A daily routine with an AI assistant (a workflow with AI) does one job. It walks with you through the place where your energy keeps dropping — the morning, the inbox, the afternoon admin. It holds. (See the pairing map for which workflow goes with which moment.)
An experiment does a different job. It grows a capacity you don't have yet. Not holding what's there — building what isn't.
Here's the frame I use for each experiment week. The same frame Track D of the Flow Design vault runs on every experiment.
Five parts
1. The question (one line, handwritten)
A question specific enough that you can live it for 7 days. Not "am I happy?" — but "if I don't check Stripe before noon, how does my day go?"
Write it by hand. There's something about the drag of the pen that keeps you from making the question too clever. A good question has three signs:
- Specific (observable)
- Within reach to do differently (not asking about the world — asking about your behavior)
- You don't know the answer (if you do, it's not an experiment)
2. The hypothesis (2–3 sentences)
What do you guess will happen? Write it down. Guess specifically. After 7 days, hold it against what actually happened. This part teaches the most — not the result, but the gap between what you expected and what was true.
3. The signal (one sentence)
What will you look at to know if the experiment is working? A number? A feeling? A behavior? The more specific the signal, the more usable the data.
4. Daily capture (5 min, handwritten or voice)
One line. How did the question play out today? No analysis. Just the record.
One small fragment a day. Seven days, seven fragments. That's the raw data — nothing more is asked of you.
5. End-of-week review (15 min)
End of day 7, sit down. Read the 7 lines. Answer:
- Was my hypothesis right? Wrong where?
- What does the signal say?
- I — after these 7 days — changed what?
- What's the next question?
The last question matters most. A good experiment often doesn't answer its own question. It opens a deeper one.
An example
Week 12 — "No Plan, All Heart"
- Question: If every morning I don't plan the day — only do whatever feels most right at the moment — does my day collapse?
- Hypothesis: Chaotic for the first three days, then self-correcting.
- Signal: At the end of the day, do I feel I lived it, or got dragged through it?
- Result: Chaos wasn't three days — only a day and a half. Then a different rhythm appeared. I decided differently with two clients. I wrote a piece I hadn't planned to write.
- Next question: That "most right at the moment" feeling — where does it come from? Is it different from the "afraid of being wrong" feeling?
Why this is Flow Design's frame
Because knowing doesn't change you. Living changes you. One experiment a week is a small piece of living. After a year you have 50 pieces — a fairly detailed map of where your energy rises, where it drops, what set-point you keep returning to, which pattern keeps protecting you. The shape of your own flow, drawn from the inside. It's also a concrete way to walk the full flow journey more than once.
The first experiment, if you want one: get the free chart and live your Strategy (the way your type is built to engage with life) and Authority (how your body, not your mind, is built to decide) for 7 days. Underneath the energetic signal, those two are the chart-level confirmation of how you're meant to move.