Push, Not Pull
On building systems that arrive instead of asking
It’s 6:15am on a Tuesday morning. The usual chaos. Socks are missing, Foolittle needs his milk topped up, and breakfast is half-eaten on the table. My phone buzzes. A chat notification pops up:
Here’s your schedule for today.
8:30am: School drop-off.
9:15am: F45.
11am: Call with client.
4:30pm: Pickup for swim class.
I didn’t open a calendar. I didn’t tap through three screens to check which kid had what. The information just arrived, like a quiet tap on the shoulder from someone who already knew what I needed.
A week ago, this same morning would have gone differently. I would have remembered, maybe, to check the family calendar while brushing my teeth. Or I would have forgotten entirely, and the swimming slot would have surfaced in my brain at 3:15pm in a flush of panic. The information was always there. I just had to go find it. And on most mornings, with two boys competing for every scrap of my attention, I wouldn’t have the bandwidth to look.
There’s a term for this. Push versus pull. Pull is when you go to the information. Open the app, navigate to the right screen, parse what’s relevant. Push is when the information comes to you. It’s a huge distinction. For anyone running on depleted reserves, the difference between push and pull is often the difference between a system that works and one that quietly collapses because nobody had the spare energy to go check.
This is not a new obsession for our household.
My husband and I have been chasing the frictionless household for the better part of fifteen years. Splitwise, BillPin, YNAB, spreadsheets, portfolio trackers. Shared grocery lists on Evernote, shared family calendars… the list went on and on. We were the kind of people who got excited about a budgeting app. I still remember the big ah-ha moment when we discovered YNAB aka You Need A Budget app. We tracked every dollar. We planned ahead. It was satisfying in the way that good systems are. You could feel the gears turning smoothly.
And then life changed shape.
By the time my second boy arrived, there came the kind of fullness that doesn’t leave room for spreadsheet maintenance (at least for me, my husband doggedly continued to stick to it). The system didn’t break dramatically for me. It just faded. I stopped opening the app and the spreadsheets went stale. The friction of maintaining the system, the pull it required from me every single day, no longer made sense in the broader scheme of things.
I think a lot of parents know this feeling especially. You build a good system. You believe in it. And then the system asks more of you than you can give, and quietly, without anyone announcing it, the system stops.
My husband, who has always been the one driving our household towards what I’d call operational elegance, took last weekend as a sprint1. He set up an AI assistant that automates a variety of household tasks including calendaring, reminders, and expense tracking. We’re still working through the kinks, still adjusting how it categorises things, still figuring out what to ask it and what to handle ourselves. But it’s running. And the fact that “still figuring it out” doesn’t feel stressful is part of the point. The cost of experimentation has dropped so low that imperfection is fine. You tinker on a Saturday, fix something the next day. The shift went from “too costly to try” to “easy enough to iterate.”
Now, when our kids’ school messages us through the school apps (yes, plural, even for one child gosh!), we take a screenshot and drop it into one of our messaging apps. It’s routed into calendars and reminders. When we buy things for the family, I take a photo of the receipt and send it there too. The bot reads it, categorises the expense, and populates our shared spreadsheet. I can ask: what’s coming up this week? And it tells me.
The killer app, really, is the conversation flow that comes with it. Every previous tool had its own logic you had to learn. YNAB had categories and reconciliation flows. Spreadsheets had formulas. The school apps each have their own navigation. But talking is something you already know how to do. That’s what I mean when I say it humanises the interaction. The bot doesn’t have a personality, but it now speaks your language instead of asking you to speak its. The interaction feels less like using software and more like asking someone who’s been paying attention.
And there’s something else worth noticing. Since the bot started pushing my schedule every morning, I’ve stopped checking the calendar myself. It’s convenient, but it means I’m placing trust in a system I didn’t fully build and don’t fully control. Some mornings the chat notification arrives and I think of HAL 9000. Except in our house it’s less “I’m sorry, Dave, I’m afraid I can’t do that” and more “I’m sorry, Yi Ning, I’m afraid you have a dental appointment at two.” The horror isn’t rogue AI. The horror is that it’s right and I forgot. That said, the kinks are real, a miscategorised expense here, a missed detail there, and the friction we removed was also, in its own way, a form of staying engaged with the details of your own life. That’s a trade-off worth noticing, even if it’s one I’d still make.
What’s disappearing isn’t the need for systems. It’s how much attention they require to keep running.
Perhaps we might want to frame all of this as a natural endpoint. A household that finally runs with minimal friction. But I think the more honest truth is that it’s just another iteration. We’ll tinker with Clawdia next weekend, and the weekend after that. Something will break and we’ll fix it. And on some Tuesday morning, the phone will buzz, and I’ll read my schedule, and I’ll think: we’re not done yet.
But we’re closer. Technology is catching up with life.
I should be honest, though. My husband and I are both fluent in technology. Fifteen years of tinkering plus careers in tech means what felt like a casual weekend project may feel like a foreign language to some people. When I say “just talk to your phone,” I’m speaking from a comfort I know isn’t universal. The barrier to entry is lower than it’s ever been, but lower isn’t the same as gone.