Sometimes the Best Way Forward Is to Stop

Two unrelated ideas have been sitting with me for a few weeks.

One is from organizational research on treating language as data. The other came from an episode of The Happiness Lab about sabbaticals. I didn't expect them to connect. They did, and the connection kept nagging at me: am I actually paying attention to the signals telling me where to go next, or am I just building?

Organizations leave clues everywhere

I came across Soren's research on using language analysis to understand how organizations are really doing.

What struck me wasn't the use of big data itself. It was the idea that ordinary language — the words people use in emails, messages, reports — can become a dataset. Instead of relying only on surveys asking people how they feel, you can study what they naturally say and look for patterns at scale.

Organizations are constantly producing clues about themselves. Data isn't always a spreadsheet. Sometimes it's language. Sometimes it's behavior. Sometimes it's the thing people keep circling back to, or the thing they've quietly stopped mentioning.

The ukulele problem

Then I listened to Laurie Santos interview DJ DiDonna about sabbaticals.

DiDonna described something he calls "identity play." When people step away from their normal professional routine, they start experimenting — learning something strange, making something, trying an idea they've been carrying around for years, not because it will become their next career, but because they're finally allowed to.

One story stuck with me. During a sabbatical, DiDonna played silly songs on a ukulele. No TAM analysis. No pitch deck. Just play. His research suggests that kind of aimless experimentation restores creativity, because people discover they can step outside their established identity, try something uncertain, and be okay.

Founders aren't wired for that. We're trained to ask what the outcome is, whether it scales, what the ROI is. Apparently, doing something purely because you're curious is still allowed. I needed the reminder.

A career is a dataset too

If language leaves clues about an organization, our experiments leave clues about ourselves.

Look back over a few years. What subjects keep pulling you back? What do you build when nobody asks you to build anything? What problem annoys you enough that you want to fix it? That's exactly the pattern I see across Curious Garden, Vowelly, and the other things I keep starting at Elel — none of them commissioned, all of them chasing the same underlying curiosity about how people learn and communicate.

One data point alone means nothing. Over time, patterns emerge. The mistake is expecting every experiment to succeed — that's like deleting every observation that doesn't support your hypothesis. Some experiments tell you what to pursue. Others tell you what to stop pursuing. Both are useful data.

Even companies need sabbaticals

There's a part of the episode that hit closer to home than I expected.

When someone important takes extended leave, the organization has to figure out how to function without them. Knowledge gets transferred. Weak processes get exposed. That sounds inconvenient. It's actually a pretty good stress test — and DiDonna's point is that this transition happens eventually anyway, whether by choice or not.

Right now, if I stepped away from Elel for even a week, most of it would stop moving. That's not a badge of honor. It's a warning sign. If the thing you built can't run without you for a while, you haven't built an organization yet — you've built yourself a very complicated job.

Stopping isn't the same as avoiding

Here's the part I had to sit with, though: stopping and avoiding look identical from the outside, and they are not the same thing.

Jeff Bezos made this distinction back in a 2001 interview. His claim is that stress doesn't come from hard work — you can work brutal hours on something you love and feel fine. It comes from something else entirely: not acting on something within your control. Or, as he put it more simply, "stress comes from ignoring things you shouldn't be ignoring."

His fix is almost too small to count as advice. Make the first phone call. Send the first email. Write the first paragraph. You don't have to solve the problem — you just have to stop pretending it isn't there. The moment you're addressing it instead of avoiding it, the stress drops, whether or not it's resolved yet.

That's the line a sabbatical has to respect. A real break is stepping away from the treadmill on purpose, with your eyes open. Avoidance is dressing up procrastination as rest. One restores you. The other just delays the stress and adds interest.

So the useful question isn't only "what am I curious about?" It's also: what have I been quietly not dealing with, that one concrete step would defuse? Sometimes the answer is to build. Sometimes it's to stop. Either way, the test is whether you're facing the thing or hiding from it.

Stop. Look. Experiment.

One finding from DiDonna's research is almost funny: among his Harvard Business School classmates, close to 90% worried about how taking a career break would look to others. Fewer than 5% actually cared when someone else took one.

We spend a lot of energy managing an audience that isn't watching.

I don't think a good product, or a good career, gets designed perfectly in advance. It emerges through observation and iteration — build, measure, step away occasionally, notice what changed, build again. Sometimes progress means pushing harder. Sometimes it means changing direction. And every so often, the most productive thing you can do is put the roadmap down and pick up the metaphorical ukulele.

You never know what the data will tell you next.