I Tracked Every Time I Used AI During My Workday — Here's What I Learned

One morning I sat down at my desk and made myself a simple promise: every time I used AI — or deliberately decided not to — I'd write it down. Not to prove anything, to myself or anyone else. I just wanted to see what reality actually looked like. Because honestly, there's a gap between what we hear about AI and what happens during an ordinary workday that not many people talk about.

Where AI Actually Helped

First thing in the morning, I had a short report to put together. I knew what needed to be in it, but getting that first sentence down always takes longer than it should. I gave AI a few key points and asked it to draft something. The output wasn't perfect — but it gave me a foundation to work from. Editing a rough draft is so much faster than starting from a blank page. That was one of the clearest moments where AI genuinely freed up my time.

In the afternoon I had a batch of similar emails to write — the kind where the content follows the same shape but each one needs to feel personal. I gave AI a general template, added the details for each person, and it drafted every one. I just adjusted the tone. What would have taken forty minutes took ten.

I also had a meeting to prepare for and needed an agenda. I listed the topics, asked AI to give them structure. The output was usable straight away.

Where AI Didn't Help

Mid-morning I ran into a situation where I needed to address a misunderstanding with someone I work with. I asked AI to suggest how I might open that conversation. The suggestion was fine — technically. But I knew this particular person, in this particular situation, needed something different. No model knew that relationship. I ended up writing the message myself.

There was also a moment where I wanted AI to help me think through a decision — choosing between two approaches for a project. Its reasoning was logical, but it was missing the context that actually mattered: the history, the real priorities, the constraints that only I was aware of. That decision stayed with me.

The Decisions That Stayed With Me — And Why That's a Good Thing

At several points during the day I deliberately set AI aside. Not because I couldn't use it — but because I knew I shouldn't. Deciding whether to take on a new project. Assessing whether a collaboration was worth continuing. Responding to a message that carried real frustration behind it.

These are tasks that require judgment — not just information. And judgment, the kind that comes from experience and genuine understanding, is still entirely mine. I don't see this as a limitation of AI. I see it as a useful reality. It tells me exactly where I need to show up myself.

The Real Lesson: Know When to Use It and When to Step Back

When I looked back at my notes at the end of the day, one pattern stood out: AI genuinely helped with tasks that had a clear structure — drafting, summarising, formatting, repetitive responses. For tasks that needed contextual understanding, relational judgment, or personal accountability — it didn't.

Most people who get frustrated with AI aren't frustrated because AI is bad. They're frustrated because they expect the first output to be perfect, or they're using it in situations where it simply doesn't belong. AI is more like a fast-working colleague who needs clear direction — not a tool that already knows what to do.

One simple question I now ask myself before reaching for AI: Does this task have a repeatable structure? If yes, AI will probably help. If not, it'll probably just slow me down.

Using AI well in your day-to-day work comes down to this: knowing when to use it and when to step back. Not everywhere. Not nowhere.

---

If you'd like to look at what this could mean for your own work — which parts could genuinely move faster and where you need to stay in the driver's seat — I'm happy to think it through with you. A 30-minute consultation (AU$50) is enough to give you a much clearer picture.