External systems, not willpower: surviving solo with a newborn

In short — Willpower is a resource that runs out, especially with a newborn. The only way to last as a solo is to build systems that run independently of your mental state in the moment.

---

A baby arrives in a few weeks. I’ve known for nine months, obviously. But it was only in the seventh month that I asked the real question: if I disappear for three weeks, what keeps running on its own at SEK — and what collapses?

The honest answer, at that point, was uncomfortable. Too much still depended on me. Not on my skill, not on my strategic decisions — just on my physical presence, on my ability to open a tab and click publish.

That’s the problem with willpower as infrastructure.

The myth of founder discipline

Discipline doesn’t scale. Systems do.

We glorify the disciplined founder — the one who gets up at 5 a.m., batches their tasks, keeps a spotless Notion. It’s a nice story. It works under lab conditions: no illness, no surprises, no baby on the way.

The solopreneur reality is that your capacity to execute fluctuates constantly. Some weeks you’re in flow, you produce in 4 hours what normally takes a full day. Other weeks you’re drained, you procrastinate on trivial tasks, you forget simple things.

If your business depends on your “good week” self, it’s structurally fragile.

James Clear popularized the idea that systems beat goals — but most solopreneurs stop at the motivational version of that idea. They make checklists, routines, rituals. Better than nothing. But a checklist still depends on you to get checked off.

An external system, on the other hand, runs whether you’re there or not.

The distinction is brutal but necessary: a routine is discipline in disguise. A system is infrastructure.

Everything that depends on your willpower breaks at the first disruption

The first real test of a solo business is the first serious disruption — and a newborn is a 3-week disruption at minimum.

I’m not talking about a sick day. I’m talking about a stretch where you’re available unpredictably, in 45-minute blocks, at any hour. Where your brain runs at 60% of its usual capacity from lack of sleep. Where priorities reshuffle completely, with no notice.

Under those conditions, here’s what breaks first:

Content publishing. If you publish manually, you stop publishing. Not because you don’t want to — because the slot you planned it for collides with a feed, a cry, or an hour of sleep you can’t give up.

Client follow-up. Emails waiting for a reply “when I have time” pile up. The time doesn’t come. Clients read the silence as indifference.

Technical monitoring. A production bug can sit active for 48 hours if nobody surfaces the alert. As a solo, if you’re not watching, nobody is.

Invoicing and reminders. Late payments don’t settle themselves. If you don’t send the reminder, it doesn’t go out.

Each of these is a dependency on your willpower. And with a newborn, willpower is the scarcest resource there is.

The real inventory of systems that run on their own at SEK

Here’s what keeps working at SEK with no manual action from me.

This isn’t a theoretical ideal — it’s the actual state of the system at the time I’m writing this article, a few weeks before the birth.

Content publishing — 100% automated. Journal articles are written, reviewed, and scheduled ahead of time in the CMS. The publishing pipeline needs no action at go-live. Content goes out on the planned day, whether I’m available or not. AI-assisted writing lets me produce and sign off on several articles in one session — I batch, I schedule, I disappear.

Error monitoring — automatic alerts. Every SEK product is wired to an alert system that surfaces critical errors (500s, timeouts, conversion rate dropping abnormally) straight to my phone. I don’t need to go looking for the problem — it comes and finds me. If I’m unavailable, I see the alert when I pick the phone back up, and I can decide whether it’s urgent.

SEO fixes — semi-automated. A weekly script scans the main pages, checks for missing meta descriptions, images without alt text, broken links. It generates a report. I don’t have to run the scan myself — I get the report, I decide what to fix first. Detection is automatic, judgment stays human.

Payments and subscriptions — zero intervention. Every SEK product that generates recurring revenue goes through Stripe. Failed-payment retries are automatic (three spaced attempts, automatic email to the client). I don’t touch anything unless a client reaches out directly.

Replies to incoming requests — automatic first filter. A contact form with an automated email sequence sends an immediate reply with the essential information (timelines, rates, process). It doesn’t replace the human conversation, but it prevents total silence during unavailable stretches.

What these systems have in common: they don’t depend on my mood, my energy level, or my being there at a specific moment. They run on triggers (a date, an event, a threshold) — not on my decision in the moment.

What they don’t do: make strategic decisions, handle a complex crisis, write content in my place with my judgment. The human stays essential — just not for every repeatable operational task.

How to decide what to systematize first

The rule: systematize first what’s frequent and low in human-added value.

Not what’s urgent. Not what’s impressive. What’s frequent and repeatable.

Here’s the filter I use. For every recurring task, I ask three questions:

1. Do I do exactly the same thing every time? If yes, it’s automatable. If each occurrence needs different judgment, it’s human by nature.

2. What happens if it doesn’t get done for 3 weeks? If the answer is “nothing serious” → low priority. If the answer is “lost revenue or a lost client” → high priority.

3. How much time does it take me per week, all in? Tasks that feel quick but happen daily are often the most time-consuming in reality. 10 minutes a day = 60 hours a year.

Apply this filter to everything you do over 2 weeks. Write down every task, even the small ones. You’ll uncover patterns you hadn’t seen.

Then prioritize in this order:

1. What generates revenue or protects existing revenue (payments, reminders, critical alerts)
2. What maintains the client relationship (automatic replies, follow-up)
3. What keeps you visible (content publishing, SEO)
4. What gives you information without you hunting for it (monitoring, reporting)

Everything else can wait. Perfectionism on systematization is itself a form of procrastination.

An honest caveat: building these systems takes time upfront. Budget 2 to 4 weeks of focused work to go from a “depends on me” business to a “runs without me” business on operational tasks. It’s an investment, not an instant fix. If you wait for the birth to start, it’s too late.

To go further on the numbers behind this approach — how many solopreneurs systematize, what impact on revenue — the solopreneur & AI statistics 2026 roundup compiles the sources that carry authority.

The test: can you disappear for 3 weeks?

Ask yourself this question now, not when the disruption hits.

This is the ultimate reality test for a solopreneur. Not “do I have systems?” — but “does my business survive my total absence for 3 weeks?”

Three weeks is the minimum for a serious parental leave. It’s also the length of a long illness, an urgent family situation, a mild burnout. It’s the duration past which an unprepared absence starts doing real damage.

To pass the test, four conditions have to be true at the same time:

Content keeps going out. Not necessarily at the usual pace, but regularly. A 3-week silence on a blog or newsletter drops organic traffic and disengages the audience. The fix: have a buffer of scheduled content before you disappear.

Critical bugs surface without you looking for them. If you have to go check manually that everything works, you can’t disappear. Monitoring has to be push, not pull.

Recurring payments land with no manual action. If a single payment needs your intervention to fire, you have a leak in the system.

Incoming leads get a reply within 24 hours. Even automatic, even partial. Total silence for 3 weeks costs you potential clients who won’t come back.

If you answer “no” to even one of these four questions, you have a critical human dependency. Not a discipline problem — an infrastructure gap.

The full test — I’d suggest running it as a simulation: pick a Friday, don’t touch anything in your business for the day, and watch what didn’t get done. It’s a miniature version of the 3-week test, and it surfaces hidden dependencies faster than any analysis.

---

I don’t know exactly what the first weeks after the birth will look like. Nobody really does. What I know is that the business can’t rest on my ability to keep the pace under those conditions.

Systems don’t replace the founder. They free the founder from repeatable execution so they can put their attention where it actually matters — decisions, direction, the human relationship. And, in my case, being present for Kelly, Émilie, and this boy who’s just arrived.

That’s the theory of constraints applied to real life: the family constraint isn’t a handicap. It’s what forces you to build better.

If you want an outside look at what’s actually blocking in your technical setup, the SEK audit is built for that — an honest diagnosis of what holds and what breaks. And if you’re stuck on a specific implementation, Unstuck exists to get you unstuck fast, with no long commitment.

---

Sébastien de Bollivier has been a freelance dev since 2008 — sebastiendebollivier.com

Got an idea to ship? A website, a SaaS, an AI automation — built with you.

Talk about your project