A guide on reporting and campaign creation with Active Intelligence
This guide is based on the Insider Tips That Work — Live Building Autonomous Marketing webinar, published on the AI Lab by ActiveCampaign.
Get the quick-start guide
What will I accomplish with this guide? By the end of this guide, you’ll know how to ask Active Intelligence plain-language questions about your campaign performance, get back charts and recommendations in 30–40 seconds, and turn those insights into a fresh campaign without rebuilding it from scratch. It mirrors the workflow Luis Fernando Castillon walked through with Mamie Hatter, the same one that took Parrish Law’s reporting from 10+ hours per month to a few minutes.
Before you start, you’ll need:
Guide on Active Intelligence
- An ActiveCampaign account with Active Intelligence enabled
- A few months of campaign send history so Active Intelligence has data to analyze
- A short list of recurring questions your leadership asks about marketing performance (this is your starting prompt library)
- At least one segmentation scheme already in place if you want segment-level comparisons
Quick reference
- Total time: 30–45 minutes for your first session
- Tools needed: ActiveCampaign Active Intelligence, your campaign data
- Key output: A reusable prompt library plus one new campaign built from your top performers
Watch this section
For full context on the following topics, watch these sections of the webinar:
- Asking Active Intelligence for a general performance overview: [11:37]–[13:23]
- Trend lines and segment comparisons in plain language: [13:53]–[16:48]
- Generating a campaign from your top performers: [16:48]–[19:16]
- Targeted prompts for subject-line and date-range questions: [19:51]–[24:17]
The workflow
Phase 1: Set up a prompt library before you start asking
After this phase, you’ll have: a saved list of 5–10 questions you can paste into Active Intelligence anytime.
- List the questions leadership asks most often: how did campaigns perform last month, which subject lines work, what should we test next. Aim for the questions you currently dread answering.
- Save them in a doc your team can reuse: Notion page, Google Doc, or a pinned Slack message. The point is fast retrieval, not pretty formatting.
- Tag each question by purpose: overview, trend, segment compare, subject lines, campaign generation. Tags help you grab the right prompt mid-meeting.
Phase 2: Run a general performance overview
After this phase, you’ll have: a top-line summary of total sends, open rate, click rate, and what’s working.
- Open Active Intelligence: the entry point lives in your ActiveCampaign sidebar.
- Ask for an overview in one prompt: paste a question like the one below.
The prompt you’ll use
How have my campaigns been performing? Give me a general overview of campaign performance with key metrics like total emails sent, open rate, and click rate, plus key insights on what’s working and what to improve.
- Wait 10–30 seconds for the response: Active Intelligence pulls the data and writes the summary itself. Luis described it as long enough to grab a glass of water, not long enough to get back to spreadsheet work.
- Read the “key insights” section: it surfaces what’s working and what’s underperforming so you don’t have to skim every campaign yourself.
- Export the summary for leadership: the response includes an export option you can drop straight into a meeting deck.
Phase 3: Visualize trends and compare segments
After this phase, you’ll have: a chart of open-rate trend over time and a side-by-side comparison of two segments.
- Ask for the trend line: prompt Active Intelligence with the question below.
The prompt you’ll use:
What’s the open rate trend in the last few months? Visualize it over time.
- Look for inflection points: the chart shows when results changed. Parrish saw the lift right when they switched on segmentation.
- Run a segment comparison: prompt for a side-by-side, e.g., the one below.
The prompt you’ll use:
Compare performance and engagement of our [SEGMENT_A] segment and [SEGMENT_B] segment. Show how each campaign is performing in each segment, and recommend optimizations.
- Read the recommendations: Active Intelligence returns specific actions per segment, not just numbers. That’s the move from reporting to decision.
I went from 10+ hours monthly in spreadsheets to asking questions and getting instant answers. So it changed how Jim and I make decisions now.
Phase 4: Generate a new campaign from your top performers
After this phase, you’ll have: a draft campaign that pulls from the elements your past sends already proved.
- Ask for your top performers and why they worked: prompt with the question below.
The prompt you’ll use:
What were some of my top performing campaigns in the past few months, and why did they perform so well? Include open and click rates.
- Ask Active Intelligence to draft a new campaign from those insights: layer the next prompt on top.
The prompt you’ll use:
Create a campaign to warm up leads that uses these insights in its content.
- Review the draft: Active Intelligence builds a full campaign (subject line, preheader, body copy, layout) using elements from the winners. Edit subject line, preheader, and copy in place.
- Swap images or tweak copy as needed: the structure is locked in; you’re customizing details, not rebuilding from scratch.
- Send when the draft passes review: click send when you’re happy. The whole loop took Luis a few minutes per campaign.
Phase 5: Refine with targeted prompts
After this phase, you’ll have: a small bank of specific, repeatable prompts that go beyond overview-level questions.
- Ask date-bounded questions when you need a clean baseline: scope to a single month or campaign batch.
The prompt you’ll use:
How did my manual campaigns from [MONTH YEAR] perform? Share insights for [NEXT MONTH] campaign improvement.
- Ask for subject-line guidance scoped to a segment: the more specific the prompt, the more useful the output.
The prompt you’ll use:
What are the best subject lines for the kind of content we share, focused on [SEGMENT] potential clients?
- Save every prompt that returned something useful: add it to your prompt library. Tomorrow’s report is half-written already.
The more specific that you can be in those prompts, the better data you’re going to get back, the better insights you’ll get back, and then you know the next action to take.
Related
More data from the AI Lab.