The single most helpful tactic for managing change

At the beginning of 2025, I was diagnosed with Type 2 diabetes.

I'm sharing that because it became one of the most useful professional lessons I've had in years. Personal, yes. Relevant to work, more than I expected.

The behavioral science literature has a name for what ended up helping me most: observation without judgment. It's a concept with deep roots in clinical practice and mindfulness research. What the diagnosis gave me was a high-stakes context where I had no choice but to actually use it.

What observation actually means

Observation means tracking what's happening before you decide what to do about it. No conclusions yet, just an honest account of what is actually true right now.

Most of us skip this step. We feel something is wrong and move straight to solutions with more willpower, discipline, or goals. Then we're confused when nothing changes, because we were solving a problem we hadn't actually diagnosed.

After my diagnosis, I started wearing a continuous glucose monitor. It tracks blood sugar in real time, all day. I didn't change anything at first. I just watched.

Within a few weeks, I had data I never would have guessed. Certain foods I thought were fine were causing significant spikes. Others I assumed were problematic weren't. My body wasn't doing what I expected, and I couldn't have known that without watching first.

There's a phrase I use in coaching sessions: one time is anecdotal, two times is coincidence, three times is a pattern. It's a useful discipline for staying in observation mode. It slows down the instinct to fix before you actually understand what's happening.

Bringing AI into the observation process

Once I had enough data, I brought it to an AI. My question wasn't "what should I do?" It was "what patterns do you see?" Keeping the question that narrow keeps you in analysis mode rather than solution mode, and it keeps the AI from defaulting to generic advice.

For this kind of pattern analysis I've come to trust NotebookLM more than other tools I've tried. It works directly from your source material and quotes it back to you rather than synthesizing from general training data, which makes it significantly more useful for finding patterns in your own observations rather than someone else's.

The patterns that surfaced helped me build a meal plan that actually fit how my body responds. My energy stabilized. My numbers improved. I did this under the supervision of a licensed clinician, with the data in hand.

I've used the same approach in my work for years, tracking what's happening in client sessions, looking for patterns, then adjusting. The glucose monitor just made the stakes concrete in a way that was hard to ignore.

The system

Start with one thing you want to understand: your energy through the day, how you show up in high-stakes meetings, or how long it takes you to get into deep work. Pick one and observe it without trying to fix it.

Track consistently. It doesn't need to be sophisticated. A note at the end of each day, a short voice memo, a simple log. The goal is data. Insight comes later.

Then bring in NotebookLM or a similar tool to look for patterns. Ask what repeats. Ask where things shift. You're not asking for a prescription, you're asking for a mirror.

This maps to the COM-B framework from behavioral science: Capability, Opportunity, and Motivation each affect Behavior differently. Changing behavior requires knowing which one is actually limiting you, and observation is how you find out.

What this looks like in practice

One of my clients was struggling with creative blocks. She'd show up to design sessions flat and distracted and didn't know why. We ran a simple observation exercise: she tracked her energy and focus at three points during the day for two weeks.

The pattern was clear once we had the data. Her best thinking happened between 9 and 11am. Everything after 2pm was maintenance work at best. She'd been scheduling her hardest creative sessions in the afternoon because that's when she had open calendar blocks.

She moved her protected design time to mornings. Her output changed. Her experience of the work changed. No willpower required, just one structural adjustment based on what the data actually showed.

Start before you're ready

You don't need a diagnosis to use this. You don't need a tracker or a lot of data.

You need one question worth answering and the patience to watch before you act. Most change fails because people are solving the wrong problem. Observation closes that gap before you spend energy in the wrong direction.