Steps to build an AI-partnered email campaign

By the end of this checklist, you’ll have planned, written, and analyzed a full email campaign using AI as your thought partner. It’s based on the workflow that Pam Covarrubias used to run an 18-email Black Friday / Cyber Monday campaign as a solo business owner. Expect about one to two weeks from planning to post-campaign analysis.

Before you start

Set up your AI tool: create an account on Claude (Covarrubias’ preferred tool) or another LLM
Have your email platform ready: you’ll need an active ActiveCampaign account (or similar) with an existing contact list
Gather your brand materials: writing samples, audience profiles, past emails that performed well, product/service descriptions, and any frameworks you use in your work

The workflow

Phase 1: Build your super prompt

After this phase, you’ll have: a comprehensive brand document that gives AI the context it needs to write in your voice and understand your business.

From Pam: “The biggest thing the doc does is it gives me guardrails to my wild mind so I don’t make things up as I go.”

Create a Google Doc (or similar) as your living brand document: this is your “super prompt,” a single reference that AI can draw on for every campaign
Add your methodologies and frameworks: describe the systems you use in your work so that AI understands your approach
Define your ideal customer in detail: include demographics, psychographics, pain points, and the stage of their journey. Covarrubias refers to hers by a persona name (“Marisol”) with specific stages
Document your constraints: marketing tactics you’d never use, topics that are off-limits, brand boundaries
Include your business history and backstory: background on your business, podcast transcripts, case studies, or other context that shapes your brand
Add writing samples: paste in emails, social posts, or blog content that resonated with your audience so the AI can learn your voice
List your tech stack: name every tool you use (ActiveCampaign, ThriveCart, MemberVault, etc.) so the AI can reference integrations without you having to explain them each time

Phase 2: Plan your campaign with AI

After this phase, you’ll have: a campaign brief with timeline, pricing, upsells/downsells, and a technical setup plan, all co-created with AI.

Share your super prompt with AI: paste or upload the full brand document into a new conversation so the AI has full context
Brief the AI on your campaign concept: describe the offer, timeline, pricing tiers, upsells, downsells, and revenue goal
Review the AI’s campaign timeline and technical recommendations: check the suggested tags, automation sequences, and milestones. Adjust what doesn’t fit your actual setup
Set up your ActiveCampaign tags: use the AI-recommended tag structure as a starting point, then refine based on your naming conventions
Set up your ActiveCampaign tags: use the AI-recommended tag structure as a starting point, then refine based on your naming conventions

Starter prompt

You are my seasoned and elite business coach. We are working on a [CAMPAIGN_TYPE] sale. I need to build the brief and instructions. But first, let’s figure out priorities. Ask me questions to clarify and develop the best possible plan.

I want to offer [PRODUCT_DESCRIPTION] at [PRICING_TIERS]. The upsell could be [UPSELL_IDEA]. The downsell could be [DOWNSELL_IDEA]. I want to make [REVENUE_GOAL].

What must be done is [TECHNICAL_SETUP_NEEDED] and connect to ActiveCampaign with tags. I don’t have a plan for ads, just organic on socials.

Phase 3: Write your campaign emails

After this phase, you’ll have: a complete set of campaign emails loaded into ActiveCampaign and scheduled to send.

“I go to my cowriting tool GPT and I say, ‘OK, I have this campaign I’m working on. Let’s write the first email together.’ And it gives me a starting point and I change things, remove words, add storytelling — I get it to 100%.”

Co-write each email with AI: work through emails one at a time. Let AI draft, then add your storytelling, adjust the voice, and edit until it sounds like you
Build your email sequence in ActiveCampaign: load each email into your automation with the appropriate triggers and tags
Enable Predictive Send: turn on Predictive Sending so that each recipient gets emails at their optimal time

Phase 4: Analyze results with AI

After this phase, you’ll have: a data-driven post-mortem with specific insights on what worked and what to change for your next campaign.

After this phase, you’ll have: a data-driven post-mortem with specific insights on what worked and what to change for your next campaign.
Feed the data to AI for analysis using the Starter prompt below
Look for insights you missed: Covarrubias found that AI helped her see past the loss (unsubscribes) and focus on the gain (engaged subscribers who stayed)
Document lessons for next time: note audience targeting changes, email sequence adjustments, and any timing improvements that the analysis surfaced

Data analysis starter prompt

Analyze which specific campaigns had higher click rates to replicate successful elements. Here is the data from my campaign: [PASTE_STATS]

Quick reference

Total time: 1–2 weeks (a few hours for the super prompt if you’re building from scratch, then roughly a week for planning and writing)

Tools needed: Claude (or another LLM), ActiveCampaign, your checkout tool (e.g., ThriveCart)

Key output: a fully planned, written, and analyzed email campaign, plus a reusable super prompt for every future campaign

Ready for the full story?

Read A 142-page prompt, 18 emails, and one solopreneur’s first Black Friday campaign, featuring Pam Covarrubias of Spread Ideas Move People, published on the AI Lab by ActiveCampaign.

Get the checklist