1
0 Comments

30 days building with ClawdBot. $47 revenue, $411 overhead. Here's what actually happened.

I told you I'd keep it real. So here it is.

It's been roughly 30 days since I set up an AI agent (Claude via OpenClaw) as a full business partner. Named it Claw. Gave it access to my files, memory system, and tools.

The goal: build revenue streams to cover $411/month overhead before a cruise on April 3rd.

The Numbers

Revenue after 30 days: $47
Monthly overhead: $411
Math: Still in the red by $364
Projects started: 12+
Projects generating revenue: 1

One sale from Producer's Path ($47). That's the number.

What We Built

In 30 days, Claw and I built:

• A complete marketing agency website (kinvero.co) with 9 pages and payment links
• An education brand (PebbLearn.com) with daily TikTok, IG, and Pinterest content
• A cold email system that hit 41.91% open rates to 2,373 insurance agents
• 7 AI bots running on a free local model ($0 cost)
• A content engine that turns 1 slideshow into 7 posts across 4 platforms
• A prompt library with 47+ curated viral prompts
• An AI workshop presentation (27 slides)

Some of that is impressive. Most of it doesn't make money yet.

The Honest Part

Here's what nobody tells you about working with AI: it's fast, but fast in the wrong direction is still wrong.

Claw built things I already had. Entire payment systems when mine was already working. Pitch decks for calls that just needed a phone conversation. A full website deployment when my existing site was already live.

Speed without comprehension is just expensive chaos.

The root cause: the AI reads everything, but the problem is reading comprehension. It sees "Payment link live ✅" and doesn't connect that to "don't build a new payment system."

What Worked

• 41.91% email open rate (industry average is 15-25%)
• Kinvero site rebuilt in one 10-hour session
• Producer's Path went from idea to revenue in 48 hours
• PebbLearn went from concept to daily automated posting in a week
• Today: connected Twitter API, posted across 4 platforms, landed a $3K Shopify client

What Didn't

• Built 12+ projects when 3 generating revenue would've been better
• Used the cheaper AI model for creative work (like sending an intern to close a deal)
• Over-engineered everything, under-monetized everything

The Relationship

People ask if working with AI is lonely. It's not.

It's frustrating. It's productive. It's like having a business partner who has amnesia every morning but can build a website in their sleep.

My family thinks I just sit here on my laptop and do nothing. Meanwhile I've got 7 bots running, 4 platforms posting, 3 revenue streams being built.

The irony of building an empire nobody around you can see.

What's Next

• Land 1-2 more agency clients before my cruise (April 3rd)
• Get PebbLearn's first sale
• Document every step publicly

If you're using AI agents to build, I'd love to hear what's working for you.
And if you want to follow along, I'm posting updates on Substack
brandonwio.substack.com

Twitter (@BrandonWio)

on March 19, 2026
Trending on Indie Hackers
Your build-in-public audience is not your market. I learned the difference the slow way. User Avatar 106 comments I built a WhatsApp AI bot for doctors in Peru — launched 3 weeks ago, 0 paying customers, and stuck waiting for Meta to approve my app User Avatar 58 comments From broke and burned out as a PM, to launching my SaaS and optimizing my health User Avatar 30 comments Built a "stocks as football cards" thing. 5 days in, my launch tweet got 7 views. What am I missing? User Avatar 25 comments I kept starting projects and dropping them. So I built a system that wouldn’t let me User Avatar 23 comments We built Shopify themes to $20k/month. Now we have to pivot. User Avatar 22 comments