I have 6 to 8 hours a day set aside for my sites, a content plan with the topics lined up, an AI workflow written down step by step, and I still can’t start. That is where I am right now, with no job and nobody telling me when to begin. Every guide on how to stop procrastinating assumes you don’t want to do the work, and in my case that’s wrong, because once I start something I keep going until it’s done.
Nobody Asks If You’re Ready When You Clock In
At a job, you clock in. Nobody asks whether you feel ready, because logging in is the start and everything after it is work. Some remote jobs still work this way, VAs have their own clock in software, so the day exists the moment they log in. Flexible remote workers don’t have that, and neither does anyone building their own business, which is me.
Why would I clock in? Flexibility is the whole point, I can start anytime and end anytime. That is also exactly what removed the start. I already wrote about how a flexible work schedule only holds up with structure, and this is the same problem showing up in the first hour of the day.
Six Hours a Day and I Still Don’t Start
Whenever I open the computer and begin, I tend to say I’ll do it later, when I’m ready. Then the day ends and I haven’t done anything. I never get ready, and I’m free to start whenever I’m ready, so nothing ever forces the moment. It has been a couple of days of this now, I planned to start last week, then this week, and I still haven’t started my newest workflow properly, I’m only doing it slightly.
This isn’t avoidance. The problem is the wind up before starting anything. When I start working, I continue until it’s done, that’s how I’ve always been. Procrastination for me is the stretch before the first move, not the work itself.
Every Trigger in My Day Is Aimed at Someone Else
I wake up at 5 AM. We leave around 5:45 so my eldest gets to school before 6, and I’m back home by 6 or 6:15. At 7:45 I take my youngest, her class runs from 8 to 11, so I pick her up at 11. My eldest is in class until 3, so I leave at about 2:55.
Those triggers work because somebody is waiting and being late costs something. The 11 AM pickup takes a few minutes, the 3 PM one takes 20 to 30, and lunch is at my table, though I leave the table on purpose so I’m not stuck there. So my work from home schedule on paper is one long block with two small gaps. But the pickups and everything that needs doing around the house never end, and none of it points at the desk, which also means work life boundaries are something I have to set myself.
The computer isn’t a trigger either. I either never turn it off at night, or I turn it on the second I leave the bedroom and go to the office, then make coffee. That’s a reflex, like a light switch, so turning it on marks nothing.
After the last drop off it’s coffee on the terrace, clearing my head. Then a little browsing: AdSense, Amazon, Facebook to see if someone liked my post, LinkedIn for possible QA topics, the news, sometimes TV. It’s all to prep my mind so I think, hey, I want to write something about this. It feels like prep because it is where my topics come from, but nothing in it ever ends.
Overload Used to Do the Starting for Me
The old problem was that people had too much to do. The house, the kids, the family, and on top of that the report you had to do manually and the form you had to fill up by hand, because that’s what working required. That overload did two jobs. It forced the start because the report was due, and it was the excuse for any delay because everything else was due too.
So when somebody asks why do I procrastinate so much, I don’t think the answer is always character. Sometimes something outside you always started the work, so your motivation to work never had to carry the first step by itself. Take that away and the first step is all that’s left.
AI Took the Pressure and Left Me the Go Signal
With AI productivity tools and AI tools for work, a lot of that manual load is gone. I’ve run my workflow enough times, and optimized it enough, that I give it less context now than I used to. I wrote about the AI productivity paradox separately, this is the day to day side of it.
AI only helps if you can judge what comes back. I’m a QA, I’ve been a PM, I can code, and I still vibe code, because I understand what I’m looking at. If you vibe code with no plan and you can’t debug, the first thing that breaks leaves you stuck. AI takes the grind away, but it can’t start the day for you.
What it leaves is the go signal. I say go and the workflow runs, but I still have to give it my own time and my own situational awareness for it to finish. As a solo operator that’s another problem, because nobody else is going to say go.
How to Stop Procrastinating When You Are the Worker, the Operator and the Boss
I’m the worker, the operator and the boss, so I’m overlapping everything. Every deferral is a management call, I have the authority to say let’s pause this first and do something else. At a job somebody else’s deadline overrules “not now”, and here the only person who can overrule it is the one asking for it.
The workflow made the choice harder in a different way. Because it needs less context from me, I’m not locked into one whole process anymore, I can run several things. That’s where analysis paralysis and decision fatigue come in, picking what goes first every morning with the whole plan open in front of me. The work got cheaper, the starting did not.
Taking the starting decision away isn’t the answer either, it feels like something is getting forced again, and I walked away from that. I don’t want a rigid 9 AM start. I want to start when I start. This is a task initiation problem, getting from not working to working, and it needs a trigger I choose.
The Feed Is Not a Trigger, It’s a Gamble
Starting from Reddit or LinkedIn is a cop out for work, and it’s a bad thing to use as the start because you’re gambling for attention. If something doesn’t catch me, I just keep scrolling. Worse, something can catch me that I’d never write, like a Division 2 post where I think, hey, I gotta fucking try that out, when that’s a cheap, saturated build guide that wouldn’t win the site any long term traffic.
To be fair, it has worked once. I was browsing Reddit, thought oh, this is a good topic for a post, and I started. But nobody said go, it happened sideways, and you can’t build a work day on that.
What I do have is the end of the last drop off, around 8:15 or 8:30. I’m already moving, I’m walking back in, and nothing has been started yet. That’s a better moment than any clock time I could pick.
Build the Trigger You Lost
A trigger is a distinct act with a clear before and after, which is why a VA’s software clock in works. Some people get used to clocking in, some people get up and take a bath, it depends on your preference. These are ideas to test, matched to situations like mine, and none of them is proven.
- If you miss the clock in: write a start line in a notebook or log, the date, the time and the first task. It takes ten seconds and works like logging in.
- If you miss the commute: take a fake commute. Walk around the block or sit outside with coffee, then come back and sit down, and treat the walk back as the arrival.
- If you work better after getting ready: use getting dressed as the marker. Work starts when you’re dressed, not when you feel ready.
- If your day runs on outside obligations like school runs: attach the start to the last one. The drop off ends, you walk in, and the first act is opening the task you already chose.
- If you start by scrolling and never leave the feed: flip the order. One line in the work doc first, browsing after, as input and not as the start.
- If you wait until you feel ready: make the first act too small to need readiness, like writing the title or the first sentence.
- If nobody is waiting on you: make the start visible. Post that you’re starting in a group chat, or work next to someone who’s working.
- If AI made everything easy and there are too many options: pick tomorrow’s first task tonight and leave yourself a note you can swap. It only works if you wrote the note yourself.
Most of these are habit stacking, where you pin the start to something you already do instead of building a new habit from nothing. I wrote the systems version of this on MomentumPath, how to stop procrastinating when your system has no start condition, where the same problem is treated as a missing input in a workflow. A morning routine for productivity works the same way if it ends in a specific act and not a vague feeling of being ready. If you want the longer version of why habits fall apart, I covered why most habits fail and how to build ones that stick in a separate post.
The Research, Short Version
Peter Gollwitzer’s paper on implementation intentions describes an if-then plan that pairs a situation you expect to meet with a specific action you’ll take when it shows up, which is different from a goal intention that only commits you to an outcome. According to the abstract, these plans help people who struggle to turn intentions into action, including failing to get started, getting distracted and slipping into bad habits. The if part is the trigger and the then part is the first action, the same shape as the list above.
On habits, the Lally et al. study had each participant tie a chosen behavior to a daily cue. Habits took an average of 66 days to reach peak automaticity, with a spread from 18 days to a projected 254. Missing a single day had little effect, though many missed repetitions gradually lowered the level people eventually reached. The researchers couldn’t draw clear conclusions about how consistent the cue itself needs to be.
The commute side is thinner. Researchers who study the boundary between work and home argue the commute works as a space where people mentally detach, and in an unpublished follow-up they tracked 80 university employees for a week. Scientific American’s write-up says their suggestion for remote workers is a substitute transition, such as a 15 minute walk at the start and end of the workday. It’s a small sample and not peer reviewed yet, but it’s what the fake commute idea above is trying to do.
Where I Am Now
My trigger is a sequence, not a clock time. The last drop off ends, I have my coffee, and then work starts, with browsing and research counted as work. That fixes what bothered me about the feed, because a quiet feed doesn’t stall the day anymore, since the browsing is already the first task.
I haven’t fixed it. The feed has no built in end, so I don’t have a signal yet that tells me the research phase is over and the next step has begun. I’m testing the trigger and I’ll find out whether it holds. If you have a start trigger and your next worry is that your work looks invisible from the outside, read why remote work discipline is really about job security.
