The Idea-to-Market Machine: How I Plan to Ship Products Over The Next Year
March 2026 · The framework behind everything
After my first product made zero sales, I had a choice: feel sorry for myself, or build a system that makes failure useful.
I chose the system.
Here's the mental shift that changed everything: I'm not building a product. I'm building a machine that produces and tests products. The individual product doesn't matter as much as the rate at which I can try things.
Why This Matters
AI has collapsed build time. With Claude Code, I can ship a full-stack product in 1-2 weeks of evening sessions. That changes the maths completely.
The old way: spend 6 months building, launch, pray it works. If it doesn't, you've lost half a year.
The new way: build in 1-2 weeks, test for 6-8 weeks, decide honestly, move on. I've already completed 2 full cycles in under 3 weeks. The build phase is fast. The testing phase is where the real time goes.
The bottleneck is no longer building. It's finding product-market fit. And the only way to find it is volume.
The Four Phases
1. Ideate (Ongoing)
This runs in the background all the time. I keep a running list of problems I spot in daily life, at work, in conversations. For each one I ask three questions:
- Who has this problem?
- Would they pay to solve it?
- Can I reach them?
I talk to 5 real people before building anything. Ideas can brew for weeks or months. No pressure to act immediately.
One hard-won lesson from Product #2: check legal viability early. My YouTube Knowledge Base got all the way to deployment before copyright risk nearly killed it. Now I check for platform TOS, IP issues, and business model legality before writing a single line of code.
2. Build (1-2 Weeks)
One idea at a time. Ship the ugliest version that works. If it takes more than two weeks, the scope is too big.
Must-haves: working product, payment flow, landing page. That's it.
Do not polish. First users don't care about design. They care if it solves their problem. I built Confident You in about three weeks. That was too long. YouTubeKB took two evening sessions for the core tool. That's more like it.
The stack: Claude Code for everything, Stripe for payments, whatever hosting makes sense. The tools don't matter. Speed does.
3. Market & Test (6-8 Weeks)
This is the phase that matters most and where my skill gap is biggest.
- Week 1-2: Launch to warmest audience (email list, personal network)
- Week 3-4: Expand reach (cold outreach, content, communities)
- Week 5-6: Talk to every user. What do they love? What's missing?
- Week 7-8: Final push or wind-down based on signals
Here's the critical insight: AI does not compress this phase. Claude can build the product in days. But trust-building and market feedback happen at human speed. You can't automate someone deciding to pay you.
This is the muscle to develop. Every cycle makes the next launch better.
4. Decide (1 Week)
Three outcomes, no ambiguity:
- Revenue growing + users engaged? Scale it. This is the one.
- Some revenue + low maintenance? Keep it running passively, move on.
- Low/no revenue + unclear demand? Archive it. Write the retrospective. Extract the lessons. Move on.
The danger zone is ambiguous middle results, where you're tempted to "just try one more thing." I set the decision date in advance. When it arrives, I decide. No extensions.
What Every Cycle Produces (Even If It Fails)
This is the part that makes the machine work even when individual products don't:
- A shipped product that proves you can build and launch
- A case study with honest content that builds credibility
- Marketing reps where the skill gap closes with each attempt
- Audience growth as each launch puts you in front of new people
- Pattern recognition so you get better at spotting what works
- Reusable infrastructure like email templates, launch playbooks, code you can reuse
My first product (Confident You) solved dozens of problems I'll never face again: email segmentation, Stripe setup, auth flows, API rate limiting. Product #2 was faster because of Product #1. Product #3 will be faster still.
The Scoreboard So Far
| Cycle | Product | Build Time | Revenue | Key Lesson |
|---|---|---|---|---|
| 1 | Confident You (AI public speaking coaching) | ~3 weeks | £0 | Don't AI-wrap services that need human connection |
| 2 | YouTubeKB (transcript search) | ~2 weeks | £0 | If your pivots remove the differentiator, the product is dead |
| 3 | TBD | Distribution first this time |
Two cycles complete. Zero revenue. But I can now ship a full-stack product from idea to payment page in under two weeks. That capability is the real product of the machine.
The Reframe That Keeps Me Going
The biggest fear isn't that one product fails. It's never finding the idea that scales to £100K side income.
The answer isn't finding a better idea. It's building a system that produces enough tested bets that one inevitably breaks through.
The system is the strategy, not any single product.
The pace is already faster than I expected. Two cycles in three weeks. The machine works. Now it's about finding the right idea to put through it.