The Support System I Wish I'd Had (And What I Did Instead)

If you're doing this without a network, I want to save you a step: you don't need the network you think you need.

I spent a lot of early weeks feeling like I was missing something. Other people building things seemed to have a scene — a group chat, a demo day, a cohort, a mentor who "gets it." I had none of that. I had a wife who was supportive but busy, two kids, a business to run, and a golf group that mostly wanted to talk about the golf.

So I made a smaller, weirder support system out of what I had. It worked.

What actually helped

• The AI itself. Not for emotional support. Just for the fact that it never got tired of my questions. That alone is worth more than most "mentors" I've ever paid for.
• One or two internet strangers. People I found in comment sections and DM'd. Not friends. Just other builders who'd reply. That was enough.
• Writing it down in public. This blog. Not for readers, at first. For me. Turning the mess into sentences forced me to see it clearly.
• A walk after dinner. Not glamorous. Every problem I've actually solved, I solved on a walk.
• My wife saying "how'd today go." That's it. That was the whole executive coaching program. One sentence, most nights.

What I stopped waiting for

• The perfect cofounder.
• The right accelerator.
• Someone "in tech" to take me seriously.
• Permission from a person I'd never actually met.

I'm not saying communities and mentors are bad. They're great, if you have them. But if you don't — and a lot of us don't — please hear this clearly:

You can build the thing anyway.

The support system that gets a first version shipped is often just: the tool, one person who loves you, a notebook, and the willingness to sit down again tomorrow.

That is not a lesser version of the journey. That is the journey.

The actual weekly structure I run

It took about eight months of flailing to land on something repeatable. This is it, and it is deliberately unimpressive:

• Monday, 20 minutes — pick the one thing. Not a list. One outcome for the week. If I cannot name it in a sentence, I do not understand it well enough to build it.
• Weeknights — build, no decisions. Decisions get made Monday. Weeknights are for execution, because tired me makes terrible strategic calls (see: every 11:45pm note I have ever written).
• Friday, 30 minutes — written review. What shipped, what did not, and the question that earns its keep: what am I avoiding? The answer is almost always the un-fun commercial work.
• Sunday — nothing. This one took the longest to accept. Seven-day weeks produced worse output than six, every time.

What I would tell someone starting with zero support

1. Get one witness, not a network. One person who knows what you said you would do this week. That is 80% of the value of a cofounder for none of the equity.
2. Join two communities, contribute in one. Lurking helps less than you would think. Answering one question a week from someone three weeks behind you does more for your own confidence than anything else I tried.
3. Write in public before you feel ready. This journal started as documentation and turned into the support structure. People find you because you wrote the thing they are currently googling at midnight.
4. Pay for one thing that removes friction. Not ten tools. One. Mine was the AI builder subscription; everything else could wait.
5. Track something other than money at the start. Days shipped, posts written, conversations had. Money is a lagging indicator and it will tell you to quit long before it tells you the truth.

The honest summary

I did not find a support system. I built a lightweight one out of a public journal, two online communities, one person who asks me what I said I would do, and a weekly half hour of writing to myself.

It is not a team. It is enough. And it is available to anyone reading this, starting today, for basically nothing.