I spent three days on a problem I should have escalated in three hours. Not because it was hard — I’ve solved harder. Because I’d built a working identity around figuring things out myself, and asking felt like breaking character.
By the time I finally sent the message, the answer took twenty minutes. Someone else had hit the same wall six months ago. The fix was documented. I’d been reconstructing a solution that already existed, in isolation, because I’d convinced myself that asking was the cost and figuring it out was free.
It wasn’t free. It cost three days.
The autonomous instinct
There’s a version of autonomy that’s actually pride dressed up as diligence.
I learned this the hard way. Somewhere in the last few years, I internalized a rule: try everything yourself first. The rule was useful when I was junior — it’s how I learned the shape of problems. But the rule didn’t update as I got more senior. It stayed at “try everything” when it should have moved to “try the right things, escalate the rest.”
The result is that I treat asking as a last resort instead of a tool. I’ll spin on something for a day, hit a wall, and instead of pinging the person who knows, I’ll open a fourth documentation page and try to derive the answer from first principles. As if the cost of the message is higher than the cost of the day.
It almost never is.
Why asking feels expensive
I know where this comes from. Asking feels like an admission. Every time I escalate a question, I’m saying: I don’t know this thing. And in a culture that rewards figured-it-out energy, not knowing feels like a tax on how I’m perceived.
So I delay. Not consciously — I don’t decide to wait. I just generate plausible reasons to try one more thing first. Let me check the logs again. Let me reread this thread. Let me see if there’s a pattern in the last ten commits. Each of these is a reasonable step. Strung together, they become avoidance wearing the costume of diligence.
The test I’ve started using is blunt: if I’m on my second hour of a problem and I know someone who could answer it in five minutes, why am I still working alone?
The honest answers are never good ones. Because I don’t want to look like I didn’t try. Because they might be busy. Because I should be able to figure this out.
None of those are about the work. They’re all about how I want to be seen.
What the delay actually costs
The cost isn’t just the three days. The cost is the shape of those three days.
When I’m stuck on something I should have escalated, my work degrades in a specific way. I become cautious. I stop starting new things because the unresolved thread is eating background attention. I check the stuck problem more often than the work warrants, because maybe I’ll see something new this time. I don’t. But I keep checking.
Three days of that is not three days of focused effort. It’s three days of low-quality work spread across multiple things, none of which get my full attention because part of me is still grinding on the thing I should have handed off.
And then I send the message, and the answer arrives in twenty minutes, and I feel stupid. That’s the part I was actually avoiding — not the asking, but the feeling of asking. The work could have been done Monday. I made it a Thursday problem because Monday-me didn’t want to feel like someone who didn’t know.
The threshold I started using
I don’t have a clean rule. I have a threshold, and it’s looser than I used to think it should be.
If I’m blocked and someone exists who could unblock me, I ask. Not after trying everything. Not after a day. After I’ve done the work that makes the question good — tried the obvious thing, checked the docs, reproduced the issue. Once the question is specific enough to be worth answering, it’s worth asking. The time between “good question” and “sent message” should be minutes, not hours.
I used to think the bar was: try everything first. Now I think the bar is: make the question answerable. Those are different standards. The first rewards suffering. The second rewards clarity.
A good question isn’t a failure of autonomy. It’s the output of autonomous work — the thing I produce before involving someone else. The asking isn’t the cost. The asking is the delivery.
The thing I had to admit
The hard part was admitting that my autonomy had become a kind of isolation. I’d built a working style that optimized for not needing anyone, and then I’d started applying it to situations where not needing anyone was the actual problem.
Asking isn’t the opposite of autonomy. Asking is what autonomy is for — you do the work that makes the question sharp, then you route it to the person who can move it. That’s not dependence. That’s what senior practitioners are supposed to do.
The junior version of me tried everything alone because he didn’t know who to ask or how to frame it. The version I’m trying to become tries the right things alone, frames the question, and asks. The difference isn’t capability. It’s honesty about where the leverage actually is.
What I’d tell a younger version of myself
If you’ve done the work to make the question specific, you’ve earned the right to ask it. Don’t sit on it for a day because asking feels like a step down.
The three days I burned were not the cost of the problem. They were the cost of my pride. And pride doesn’t ship anything.
Ask earlier. The work you’re protecting by waiting isn’t the work. It’s the version of yourself you’re trying to look like. That version can afford to send the message.