Today is day 2154 of my daily running streak. It is also day 10 of a 30-day experiment. I am taking a business concept through build, launch and delivery, aiming for profit within 30 days, with AI doing 90% of the work. I count working days only, so each week is five days.

A third of the way in, one thing explains where I stand: the AI business plan. Without it, I would have no idea whether I am on track.

Where the build is today

I am about 50% through building the product. I built that half in ten days, which tells me roughly how long the second half takes. I have also learned more about vibe coding in a structured way, using AI to write the code while I direct the work from a clear plan. I now know what a good instruction looks like and what a vague one costs me.

So I wrote a project plan for the next two weeks. The second half of the build gets the next ten working days. The final ten days cover marketing assets and launch preparation. They also include a large block of migration work, moving the product onto a sturdier back end so it suits clients on a subscription. Onboarding sits in that block too. I want a new customer to join with as little friction as possible.

What the planning tools gave me

Some of the modules I have already built are doing real work for me. One is a task and project management system. Its projects are quarterly, and they come down from an objective and strategy system. In that system I set the vision for ten years, five, three and one, then work backwards into quarters. From the quarters I set monthly milestones and weekly tasks.

The result is clarity. I see what needs doing and by when, for this business and for my real estate business, which runs alongside it. The leadership dashboard I built handles cashflow, forecasting and profit and loss in real time. I use it daily, so I know the work done so far has a real use.

Using the software also shows me the bugs. I find glitches and fix them as I go. That is part of building software with AI, and I would rather meet them now than in front of a client.

Owning the asset

The original model for this business was to build systems and processes on top of external software. Now I own the software. Ownership matters because an owned product is a real asset, and a saleable one. That shapes the whole plan.

The numbers I am working to

I am still setting pricing. The current thinking is a £2,000 implementation fee, because setting this up properly for a client is intensive work, then an ongoing £249 subscription. On those metrics, 30 clients would value the business at around £550,000. I would then look to exit between 18 months and three years from now, and possibly go again with what I have learned. These are targets and projections. They are not results.

The lesson

If you do not have a clear and robust plan, delivering what you set out to deliver is close to impossible. Failing to plan is planning to fail. The planning tools gave me a timeline I trust. Now the work is execution, and if I miss the target, I have only myself to blame. I prefer that. A plan with my name on it leaves no one else to point at.

I see the same pattern in my running. The streak does not run on motivation. It runs on a plan that is simple enough to repeat: one run, every day, in barefoot-style shoes. I do not decide each morning whether to run. The decision is made. Business building works better the same way. Set the plan, break it into small pieces, and do today's piece.

The mission behind it

Every episode of this vlog is part of a fundraiser. I am running a lap of the world, 40,075km, on consecutive days, in barefoot-style footwear, with the aim of raising £1M for children's causes, including Great Ormond Street and BBC Children in Need. With today's run I have covered 16,541.80km. That leaves 23,533.20km. I am working to a streak vision of around 16.5 years, so the pace has to be steady, not fast.

The business matters to the mission too. Building something I own, and building it with a clear plan, gives me more control over my time and resources over the long run. That protects the daily run. The more people who see this vlog, the more the fundraiser gains, so I keep showing up every day.

If you run a business or are curious about building with AI, watch the episode and tell me what you would plan differently. If you believe in the mission, subscribe, share and comment. I will be back tomorrow with day 2155.

Watch the full episode: https://youtu.be/YN1GYuS2-6E