The cursor blinks on a half-finished portrait of a dog named Captain. $200 deposit. Due Friday. It is Tuesday, 2:43 in the afternoon, and I have spent the last forty minutes not drawing Captain but updating a spreadsheet row that tracks the commission’s progress, toggling between three browser tabs to compare reference photos I already saved to a folder on my desktop, and reconsidering whether the invoice template I built last March needs a revised late-payment clause. Captain’s left ear is finished. His right ear does not exist yet. The spreadsheet says I am 60% done, which is a number I made up to feel like something is being measured.
The apartment is quiet except for the radiator, which started its annual clicking two weeks early this October. My desk is a door on two filing cabinets — a setup I copied from a cartoonist I met at Portland Zine Fest in 2022 who swore it was the cheapest standing desk in the city. She was right. The filing cabinets cost $23 at the ReBuilding Center on Mississippi Avenue. The door I found on the curb outside a house on Alberta Street. The desk holds a Wacom tablet, a lamp from Goodwill, a jar of ink I have been using since 2021, and the lid of an oatmeal container I use as a palette. These are the tools I can name without embarrassment. These are not the ones taking up most of my attention.
The tools taking up most of my attention are the fourteen application icons in my dock. I opened six today. I used two. There is the spreadsheet. The invoice generator. The time tracker I installed during a panic in April when I thought if I could just see how many hours I was losing to administrative tasks, I would stop losing them. I have not opened it since May. There is the project management app I used for exactly three weeks before it became another place to avoid looking at unfinished work. A note-taking app with 847 notes in it, 840 of which I will never read again. And a drafting app with a half-written client email I started composing on Sunday, stopped mid-sentence, and have been avoiding because I cannot figure out how to say “I need an additional week” without sounding like I am failing.
I am not failing. I am also not drawing Captain’s right ear.
What I am doing is something I do not have a clean name for. It is not procrastination, exactly, because the spreadsheet needed updating and the invoice template probably does need a late-payment clause after the thing with the bakery in June. It is not client work, because Captain’s portrait is the client work and I am not doing it. It is not rest, because my jaw is tight and I have checked the time eleven times in the last hour. It is the administrative labor of maintaining a creative practice — the emails, the tracking, the organizing, the planning, the tool-evaluating, the contract-reading, the folder-structuring, the password-resetting, the app-trying, the app-abandoning — and it has quietly become most of what I do on any given Tuesday.
That same discipline applies to editorial structure: before publishing, editors need a way to test scattered notes become an argument readers can follow, which is where an AI writing software that fits the project can function as a planning aid rather than a substitute for domain evidence.
I did not used to have fourteen apps. When I started freelancing four years ago, I had a sketchbook, a laptop, and a Google Doc where I kept a list of people who owed me money. That was it. The list was ugly and effective. I checked it on the first of every month, sent awkward emails, and got paid or did not. The system had no dashboard, no color-coded status flags, no automated reminders. It was a list. It worked because it was too simple to avoid using.
Then I started reading about how other freelancers organized their businesses. I read a post by an illustrator in Seattle who used a Kanban board for every commission, with columns for “Inquiry,” “Quoted,” “In Progress,” “Revisions,” and “Delivered.” It looked clean and professional in the screenshots. I set one up the same afternoon. I moved two projects into it, felt a surge of competence, and then never updated the board again. The columns sat there with their two lonely cards while I went back to the Google Doc. But I did not delete the board for eight months, because deleting it would mean admitting I had spent two hours building something I never used.
This is the pattern. I encounter a tool that promises to make the business side of freelancing feel less like drowning. I adopt it with a kind of desperate optimism. I use it for somewhere between three days and three weeks. I stop using it. I do not delete it. It sits in my dock or my bookmarks bar or my app drawer, accumulating a small charge of guilt every time I see it. The guilt is minor. The cumulative weight is not.
There is a term for what I am describing, and it does not come from the creative world. In site reliability engineering — the discipline of keeping large-scale software systems running — there is a formal distinction between substantive engineering work and the repetitive, low-value operational tasks that surround it. Google’s engineers call that second category “toil,” and they have written entire chapters about how to identify and eliminate it, because they noticed something that sounds familiar: the tools designed to reduce operational overhead can themselves become a form of that overhead. The system you build to track your work becomes work. The dashboard you create to monitor your tasks becomes a task. The automation you install to save time costs time to maintain. I read about this in a site reliability engineering reference from Google one night when I was supposed to be answering client emails, and I recognized my own desk in it with a specificity that was not comforting.
The engineers have a vocabulary for this. I have a dock full of icons and a vague feeling that something is wrong.
The something that is wrong is not that I use tools. Tools are fine. The invoice generator saves me twenty minutes per invoice. The spreadsheet, when I actually update it, tells me whether I can afford to buy the good watercolor paper this month or whether I need to stick with the student-grade Cotman set I have been using since college. This month: student-grade. The good paper is Fabriano Artistico at $32 a block at Muse Art and Design on Hawthorne, and I buy it when I have had a month over $1,800, which has happened twice this year. The tools are not the problem. The problem is that the tools have started shaping what I consider worth doing.
I noticed this most clearly with the time tracker. When I installed it, I told myself I was gathering data. I wanted to know how much of my workday was spent drawing versus how much was spent on email, invoicing, scheduling, file management, and the other unnamed tasks that fill the hours between picking up a pen and putting it down. What I found, after three weeks of tracking, was that I was spending roughly 40% of my working hours on drawing and 60% on everything else. This was useful information. It was also devastating information, because the time tracker did not just measure what I did — it started judging it. Every hour I spent drawing began to feel slightly wasteful, because drawing was the thing I was supposed to be doing and therefore the thing that did not need tracking. The administrative tasks, by contrast, felt productive in a way that drawing did not, because they generated data. They filled cells. They produced charts. The time tracker turned my creative practice into the thing that happened between the tasks that counted, and I started unconsciously prioritizing the tasks that counted.
I stopped using the time tracker. I did not delete it. It sat in my dock for three more months, its icon a small reminder that I had tried to measure my way out of a problem that was not really about measurement.
The problem, I think, is that I have been treating the administrative labor of freelancing as something separate from my creative practice — as overhead, as tax, as the cost of doing the work I actually care about. But the overhead has grown so thick that it is now the thing I do, and the drawing has become the break I take between emails. The ratio has inverted, and I did not notice until I found myself on a Tuesday afternoon with a half-finished dog and a fully updated spreadsheet and no clear sense of which one was the work.
Both are the work. I know that. The emails are how I get commissions. The invoices are how I get paid. The contracts are how I avoid getting exploited. The spreadsheets are how I know whether I can buy the good paper. None of this is optional. A freelance illustrator who does not handle the business side is a freelance illustrator who does not eat. But there is a difference between handling the business side and letting it colonize the desk, the dock, the attention, the day. There is a difference between using a tool and being used by it.
I think about this when I look at the drafting app with the half-written email in it. The email says: “Hi Sarah, I wanted to let you know that I’ll need an additional week to finish Captain’s portrait. The left side is complete and I’m really happy with how his ear turned out, but I want to make sure the right side matches before I send the final. Would a deadline of October 18th work for you?” I wrote that sentence on Sunday. I stopped because I was afraid Sarah would think I was unprofessional. I stopped because the drafting app suggested a more formal revision that sounded like a press release. I stopped because I started wondering whether I should restructure the email to lead with the new deadline, then the reason, then the reassurance — a structure I read about in a freelance communication guide that I have never successfully applied because it makes me sound like a project manager and not a person drawing someone’s dog.
The drafting app is one of several AI writing software tools I keep open in tabs during a typical workday, none of which I have fully committed to and none of which I have fully rejected. There is one I keep meaning to actually evaluate for the client communication side of things, but I have not gotten past the stage of opening it, staring at the empty input field, and closing it again because I cannot think of anything to type that I do not already know how to say. The drafting app I did use, once, generated three versions of an email declining a commission I did not want to take. All three versions were polite. None of them sounded like me. I sent my own version, which was also polite and also did not sound like me, because declining work never sounds like the person doing the declining. It sounds like someone trying to survive.
I keep these tools open the way I keep the time tracker in my dock — not because I use them, but because I have not yet decided whether to trust them, and deleting them would mean making a decision I am not ready to make. The decision is not really about the tools. It is about what kind of creative practice I want to have. Do I want a practice that is organized, systematized, optimized, and efficient? Or do I want a practice that is messy, slow, occasionally unprofessional, and mine?
The honest answer is that I want both, and the tension between those two things is where I live most of the time. I want to send invoices on time. I want to answer emails within two business days. I want to have a contract that protects me when a client disappears without paying, which has happened twice. I also want to spend an entire afternoon drawing Captain’s right ear without checking a spreadsheet, without updating a status flag, without opening an app that will measure how long it took and tell me whether that was an efficient use of my time. I want the business side to be small and quiet and the drawing side to be large and loud. Instead, the business side is large and loud and the drawing side is small and quiet, and the tools I installed to fix this have made it worse.
There is also the ethical dimension, which I have been avoiding thinking about because it is easier to avoid thinking about it than to think about it. The AI writing tools I have been half-using are built on large language models, and those models were trained on text that was not freely given. The Authors Guild has published guidelines for writers navigating AI tools that are worth reading if you are in this same liminal space, because they make clear that the foundational models underlying most of these apps were trained on pirated, unlicensed books without compensating the authors. I am an illustrator, not a writer, but I am also a writer — I write emails, contracts, grant proposals, artist statements, and now this — and the idea that I might be using a tool built on stolen work to draft an email declining a commission feels like a specific kind of hypocrisy I am not sure how to sit with. The guidelines are not rules. They are offered as a way to help people make their own informed decisions. I have not made mine yet. The tabs stay open.
This is the part where a productivity blog would tell me to delete everything and start fresh. Do a digital declutter. Keep only what you use. Be ruthless. I have read those posts. They always feel like they were written by someone who has never stared at a tool they might need someday and felt the specific anxiety of throwing away a safety net they have never actually used but might need in the month three clients pay late at once.
I am not going to delete everything. I am going to close the spreadsheet. I am going to close the drafting app with the half-written email in it and write the email myself, in my own voice, which will be slightly too casual and will probably not lead with the deadline. I am going to leave the time tracker in my dock because deleting it would feel like a declaration I am not prepared to make. I am going to pick up the pen and draw Captain’s right ear, and I am going to try to notice whether the shape of his ear matches the shape of his left ear without checking a reference photo I already have saved in a folder I can find without opening an app.
The radiator clicks. The light at 3:47 has moved to the spot on my desk where the good paper would go if I had bought the good paper. Captain looks at me from the left side of the page with one finished ear and one empty space where the other ear will be. The spreadsheet says 60%. The drawing says something else. I am not sure which number to trust. I am not sure it matters. The ear needs to be drawn either way, and the tools that could help me draw it are not the ones that will.
I pick up the pen. I do not open anything.
