The hazard of working alone is not loneliness, exactly. Loneliness is immediate and obvious — you notice it the way you notice cold. The real hazard is a subtler erosion: the slow drift in your calibration that happens when no one around you is pushing back on your assumptions, mirroring your schedule, or noticing that you have eaten the same meal three days running and called it "optimizing." You do not go feral all at once. You go feral incrementally, one unchallenged habit at a time, and the trouble is that the drift feels like productivity for a while because you are not being interrupted.
Solo work has genuine advantages — no open-plan offices, no meetings that could have been emails, no colleagues who materialize at your desk to talk about the weekend. But the same isolation that keeps the interruptions out also keeps the feedback out, and without feedback, small dysfunctions compound undetected until the day you surface from a two-week sprint and realize you cannot quite remember what you were supposed to be building. What follows are five rules, each specific enough to be actionable rather than aspirational.
Rule 1: Schedule one meeting per week that you cannot cancel
The meeting does not need to be about work. It needs to be a standing commitment with a real person — coffee with a friend, a call with a former colleague, a weekly walk with someone who notices if you cancel twice in a row. The purpose is the fixed point on the calendar that you structure the rest of the week around. When you work alone, the days telescope — Monday and Thursday become indistinguishable, and the week becomes a featureless block of time easy to lose track of. One uncancellable meeting per week reinstalls the scaffolding that an office provides by default.
The specific constraint matters. A dentist appointment works for one week. A standing Thursday morning coffee with a peer who is also building something — who will ask what happened to you if you bail — works for fifty-two.
Rule 2: Write down what you did, not what you plan to do
Most productivity systems are future-facing: here is the list, here is the plan, here is what tomorrow looks like on paper. Solo workers need the opposite more urgently — a backward-facing record of what actually happened. Spend five minutes at the end of each working day writing three to five sentences describing what you completed. Not what you intended, not what you started. What you finished.
The backward-looking log does something the to-do list cannot: it tells you the truth about where your time actually went.
This serves two functions. First, it gives you a realistic picture of your actual output over time, which turns out to be both more and less than you imagine. Second, it provides the raw material for calibrating your estimates. If you have logged fifty days and can see that "refactor the authentication flow" reliably takes three days when you estimated one, that data is worth more than any productivity framework. The log is not a journal. It needs to be honest.
Rule 3: Define the end of the workday in advance, not in the moment
When you work from home, the workday has no natural terminus. There is no commute that marks the end, no office lights going off, no colleague putting on their coat. The only thing that ends the day is a decision, and decisions made in the moment — especially by people who are in the middle of something interesting — tend to get deferred. "Just one more thing" is the phrase that has cost more lost evenings than any other four words in the remote-work vocabulary.
The rule is to set the end time in the morning, write it down, and treat it as fixed. Not "I'll stop when I feel done" — that feeling arrives approximately never. Not "I'll stop at a natural stopping point" — there are no natural stopping points in software; the work is fractal. Six o'clock. Or five-thirty. The specific time matters less than the habit of deciding it before the day starts, when your judgment about time is not yet corrupted by momentum.
Rule 4: Keep one item on your list that belongs to someone else
Accountability structures fail solo workers for a predictable reason: they are voluntary. You can join a mastermind group and stop showing up. You can sign up for a coworking space and stop going. The commitments that stick are the ones with a real person on the other end who is blocked until you deliver. The most reliable way to build this in is to keep at least one item on your active list that someone else is waiting on — a contractor waiting for your feedback, a user waiting for a reply, a collaborator waiting for your draft.
This sounds like advice to create more work for yourself. What it actually does is wire in an external obligation that holds even when your internal motivation is weak. Some days the only reason to open the laptop is that someone else needs you to. That is not a failure of discipline. It is a structural feature of how humans work, and building it in deliberately is better than discovering its absence when you have been unproductive for three weeks and cannot explain why.
Rule 5: Leave the building at the same time on at least two days per week
Leaving the building means leaving — not stepping onto the porch, not walking to the mailbox. A walk long enough that you pass several strangers. A run. A trip to a coffee shop where you sit and do nothing for an hour. The physical break matters, but what matters more is exposure to the ambient world outside your project: strangers talking, the texture of weather, the fact that life is proceeding at its ordinary pace outside the walls where you have been living inside a problem.
Solo workers who stop going outside do not notice the moment they stop. They notice it three months later when they realize that every conversation they can remember happened through a screen, and that their opinions about the product, the industry, the competition have calcified quietly. The world does not care about your product the way you care about your product. Leaving the building is the cheapest available reminder of that.
The underlying problem all five rules solve
Every rule above answers the same structural problem: when you work alone, you are both the worker and the manager, and the manager half tends to atrophy first. The worker half is easy to sustain because the work is interesting. The manager half — scheduling, honest output evaluation, external relationships — requires friction from the outside world to stay calibrated, and solo work removes that friction by design.
None of these rules is a cure. Working alone is what it is, and some of its discomforts are the price of its advantages. But the discomforts are manageable, and managing them deliberately is better than waiting to notice you have gone feral and working backward from there.