How to watch external data to subscribers and trigger the right email automatically

What will I accomplish with this guide? By the end of this guide, you’ll have an AI-driven segmentation flow that watches an external data source, scores your subscribers against it, and triggers an ActiveCampaign email to the right people at the right time — no human drafting, scheduling, or list-building. It’s based on the workflow Adam from electricityrates.com walked through in this webinar, where the same approach doubled monthly conversions and would have taken six people to run by hand.

Before you start, you’ll need:

ActiveCampaign account with Automations and a populated subscriber base
Access to the external data source you want to react to (rate changes, inventory, weather, pricing — whatever drives your customers’ decisions)
An LLM with API access (Adam used Gemini, but Claude, ChatGPT, or others work)
A connector between your LLM, your external data, and ActiveCampaign — Make, Zapier, n8n, or a custom script
Subscriber attributes that the AI can match against the external data (geography, plan type, contract end date, etc.)

Quick reference

Total time: Three to six weeks for the first version, depending on how clean your subscriber data is
Tools needed: ActiveCampaign, an LLM API, your external data source, an automation connector
Key output: Email campaigns where AI selects the audience and timing based on a real-world signal, with no human triage step

Watch this section

For full context on the following topics, watch these sections of the webinar:

The starting point with Gemini and the question Adam asked it — [19:36]–[20:10]
How the AI matches external rate changes to specific subscribers — [20:43]–[21:17]
The results and the team-of-six counterfactual — [21:50]–[22:23]

The workflow

Phase 1: Get your subscriber data and external data into a place AI can read

After this phase, you’ll have: a clean export of your ActiveCampaign subscribers plus a feed of the external data your AI will react to.

Export your subscriber base from ActiveCampaign: include the custom fields the AI needs to match on — geography, plan type, contract end date, account age.
Identify the external data feed that drives your customer decisions: for Adam it was utility rate changes by state; for you it might be inventory levels, competitor pricing, weather, sports schedules, or anything else.
Set up a connector that pulls the external data on a schedule: Make, Zapier, or a Python script the AI can call.
Confirm both data sources can land in the same place: a shared Google Sheet, a database, or directly in your AI tool’s context.

Phase 2: Use AI to identify which subscribers match the moment

After this phase, you’ll have: a daily or weekly AI run that returns a list of subscriber IDs who are best-matched to act on the latest external signal.

Start simple — ask the AI to look at both data sets together: Adam’s first question to Gemini was the unstructured version of “look at this data and tell me which customers have the highest likelihood to switch.”
Refine the prompt to ask for reasoning: request the matching subscriber IDs and a one-line reason per match. The reasons become useful for QA and for the email body itself.
Run the AI on a recurring schedule: daily for fast-moving data (rates, inventory), weekly for slower-moving data (contract end dates, lifecycle).
Pipe the AI’s output back to a place ActiveCampaign can read: a tag, a custom field update, or a list import.

We started by telling Gemini, just have a look at this and tell me what you think. Look at all of this data that I’m gonna feed into you and tell me which customers have the highest likelihood to switch to a new energy provider, and why.
Adam Cain, ElectricityRates.com

Phase 3: Connect the AI’s output to ActiveCampaign segmentation

After this phase, you’ll have: an ActiveCampaign segment that updates automatically based on the AI’s matching, ready to fire an automation.

Create a custom field for the AI’s match reason: so the email content can reference what the AI saw (the rate change, the trigger event).
Create a tag for “AI-matched this week”: apply it to the subscribers the AI identified, remove it from prior matches before applying.
Build a segment that selects “tagged AI-matched”: this is what your automation will target.

Phase 4: Trigger the right email automatically

After this phase, you’ll have: a fully automatic flow where the AI’s matching kicks off an ActiveCampaign email to the matched audience.

Build an automation that triggers when the tag is applied: start node = tag added.
Use conditional content to insert the AI’s reason field into the email body: so the message references the specific event for that subscriber.
Add a wait step or send-time logic: so subscribers who match on the same day get sent during their best engagement window.
Set the goal: the action you want the subscriber to take (click, reply, switch plans).
Test end-to-end with a small list before scaling: verify the AI matches are sane, the email content is on-brand, and the conversion path works.

Phase 5: Measure, refine, repeat

After this phase, you’ll have: a feedback loop where the AI’s matching decisions get scored against actual outcomes and the prompt improves over time.

Track conversions per AI-matched campaign: Adam saw double the monthly conversions versus his prior approach.
Track unsubscribe rate: if it stays low, the AI’s matching is on target. If it spikes, the prompt is over-matching and needs tightening.
Feed the conversion outcomes back into the prompt: add a “subscribers who looked like X converted at Y rate” note so future runs weight that pattern higher.
Expand to additional triggers: once one external-data trigger works, add another (a second product line, a second region, a second event type).

Watch the webinar

Watch now

Get the quick-start guide