How I Use AI to Get More Done in Less Time

Okay so I want to be upfront about something: I was deeply skeptical of all the 'AI makes you 10x productive' content online. It felt like the same energy as those '5am routine' videos where someone's alarm goes off and they immediately start doing pull-ups. Like, cool for you, but that's not my life.

But then I actually started paying attention to where my time was going. And I was horrified. I was spending probably two hours a day on things that didn't require my brain at all, writing boilerplate emails, reformatting notes, trying to remember what I was supposed to do next, rewriting my own drafts because the first version was garbage. Stuff that felt like work but wasn't really doing anything.

That's when I actually gave AI tools a real shot. Not as a magic fix, but as a way to get the boring stuff off my plate so I could focus on the things that actually require me.

The email problem was the first thing I tackled

I'm not a bad writer. I actually like writing. But professional emails have this weird middle zone where you have to be warm but not too casual, clear but not blunt, and somehow convey exactly what you need without sounding demanding. I was spending 20 minutes on emails that should've taken 5.

Now I draft a rough version, like, genuinely rough, the kind of thing I'd never send, and then I paste it in and say 'make this sound like a normal professional email, keep it short.' Takes 30 seconds. And honestly the output is usually better than what I'd write if I spent 20 minutes on it, because I'm not overthinking it.

The key thing I learned is that you still have to read it and tweak it. AI will sometimes make things more formal than I want, or it'll miss a specific nuance I was trying to include. But it gives me something to react to instead of starting from a blank page. That alone is huge.

I use it as a second brain for notes

My notes are a disaster. Were a disaster. I have notes in three different apps, voice memos I never transcribed, screenshots of things I meant to come back to. It was chaos.

What I do now is honestly I dump everything into a single prompt, all the random stuff from my head, the voice memo transcript, the bullet points from a meeting, and I ask it to organize this into action items and reference material. Five minutes later I have something I can actually use.

I'm not going to lie, the first few times I did this I felt like I was cheating somehow. Like I should be organizing my own thoughts. But here's what I realized: the thinking happened when I was in the meeting, or when I was walking and talking into my phone. The organizing is just admin work. I don't need to do admin work.

First drafts on everything

This is the big one. The thing I use AI for most is getting a first draft of anything down so I'm not starting from nothing.

Doesn't matter what it is. A proposal, a recap email, a social media post, notes for a phone call, a rough outline for a project. I give it context and ask for a first pass. Then I rewrite it. Then it sounds like me and actually has the right content in it.

The blank page problem is real and it's the biggest time suck I had. I'd sit down to write something and spend 15 minutes just staring at the cursor. Now I spend 30 seconds on a prompt and then I have something to react to. My brain works way better in reaction mode than creation-from-nothing mode, and I think that's probably true for most people.

Research that would've taken an hour

I do a lot of research for work, background on companies, summaries of topics I don't know well, pulling together information from a bunch of different sources. This used to eat my morning.

Now I'll ask for a summary of something, or ask it to explain a concept to me like I'm coming in cold, or ask it to give me the main things I need to know before a meeting. It's not perfect. I always double-check anything that matters. But getting the lay of the land in five minutes instead of an hour? That change alone has been significant.

The thing I had to get over is the feeling that I should be doing this research myself. But doing research isn't the skill I'm paid for. Understanding what I find and making good decisions with it is. Those are different things.

What I don't use it for

I want to be honest here because I think a lot of AI productivity content skips this part.

I don't use AI for anything where the relationship is the point. If I'm writing something personal to someone I care about, I'm writing it. Full stop. AI-generated messages to real people feel hollow to me and I think people can sometimes tell.

I also don't use it to replace actual thinking on things that matter. Strategy stuff, decisions, anything where I need to actually grapple with a problem. I do that myself. AI is good at producing output, but it's not great at sitting with ambiguity or making judgment calls that require real context about who I am and what I care about.

And sometimes it just gets things wrong. Confidently wrong. Which is its own kind of annoying. So anything high-stakes gets verified.

The actual time math

I've started being more deliberate about tracking this. My honest estimate is I'm saving somewhere between 90 minutes and two hours a day. Which sounds like a lot. But when I look at the list, emails, note organization, first drafts, research, it adds up fast.

What I'm doing with that time is not, like, meditating or going for more walks (though maybe I should be). I'm using it to do more of the actual work. The stuff that matters. The stuff that requires me specifically and not just any capable person with a computer.

And that's the reframe that made all of this click for me. It's not about being lazy or cutting corners. It's about figuring out where your actual value is and protecting that time. For me, the thinking and the judgment and the relationships are where I add value. The formatting and the first drafts and the summarizing? That's not where I'm irreplaceable.

Once I got honest about that, using AI started feeling less like cheating and more like just... being smart about my time. Which is what I was trying to do all along.