One Place Where the Truth Lives

Last time I said I’d show you how I built it. The honest answer is that I didn’t build it all at once, and I didn’t build it cleverly. It started as one spreadsheet with a clunky dashboard bolted on. It grew from there, a piece at a time, each piece earning its place by solving a real problem. That became the rule: start crude, keep what works, and only add the thing last week proved was missing.

The first decision was the one that mattered most. There is one place where the truth lives. Everything goes into a single sheet, every day, so there is nowhere to hide. The reason diets quietly fall apart is that the evidence is scattered and easy to ignore. You half-remember the week as “pretty good”, the scale disagrees, and you don’t know which one is lying. Put it all in one place and the arguing stops. The number is the number.

One gap runs the whole thing: what I eat against what I burn. I track my weight every morning in the same state, after the toilet and before the coffee, taking every ounce of credit I can get. But weight is the gauge, not the engine. The engine is the calorie deficit. The burn, active and resting, comes from my Apple Watch. I don’t treat it as gospel. No wrist gadget deserves that. But checked against the weight trend, it is good enough to show whether the gap is real. Log the food, pull the burn from Apple Health, and the difference between them is what matters.

The logging itself is deliberately dull. Through the day I note what I eat, mostly keeping an eye on calories and protein. I record the training. And I log the beer, honestly, because leaving it out would only fool me. It asks you to be honest, not perfect, and those are very different jobs.

Then the sheet turns it into a picture, and this is the part that stopped me quitting. A single day’s weight is noise. You can be two pounds heavier on a Tuesday because of a salty dinner. None of it is fat. So I don’t look at days, I look at the trend, a rolling line that smooths the wobble out. Once you can see the line heading the right way, one bad morning stops being a verdict and goes back to being weather.

I also don’t trust any single gadget. The home scale gets read every day, but every few weeks I check it against a body-scanner app, and now and then a proper scan at the gym, usually every four weeks. They never agree exactly, and that’s fine. I’m not after the perfect number, I’m after agreement on direction. When the scale, the app and the scanner all point the same way, I believe them. When one of them does something odd on its own, I ignore it. Three honest witnesses beat one confident liar.

The spreadsheet came first, dashboard and all, and it is still the spine. But I am not always at my laptop, so I built a stripped-back view for my phone that lets me log a meal in seconds, wired up through iPhone Shortcuts and refreshing through the day. Later I built the proper dashboard, the one I live by now. It turns everything into trends, projections, and a weekly read on what is working.

I should be straight about how these got built. I do not write code. I specified what I wanted and directed the build, the way I have directed delivery teams for twenty years, except this time the team was AI. That is the part that surprises people, and it is the part that means you do not have to be technical to do this.

Not all of it worked. We spent a while trying to pull weight and calorie burn straight out of Apple Health automatically, and in the end gave up because it was more trouble than it was worth. That is the method in miniature: keep the parts that earn their place, and bin the rest.

The piece that turned a pile of data into an actual system came later. A few weeks in, with the numbers stacking up and not much being made of them, I added an automated weekly review. That quickly became the engine.

Once a week the setup takes the week’s data and tells me, in plain terms, whether things are still moving the right way and what the single thing to change is. One change, not five. It has an aim, a rate of loss I am looking for, and a few trip-wires, the rate stalling, the drinking creeping up, muscle dropping, so a soft week gets caught early instead of three months too late.

And it does something I did not expect to need: it tells me when the scale is lying in my favour as much as against me. Some of my best weeks have shown almost nothing on the scale, because the fat was coming off while muscle built underneath. Reading one number, I would have called those weeks a failure and probably packed it in. Reading the trend across more than one measure, the review could see they were the opposite.

None of this had to be clever to work. It began as one honest spreadsheet, and that first version, the original Excel dashboard, is the thing I want to hand you, stripped back to the bones, so you can point it at your own life and start seeing the real picture for yourself. That is where this goes next.