Shipping evenings-only: the discipline that lets a side empire survive a day job

There’s a whole genre of side-project advice that assumes you have either a) twelve free hours a day or b) the kind of partner who’ll do all the dishes for a year. I have neither. The empire ships because I learned what to cut, not what to add.

This is the honest list.

The actual schedule

That’s roughly two to three hours a day on the empire, split across morning and evening. The morning slot is for hard problems (architecture, debugging, new code). The evening slot is for low-stakes work (publishing, redirects, dashboard checks, replies).

The split matters more than the total. Three hours in one block at 11pm is worse than ninety minutes at 6am plus ninety minutes at 9pm. Cognitive fatigue compounds.

What I cut

To make those three hours productive, here’s what I stopped doing:

What survives

Three things stay in the schedule no matter what:

1. Saturday review (45 minutes). I open the empire dashboard, scan traffic numbers, decide what to work on next week. Not “everything I want to” — what fits in the slots I actually have. If I skip this, by Wednesday I’m working on whatever feels urgent, which is rarely what’s valuable.

2. Sunday content (90 minutes). One long-form post or one batch of small content (snippets, notes). The week’s traffic flows through whatever I publish Sunday. If I skip Sunday, Monday’s traffic dips noticeably.

3. Tuesday deploy (whenever the morning slot allows). Whatever was built over the weekend ships Tuesday. Friday deploys have bitten me too many times — Saturday’s family time becomes “fix the regression” time. Tuesday gives me four working days to catch issues before the weekend.

The energy management problem

This is what most “side project” advice misses. The hard problem isn’t time — it’s energy. A 9-hour day job consumes most of the cognitive capacity available for the day. By 9pm, hard problems are out of reach.

I match the work to the energy:

• 6am block (peak energy): new code, architecture decisions, hard debugging
• 9pm block (depleted energy): publishing, redirects, dashboard reviews, replies to support emails, design polish on existing UI

If I try to write code at 9pm, I produce code I have to revert at 6am the next day. The 9pm block has rules: no git push origin main without re-reading the diff in the morning.

The family negotiation

Honest part most posts skip.

Three things made the family-empire balance work:

1. No laptop after dinner before 9pm. The two hours between dinner and the second work block are family-only. No “I’ll just check this real quick.” Phones face-down. The empire can wait 120 minutes.

2. Weekend Saturday = family. Weekend Sunday = 4-6 hours empire. Split, not blended. Trying to “work a little, then play with kids, then work a little more” makes both worse.

3. Public commitments. Telling family “I’m shipping X by next Friday” turns deadlines into shared knowledge. Slipping a deadline becomes a conversation, not a stealth crisis.

The compounding bet

The reason any of this works at all: side empires compound on a different curve than full-time work.

Full-time: linear effort → linear results. Two hours twice as good as one hour.

Side empire: linear effort → compounding results, but only past month 12. Months one through twelve produce nothing visible. Months thirteen through thirty-six produce most of the result. Most people quit in month nine.

I started this empire in March 2024. Most of the traffic, most of the revenue, most of the meaningful content arrived in 2026. Three years of evenings to escape velocity.

The cost — said plainly

This isn’t a victory lap. There’s a real cost to evenings-only shipping:

• You don’t get to be a normal developer. Conferences? Sometimes. After-hours hackathons? No. Open-source contributions to projects you love? Limited.
• You miss things. Some birthdays, some school events, some shows everyone watched. Not many, but some.
• The day job suffers a little. Energy management means I’m sometimes 80% at work instead of 100%.
• Side income is small for a long time. Year three revenue is ~$1,300/year. Below minimum wage if you count hours.

It’s a long bet. The bet only pays out if you survive the boring middle.

What I’d tell past-me

If you’re starting an evenings-only side empire:

1. Pick a stack you can sleep with. Boring matters more than optimal.
2. Build one platform thing before you build product things. Even if “platform” is just shared auth.
3. Cut tools, not commitments. Fewer frameworks, fewer accounts, fewer dashboards.
4. Saturday is for review, Sunday is for content, Tuesday is for deploy. Not negotiable.
5. Tell your family the timeline. Real timeline, not the optimistic one.

The empire doesn’t win because I have unusual time. It wins because I cut everything that competes for the time I do have. That’s the only real discipline.