If I started over today, I would not try to recreate the path that got me here. That is one of the traps experienced entrepreneurs fall into. We confuse lessons with formulas. The market that formed us is not the market a new founder is entering now.
I founded CI Web Group in 2006, when the internet was still misunderstood by many business owners. If I were starting now, I would assume AI, data, automation, agentic workflows, and machine-readable trust were not future features. I would build them into the foundation.
What I would build differently
I would build the smallest serious version of the company first. Not small in ambition. Small in drag. I would document the entity, the offer, the customer, the proof, the workflows, and the decisions from day one so both humans and machines could understand the business clearly.
I would also design for replacement earlier. Not replacement of humanity. Replacement of founder bottlenecks. Every repeatable explanation, decision, and workflow would be treated as a future system. That is the heart of the modern CEO operating system.
Why starting over requires humility
Starting over means admitting that some of what made you successful would slow you down now. The agency model that once made sense may not be the model the AI era rewards. The org chart that once signaled growth may now signal unnecessary coordination. The founder habits that once rescued the company may now keep it dependent.
Experience is only an advantage if you let it evolve.
That is why I keep coming back to the broken agency model. Starting over today would mean refusing to build a company around yesterday's billable assumptions.
What fails when founders copy their old playbook
The first failure is nostalgia with a business plan. Founders remember the tactics that worked and forget the conditions that made them work. They hire too soon, add complexity too soon, or build around channels that no longer carry the same trust.
The second failure is treating AI as a later phase. A founder who waits to add intelligence until the company is already messy will spend years cleaning up what should have been designed correctly at the beginning. It is much harder to retrofit clarity than to start with it.
What proof guides my answer
My proof is the rebuild. From 320 to 38 taught me that more people do not automatically mean more capability. The right systems, the right standards, and the right operators can create a stronger company than a large organization carrying too much coordination work.
My proof is also the long arc of learning. I wrote Why I Never Stop Learning because staying teachable is one of the few advantages that survives every market shift.
What I would tell a new founder
Build the company as if it will need to be understood by people, customers, search engines, AI systems, and your future self. Name things clearly. Keep proof close. Make decisions visible. Do not create mystery where structure would serve you better.
And do not worship size. Build something useful, intelligent, and honest enough to adapt. The future does not need bigger versions of old companies. It needs clearer builders.