JournAIing

My reflective practice is ever evolving and this is another post on that evolution.

I’ve written elsewhere about the mechanics of journaling ( see Journaling, Upgraded), you know things like sitting at the same time each day in a quiet space and being focused. Now that still matters, because none of what follows works if you’re inconsistent.

This isn’t about inspiration. It’s about rhythm.

Treat journaling as a system, not a habit

The shift for me was going from, “What should I write today?” to, “What am I trying to notice over time?”

That led me to build a simple structure:

• A dedicated project (in my case, a GPT project folder) for me to store my reflections

  • Labeled by date and month

• A standing prompt that shapes how I reflect

• A consistent set of questions that I return to

And I have found that this doesn't constrain my writing, happily it is training my attention.

Build a meta prompt that does some of the thinking for you

Instead of starting from scratch each day, I use a prompt that guides the reflection.

It does a few things:

• Tracks recurring themes, tensions, and patterns

• Notices contradictions or dropped ideas

• Challenges easy or sentimental conclusions

• Helps mature ideas rather than just generate them

In other words, it acts a bit like a skeptical, slightly annoying thinking partner.

You don’t need GPT for this. But it helps. Because it remembers what you forget.

* (available for download here) not that this is a starting point, there are areas in the prompt for you to add your own customization

Capture first, shape later

Before any structure or questions come into play, I start in a very simple way. I’ll open with the phrase, “This is a journal entry for the date…” and then I just talk.

At this point I am not trying to organize my thoughts or make them sound coherent. I’m just getting the day out of my head and onto the page. That usually includes a mix of things. What went well, what didn’t, conversations I had, ideas that are starting to form, frustrations that are still sitting with me, and the odd thought that doesn’t fully make sense yet but feels like it might lead somewhere. It’s not tidy, and it’s not meant to be.

Once I’ve done that, I ask for it to be transcribed verbatim. Not summarized or improved, just cleaned up enough so that I can read it back clearly. That distinction matters more than it looks, because the moment you start editing or interpreting too early, you lose something of what was actually there.

What I’m doing here, whether I realized it at the start or not, is separating two different modes.

First, there’s capture, which is about recording what actually happened and how it felt. Then there’s reflection, which comes later and is about making sense of it, spotting patterns, and asking better questions. Most of us try to do both at the same time, and that tends to blur things. We tidy up the story before we’ve really seen it.

So I speak the day then read the transcription, and when I do that, something interesting happens. I’m effectively creating a small distance between the experience and my interpretation of it. From a cognitive perspective, that shift moves me from being inside the experience to observing it, which engages more reflective, pattern-recognition processes rather than reactive ones. It’s a simple way of turning memory into something you can actually work with. In that reread I often notice things I didn’t realize I was thinking or feeling at the time. And that’s usually where the more interesting insights begin to show up.

Ask better questions**

Most journaling fails because the questions are too vague. “What did I do today?” will get you a list. Better questions get you insight. So I sat down an wrote a list of questions that sits inside the project folder (and that the prompt explicitly asks the GPT to use)**

Here are a few I’ve been using:

• What did I notice today?

• Who did I hold space for today?

• What felt easy that used to feel hard?

• Where did I feel tension or resistance?

• What idea is trying to form but isn’t fully there yet?

None of those were the questions I started with in July 2025 when I started this experiment, they've evolved out out an almost daily practice

 **(available for download here) again, these are a starting point, get creative and build your own list - the GPT can pick at random to keep it fresh for you

Accept that not everything becomes output

Accept that not all ideas arrive fully formed. Some need to be revisited, turned over, composted. Some days you’ll write something and notice something about it immediately.

Some days you’ll write something, leave it for a few days, and come back to it later

Somedays you'll have an idea for a course, a blogpost etc. For those I put

them in another folder called the “idea bank” and return to it regularly.

Not to force output. To resurface what still has energy.

Build in periodic reviews

Daily journaling is useful AND another layer of value to me is when I zoom out.

Every few weeks, I scan back through what I’ve written and look for:

• Repeated themes

• Persistent tensions

• Ideas that keep resurfacing

• Patterns in how I’m thinking or reacting

And then, on a quarterly basis, I do a deeper review. Not just “what happened this quarter,” but:

• How has my thinking evolved?

• What have I been avoiding?

• What am I starting to see more clearly?

This is where journaling stops being reflection and starts becoming direction.

What does this gives you?

A better memory of your days, and a clearer sense of what actually matters. It gives you a way to notice the important things, those moments that pass too quickly to register, and the ones that happen so often you start to take them for granted.

As a coach and a leader, as a husband and a parent, it gives me a clearer sense of who I am and how I’m showing up with the people around me.

I think that’s a good thing.