I Built My Own Blood Pressure and Exercise App at 71
I was tired of writing my blood pressure on a scrap of paper and losing it. So I built my own app.
Not a real app from the app store. A private page on my own website that behaves like one on my phone — I tap the icon on my home screen, it opens, I punch in the numbers, and it remembers everything. Nobody but me can see it. It cost nothing beyond the time to build it, and it does exactly what I want instead of almost what I want.
There are four screens. One for blood pressure, one for my daily exercises, one for history, and one for supplements.
Why Not Just Use One of the Hundred Existing Apps
The problem with every health app I looked at is that it wants to be everything, and in the process it does none of my specific things.
I do not need a social feed. I do not need it nagging me. I do not need an account, a subscription, or my blood pressure sitting on somebody else's server being sold to whoever buys that sort of thing. And not one of them tracks the particular set of exercises I actually do, or lets me tag a reading with the things I want to compare.
What I wanted was a notebook that does arithmetic. That turns out to be a small thing to build.
The Blood Pressure Screen
I take my pressure most mornings. I enter four numbers and it does the rest.
Systolic, diastolic, pulse, and my sleep score off the ring. There is a weight field too, because I was already weighing myself and had nowhere sensible to put the number.
Then the part that makes it worth having. Three cards across the top show my last reading, my seven-day average and my thirty-day average, each with the category it falls in. A single reading tells you almost nothing — blood pressure bounces around based on what you did an hour ago. The seven-day average is the number that actually means something, and until I built this I never had it.
The Stat Most People Have Never Heard Of
Look at the screenshot and you will notice a number on every card that your doctor has probably never mentioned to you: pulse pressure.
It is the simplest calculation in this whole app. Take your systolic — the top number — and subtract your diastolic. That is it. A reading of 120 over 80 gives you a pulse pressure of 40 mmHg.
Here is why it is worth knowing. Your systolic is the pressure when your heart squeezes. Your diastolic is the pressure when it relaxes between beats. The gap between them is a rough measure of how stiff your arteries have become. Young, elastic arteries expand when the heart pushes blood into them, which absorbs some of that force and keeps the gap small. Arteries that have hardened with age cannot expand, so the systolic spikes higher while the diastolic stays put or even drops — and the gap widens.
That is why pulse pressure tends to climb with age even in people whose blood pressure looks acceptable. Two people can both read 130 over something and be in very different shape. 130 over 85 is a pulse pressure of 45. 130 over 65 is 65, and that second person has considerably stiffer plumbing.
The way I score it:
- Under 50 — great. Green on my screen.
- 51 to 60 — good. Amber. Where I currently sit, if I am honest.
- Over 60 — poor. Red, and worth a conversation with your doctor.
Nothing about this requires a new device or a new test. If you own a blood pressure cuff you already have the number — nobody ever taught you to do the subtraction. Do it on your next reading. I also watch it on my twice-yearly blood panels, where the same stiffness story shows up in other markers. It takes two seconds and it tells you something the two numbers alone do not.
Below that is a thirty-day chart, and underneath it the feature I am proudest of: tag buttons. I can mark a reading as taken after nicotine gum, after modafinil, after alcohol, or after a low sleep score. The app then shows me my average with that tag against my average without it. That is a controlled experiment on myself, running quietly in the background, and no commercial app was going to give me that.
It exports everything to a spreadsheet, so my data is never trapped.
The Exercise Screen
The second screen is a checklist of everything I try to do in a day. Heel raises, Asian squat, grip trainer, Jawzrsize, curls, triceps, twists, red light mask, walk. Several of those have their own write-ups here: daily jumps for bone density, dumbbells on the vibration plate and interval walking.
Some of those are one-and-done. Others I break into sets across the day — the grip trainer is four rounds, Jawzrsize is five, the walk is two. So each of those shows its own counter: 1 of 4, 2 of 4, and it turns green when I finish. That solves the real problem, which is standing in the kitchen at four in the afternoon genuinely unable to remember whether I already did that round.
Every tap saves itself the moment I make it. Close the app, come back hours later, and it is exactly where I left it. There is a running tally at the top of how many of the day's exercises I have checked, a progress bar, and a streak counter for consecutive days.
The Asian squat has a timer field rather than a checkbox, because with that one the number of seconds is the whole point.
The Supplement Screen
The fourth screen is the one I use most when I am not thinking about my health at all.
It is every supplement I take, in one alphabetical list. When a bottle runs out, I tap it and it lights up green. When I restock, I tap it again and it goes dark. A counter at the top tells me how many I need to reorder.
That solves two problems I had not realised were problems. When I am standing in a store or about to place an order, I know exactly what I am out of instead of guessing. And when someone asks me what I take — which happens more than you would think — I pull out my phone and show them the list.
What I Am Building Next: Steps
My goal is 7,500 steps a day. My iPhone already counts them whether I open anything or not, so the question was never tracking — it was getting the number somewhere I would actually look at it.
I am connecting the Health app on my phone to this tracker, so each day's step total lands on the history screen on its own. No typing anything in. Once it is running, I will be able to see my step trend next to my blood pressure and my exercise days, which is the whole point: finding out which of these things actually move together.
That number is deliberate. Ten thousand steps traces back to a Japanese pedometer marketing campaign in the 1960s, not to research. The mortality studies suggest most of the benefit arrives somewhere around 7,500, so that is where I set the bar. Some days I blow past it. The goal is the floor, not the ceiling.
What Building It Actually Taught Me
Three things I did not expect.
The averages changed how I think. I used to react to single readings. A high one would bother me all morning. Now I look at the seven-day line and the individual numbers stop mattering, which is both more accurate and considerably better for my blood pressure.
Tagging found things I would have argued about. When you can see your average on the days you did a thing against the days you did not, you stop guessing. Some of what I believed did not survive contact with the data.
Checking a box is a surprisingly strong motivator. The progress bar at four in the afternoon showing eight of twelve has got me off the couch more times than any amount of discipline.
I Did Not Write a Line of Code
I should be straight about this, because it changes what the rest of this post means to you.
I am not a programmer. I have never written software in my life. What I did was describe what I wanted, in plain English, to an AI — Claude — and it wrote the code.
It went something like this. I said I wanted a page where I could enter my blood pressure and see my seven-day average. It built that. I said the single readings were bouncing around too much and I wanted a thirty-day average too. It added one. I said I wanted to tag a reading as taken after nicotine gum so I could compare. It built the tagging and the comparison cards. I said the numbers were hard to read on my phone in the morning. It made them bigger.
That back and forth went on for a while. Nothing about it felt like programming. It felt like explaining to somebody what I wanted and then saying “no, more like this” until it was right.
The part I did not expect is how much of the work was deciding what I actually wanted. The AI never once pushed back on a bad idea about my own health, and it never asked me why I was tracking something. Those questions were mine. Knowing that pulse pressure was worth showing, that a seven-day average matters more than any one reading, that a comparison needs a minimum number of readings before the number means anything — none of that came out of the machine. That came out of a year of paying attention to my own numbers.
So the honest split is this: I did the thinking, the AI did the typing. That is a real skill in 2026, and it is a skill anybody reading this can pick up. It is also worth knowing before you read the next section, because it completely changes the answer.
Could You Build This
Now you know the real answer: yes, if you can describe what you want. If you have a website it is one page, locked to a logged-in user, with the data saved on your own site and nothing going anywhere else. If you do not have a website, a spreadsheet on your phone does ninety percent of the same job — the averages are the part that matters and a spreadsheet will do those all day.
The point is not the software. The point is that measuring anything consistently beats measuring everything for a week and quitting. I have logged my pressure for months now because it takes fifteen seconds and the numbers are right there waiting.
You cannot fix what you do not measure. I say that a lot. This is me taking my own advice.
Want to see the rest of what I do? Here is my full protocol list, my fitness routine, and every biomarker I track.
Live long and prosper,
Gary
A note from Gary: I am not a doctor, and I am definitely not your doctor. I am a 71-year-old running a 129-year experiment on myself and publishing the results as they come. Talk to your physician before changing anything about your own health.
These are products I use on my push to 200. I only put something on this list after I’ve tested it on myself. The links are affiliate links, which means I may earn a small commission at no cost to you.
See more products →
