I'm mind-blown every day by how powerful it is to have your own agent!
For example, today.
I'm working on the Lisbon Connection Week — a 10-day urban festival where we will organize electronic music events across the city.
Our goal: we have to get it into the press.
What I got: my colleague Virginia found 6 magazines/websites that could help out.
I just dropped my agent a voice note about what he should do — contact the magazines, get the best emails, draft personalized messages to inquire about media partnerships.
Then he gets back with a bunch of questions. So I just answer him back. By talking to him.
I then drop the list of the six magazines, the link to our Notion page where we have a lot of comments on the different events we want to host (even though it's quite messy). It gets it all.
Then it starts the research.
It is also smart enough to update files, so it automatically saves all of the information I just gave him in our core Notion. And remembers it.
"My agent knows not to bother me between 13h and 14h30. Got to respect food!"
I go for my lunch break — and come back to the following message: 6 fully drafted, personalized outreach emails ready to send.
And that was it. I just had to copy-paste the emails. They were all individualized. And I just sent them out.
Now, I could also give him access to my email and have HIM send out the emails.
But there's a reason why I didn't do this yet. Setting up your own super-powerful agent comes with several risks.
Just yesterday, I had a call with my friend Gustavo (name changed) who was inspired by our last webinar.
After watching me in action, he decided to create his own agent and had the genius idea to give him access to his business WhatsApp.
Two days later, we're on the phone.
"The agent is doing really well. He is answering all of my customers — non stop!"
I was thinking that this was not how you should start using your agents, but didn't say anything — seems like it was working out for him.
But here's what he says next:
"The only issue is that he does it without my approval. And when I ask him why he writes the messages, he says it's not him!"
We had to figure out why the agent was sending messages without approval, why he was denying it, and reboot him from scratch.
This wouldn't have happened if Gustavo had treated the agent like a new employee.
You do not give access to all your passwords and platforms to your agent on day one. You slowly onboard an employee into your processes. You give them their own email, their own drive, and they slowly gain your trust the better they perform.
There are certain principles in setting up your agents and making them work!
This is why I launched the Timeback Bootcamp, where I walk you through how to set up your team of agents in a secure way that creates reliable results — and automates the workflows that currently control your time.