TL;DR
- Three weeks of data is noise. Three months is a map. Wait for the map.
- Four patterns worth looking for: the weekly low, the dominant pillar, the unexpected lift, the seasonal floor.
- The mistake that ruins the read: looking for what you already believe instead of what the data actually says.
- One small action per pattern is enough. Don't redesign your life from a heatmap.
Three weeks vs three months
People look at their mood data after two weeks and either declare it useless or read everything into it. Both are wrong. Two weeks is too noisy — a bad cold, a single rough conversation, and the whole graph is skewed. Three months is when the signal lifts above the noise. Six months is when you can argue with your own beliefs about yourself.
Be patient. The data is not for the daily — it's for the retrospective.
The four patterns to look for
Once you have 90 days, four things tend to surface, in this rough order of obviousness:
- The weekly low. One day of the week consistently sits below the others. Often not the day you'd guess. Sunday afternoon and Wednesday are the usual culprits, for different reasons.
- The dominant pillar. One pillar correlates with your overall mood more than the others. When it's high, the week is high. When it's low, nothing else compensates. That's your load-bearing pillar, and it deserves disproportionate care.
- The unexpected lift. Something you didn't think mattered turns out to. Often it's a pillar you'd dismissed as "minor" — a creative outlet, a small social ritual — that's quietly carrying you.
- The seasonal floor. One month is reliably worse than the others. The first time you notice it, it's interesting. The second year, when February or November land the same way again, it's actionable.
Most people stop at #1. The deeper payoffs are #3 and #4.
The pattern you'll resist seeing
The most useful pattern is usually the one you don't want to see. The thing you've told yourself for years is energising might show up as draining in the data. The relationship you defend as "fine" might correlate with the lowest weeks. The career commitment you take pride in might be where the floor lives.
Don't argue with the data on first sight. Sit with it for two weeks before deciding what to do. The instinct to dismiss is itself information — it usually means the pattern is true and uncomfortable.
What to actually do once you see something
The most common mistake at this stage is doing too much. You spot a pattern and immediately want to restructure your week, change your pillars, write a manifesto. Don't.
One small experiment per pattern is enough:
- Weekly low on Wednesday? Protect Tuesday evening for one month and see if it changes.
- Career is the dominant pillar dragging you down? Pick one meeting to move and look at the next four weeks.
- Seasonal floor in February? In January, plan one small visible thing — a trip, a project, a book — to anchor the dip.
The experiment is small enough that you can read the result. That's the point.
A note on overreading
Some weeks are just bad. Some months are just bad. Mood data isn't a horoscope — it doesn't tell you who you are; it tells you what the last 90 days looked like. Hold the patterns loosely. They're a starting point for a question, not an answer.
The honest framing: this is a thoughtful person noticing themselves more carefully. That's all. And that's plenty.
Related reading: Our approach to journaling · How to choose your five pillars
← Back to all resources