A 5K in 28:25 on September 27. 26:28 seven days later. Same hill, same shoes, same stretch of sidewalk.
My ninety-day target was 26:00, and I'm 28 seconds off it after one week.
I did not get two and a half minutes fitter in seven days. Nobody does. So before anyone congratulates me, here's what actually happened.

26:28. The watch called this one hard.
The number that changed was effort.
Day 0: average heart rate 157. My watch scored the effort 6 out of 10, moderate. My quads and knees were still sore from squatting six days earlier.
This week: average heart rate 162. The watch scored it hard. Fresh legs. Same course, same conditions.
The difference was that I went out planning to race it. Last week I went out planning to finish.

The numbers. The rest of the story is in the heart rate.
What made me plan to race it was one word.
When I looked at the day 0 data, "moderate" sat there next to a time I wasn't proud of. I'd written in the issue that a moderate score meant 28:25 was a floor rather than a ceiling. Then I had to go run again knowing I'd said that in public.
So I picked a better playlist and went hard.
There's a version of this problem in every lab. Put a thermometer into a small sample and the thermometer changes the sample's temperature. The act of measuring moves the thing you're measuring, and a good method accounts for it instead of pretending it isn't happening.
Publishing that number became one of the conditions of the next run. The clock didn't make me faster. Knowing someone would check made me try.
What the hard part actually felt like.
Around the climb, it stopped being a run and started being a negotiation. Legs burning, chest tight, the part where your brain starts offering you reasonable-sounding reasons to ease off.
What I do there is narrow everything down to my breathing. Count it, hold a rhythm, make the exhale longer than it wants to be. All my attention goes onto the one thing I can still control.
I used to think that was about getting more oxygen. It isn't. At that effort my blood is already carrying close to all the oxygen it can. What's driving the urge to gasp is carbon dioxide building up and my blood turning more acidic, and no amount of breathing technique fixes that in the moment.
So the breathing isn't doing much physiologically. It's doing almost everything attentionally. Pain expands to fill whatever attention you give it. Counting breaths gives my attention somewhere else to go.
The climb is fixed and the breath is yours.
That’s the hill. Here's the part that transfers, and it took me a while to see it.
Think about a business quarter that's going badly. By week eight, most of the inputs have stopped being yours. The market is what it is. Your pricing was set last year. Headcount is frozen. The deal that slipped to next quarter already slipped. You can spend the remaining weeks resenting all of that, and plenty of people do, and the resentment costs real energy and returns nothing.
What's still yours is smaller and more boring. Which three accounts you call today. Whether your team knows what matters this week or is guessing. Whether you tell your boss the number is going to miss now or on the last day.
Most of what burns people out at work isn't the difficulty. It's attention spent on inputs that stopped being theirs weeks ago. That's not a mindset problem. It's an allocation decision.
Why a 40-year-old is timing himself in public.
I'm not trying to become a fitness guy. I'm eighteen years into a career, I run revenue for a company with real investors behind it, and I have a four-year-old and a one-year-old at home.
What I've noticed is that all of it runs on one tank. The quality of my judgment in a Tuesday pricing meeting is downstream of how I slept, how I trained, and what I've been carrying. I've spent most of my career treating my capacity as a given and optimizing everything else around it. That worked until it didn't.
So I'm measuring the tank. Ten numbers, ninety days, published either way.
There's a version of this conversation online that's aimed at 25-year-olds with no dependents and infinite runway. This isn't that. I'm building something that outlasts my current role, in the hours that are actually left over, at an age where recovery is slower and the stakes at home are higher.
If you're somewhere similar, the mechanics matter more than the inspiration. That's what I'm trying to write down.
I'm not changing the prediction.
26:00 on December 20 is probably too easy now. I'd be surprised if I don't beat it.
Moving it would be the obvious thing to do and the wrong one. A prediction you revise when the first week goes well isn't a prediction, it's a running commentary. The whole point of writing ten numbers down in public was to be gradeable, and part of being gradeable is being wrong in the generous direction sometimes.
In December I'll tell you I set that one too low.
Two things I don't know.
How much of the 1:57 was effort and how much was fresh legs. I can't separate them. One run isn't a dataset, and I changed two variables at once, which is bad method and I knew it while I was doing it.
And whether chasing the number is going to hurt me. That's the real risk here. A published prediction creates pressure, and pressure is how a 40-year-old with a garage gym and two kids ends up with a strained hamstring in week three. Ninety days doesn't work if I'm hurt in week two.
So the rule I'm setting now, while I'm not under pressure: I race the 5K once a month. The rest of the runs are runs.
What I'm testing next.
My course has 225 feet of climbing on sidewalk concrete. I've been wondering what the same effort looks like on a flat track with some give in the ground.
I'm going to go find out, and that number won't replace anything. The benchmark course stays the benchmark, because changing the measurement to get a better result is exactly what I spent last week saying I wouldn't do. The track is curiosity, logged separately.
What's one advantage you could compound from here?
Hit reply and tell me. I read everything, and what you send shapes what I write next.
— Austin
Austin Stevenson is CRO at ACT Global Lab Services and a co-founder of Vertosa.

