Meet the Growth Agents is a video series that spotlights each of the AI agents that power the GrowthLoop Compound Marketing Engine.

The series features Anthony Rotio, GrowthLoop Chief Data Strategy Officer, who walks through each agent, how they work, and how they help teams accelerate the marketing cycle and launch more personalized campaigns faster.

In episode one, Anthony showcases the Research Agent (formerly called the Search Agent). The Research Agent acts as your personal brainstorming partner who searches the internet to retrieve contextual answers about campaign decisions.

Video transcript:

Hi everyone. I'm Anthony Rotio with GrowthLoop and welcome to Meet the Growth Agents, where I'll take you through in each episode, one by one, our AI agents that power the Compound Marketing Engine. So let's jump right in. Today we'll be talking about the Research Agent. Now each of our AI agents have access to the customer data that you've connected in GrowthLoop and also the world data that the foundation models have been trained on.

Now what the Research Agent does is it extends that data that your agents have access to, to allow for public internet data to be accessible. So you can search for data that may be outside the context of the training data of these foundation models.

So let's walk through an example. I'm going to pretend to be a retailer here. I'm connected in an environment that has retail data connected. So let's go ahead and see what the Research Agent can help us out with.

Let's start with a case where we're researching events in Q4 2025 like retailer marketing campaigns. Now this is something where maybe I'm in the ideation phase as a marketer I'm deciding what actually I want to target in Q4. So maybe I have an insight, maybe I've written a brief, but I don't know how I'm going to activate that yet in Q4. So now what happens is the Supervisor Agent, this is the kind of the air traffic controller for all the agents in GrowthLoop.

And what the Supervisor Agent did was reach out to the Research Agentand say, hey, we need some information here. And very quickly in real time, the Research Agent came back with a report about some events in October 2025, November, and December on some of the events that we might want to attach to for this campaign or activate in Q4. That's interesting, right? And this is the agent kind of acting on its own to do these things.

Now when it really starts to get interesting is when you start to see the Research Agent work with other agents to actually produce assets that you can activate. So let's go ahead and look at another example. Here we'll say I'm a marketer at Target and we want to research recent earnings calls from my top competitors and identify audiences that might be at risk for us for losing share of wallet. Right. So this is a typical case that you might have when you're competing right in an environment with other retailers and you know, you want to make sure that your customers, especially your top customers, aren't going to be lost to competitors for some of those key buying opportunities.

So here we have the supervisor saying, okay, this is kind of a more complex case. We want to actually do something here where the Research Agent is going to look at recent earnings calls and produce a report. Now we'll get a report back similarly to in the first example. But what I want to show after we get the report back is how we actually use this to then go build an asset that we can activate against in GrowthLoop.

So here we're getting some very specific information about three of our top competitors here with a conclusion and some recommendations. So now that we have that, the supervisor says, okay, we actually could do something useful with this information. So it automatically decided to reach out to the Audience Agent, which is another agent we'll cover in another episode.

But here the Audience Agent is looking at the connected data in my data warehouse, right? All the customer data that I have. And it's deciding which audiences might be at risk for losing share of wallet to these other competitors. And then it builds these audiences for me. So I can actually target these audiences and activate them across the different sales and marketing channels where my teams are already working and very quickly get them into the hands of the marketers that need to activate this, whether they're me, myself, or other folks on the team. So this is an example of how the Research Agent can be a really powerful tool to connect kind of the world model data, the foundation models, my customer data in the data warehouse, and also public recent information on the internet to give us really tight targeting for our next campaign.

So drop a comment if you have questions about a particular use case, or if you want to see something in a future episode.

If you want to see something specific for your industry outside of retail, that's great too. Let's work together and apply AI to grow your business. Thanks for stopping by today.