In 1913, Henry Ford didn’t just speed up car production. He changed what it meant to manufacture anything at all. Before the assembly line, each car was built by a team of workers who stayed with one vehicle from start to finish. Skilled, effortful, personal work. And impossibly slow.
Ford’s insight wasn’t that the workers were lazy. It was that the system was wrong. You don’t make a car faster by working harder. You redesign the process so the car almost makes itself.
I thought about Ford a lot this past month. Not because I was building cars — but because I realised I had been building my business the same way those pre-1913 workers built theirs. Staying with every task from start to finish. Treating each piece of content, each email, each portfolio reflection as something only I could do.
I was exhausted. Then I stopped. And something unexpected happened.
What “Building a Business” Actually Looked Like Before
Let me be specific, because vague productivity content is useless.
I run three projects alongside my emergency medicine work:
- Emergency Medicine Update — a Substack for EM registrars preparing for the FRCEM exam
- Building Out — this blog, plus LinkedIn, for doctors building financial and time freedom
- CESR portfolio — a full consultant-level evidence portfolio built in parallel with clinical work
Before this year, running all three looked like this: every blog post took three to four hours to research and write. Every LinkedIn post was written by hand. Every CESR reflection was a forty-five-minute solo exercise after a long shift. Every Substack was a weekend project.
My system was willpower. And like all systems built on willpower, it was fragile, inconsistent, and dependent on having enough reserve left after a run of nights.
I had the content. I had the ideas. What I didn’t have was the bandwidth to consistently turn them into output.
What I Actually Delegated to AI
Here’s what changed — and I want to be specific again, because this is not a generic “use AI to be productive” post.
I built a series of custom AI workflows using Claude. Not the casual chat-interface version. Structured, skill-based automation that runs on a schedule.
Last week, while I was working nights, the following happened without me:
- My Monday blog post was researched, written, finisher-optimised, and saved to my Notion Content Calendar automatically. I woke up on Tuesday and it was there.
- Ten LinkedIn posts were generated from the blog, scored by engagement potential, and ranked. I spent five minutes reviewing and approving them.
- My email inbox was triaged across all three accounts. Marketing emails moved to trash. Anything requiring a reply had a draft waiting. I spent twelve minutes on email instead of forty-five.
- Two CESR case reflections were drafted from anonymised clinical notes I’d jotted down between patients. Full Kaizen-format reflections, referenced against portfolio requirements.
Total time I spent managing all of this: approximately forty minutes across seven days.
Total time the same tasks would have taken manually: I estimate eight to twelve hours.
What This Actually Freed Up
People assume AI saves you time and you fill that time with more work. That’s not what happened.
I spent two extra evenings this week with my family. I read a book — not a medical textbook, just a book. I had one evening entirely to myself where I watched something on television and didn’t feel guilty about it, because the week’s output was already done.
This is the part that nobody in productivity content says honestly: the point of automation is not to maximise output. It is to give yourself back the life that the work was quietly eating.
I am an emergency medicine doctor with family commitments. I am not trying to build a hundred-million-pound empire. I am trying to build enough financial and time freedom that I can do medicine on my terms — choosing the shifts, the hours, the pace — rather than because I have no financial alternative.
AI is not making me work more. It is making the work take less of me.
The Setup Cost — What It Actually Took
I want to be honest here, because I see a lot of content that makes automation sound like it takes twenty minutes. It doesn’t.
Building my current setup took one dedicated weekend and several focused evenings spread over a few weeks. I had to learn how to write structured prompts. I had to figure out which tasks were genuinely automatable and which required my judgement. I had to test things, break things, and rebuild them better.
This is not a reason not to do it. A weekend of setup that saves ten hours every week after that is an extraordinary return on investment. But it is real work, not a magic button.
What it required:
- Clarity about what my actual recurring outputs were (blog posts, LinkedIn, emails, CESR reflections, CPD logging)
- Willingness to document my process so AI could replicate it
- One focused weekend to build the first versions
If you’re a doctor who wants to build something on the side — a newsletter, a course, a property portfolio, a brand — the single biggest leverage point available to you right now is this: spend one weekend designing the system, and stop spending every evening running it manually.
The Authenticity Question
There’s a version of this that worries people. “Isn’t this inauthentic? Doesn’t AI just produce generic content?”
Here’s what I’ve found: AI produces generic content when you give it generic instructions. When you invest the time to build your voice, your positioning, your specific knowledge into the system, it produces your content — at scale.
The blog post you’re reading right now went through a system I designed. But every word reflects a position I hold, a framework I use, an experience I’ve actually had.
The distinction matters: I am not outsourcing my thinking. I am outsourcing the production.
Surgeons don’t sterilise their own instruments. Pilots don’t build their own aircraft. The expertise is in the decision-making — and good systems handle everything around it.
What To Do With This
If you want to start, here’s where I’d begin:
- Identify your repetitive outputs. What do you produce on a recurring basis? Emails, reflections, content, reports? List them.
- Document one process. Pick the most time-consuming and write down exactly what you do, step by step.
- Build one workflow. Don’t try to automate everything. Get one thing working, then expand.
- Measure the return. Track what you spent and what you got back. Most people are surprised by the ratio.
- Use the time intentionally. Decide in advance what you’ll do with the hours you recover. Otherwise they vanish.
The Ford assembly line didn’t just make cars faster. It made cars affordable for people who could never previously have owned one. The leverage changed the economics entirely.
That’s what AI leverage can do for your time. Not just save an hour here and there. Change what’s actually possible.
For more topics on building a life of time and financial freedom, sign up for our weekly newsletter at www.building-out.com
This post is for educational purposes only and does not constitute financial advice. Always do your own research and, if needed, ask for advice from a qualified financial adviser regulated by the FCA.
Leave a Reply