitability, retention, purchase frequency, and customer value. Claude Code can accelerate the analysis, but it can’t make a small or biased sample representative.

How do you connect Google Ads and Meta Ads to an n8n reporting workflow?

Connect both advertising platforms through authorized reporting integrations, such as their APIs or an appropriate MCP connection. Google Ads API access is managed through a Google Cloud project, OAuth 2.0 credentials, and the required production access level. For Meta, use a reporting connection with the necessary account permissions. Once the data is accessible, n8n can retrieve campaign metrics on a schedule, combine the results, analyze performance against your targets, and distribute a digest. Google for Developers+2

Why do Google Ads, Meta Ads, and GA4 sometimes show different conversion numbers?

The platforms can use different attribution models, conversion windows, counting methods, and reporting dates. Google Ads and GA4 may even attribute a conversion to different dates, depending on whether they’re reporting the ad interaction or the conversion itself. Combining the data doesn’t automatically resolve these differences. Before automating campaign grades, decide which source of truth and conversion definition will drive each decision, and make sure the comparison is consistent.

Should an AI reporting agent automatically pause or scale campaigns?

Not by default. An AI reporting agent can identify campaigns that fall outside predefined performance targets and recommend actions such as pausing, scaling, or testing. That doesn’t mean it should execute those changes without oversight. Start with reporting and recommendations, validate the accuracy of the underlying data, and keep budget decisions under human control. If you later automate campaign changes, use explicit spending limits, approval rules, and safeguards against incorrect data.

What to do next

If you want to build these workflows yourself, CXL’s B2B ad campaigns with Claude Code and n8n course covers both sides: turning ICP research into scored ad variants and setting up automated Google Ads and Meta reporting through n8n. It follows the same practical approach Nick demonstrated, with more detail on the build and implementation.

Join CXL’s AI Native Marketing program

CXL’s AI Native Marketing program takes this further with hands-on workshops where marketers build working AI systems using their own tools and data. The current program includes building a marketing context system, a campaign engine, and a reporting agent that brings performance insights into your day-to-day workflow.