I run a screen printing shop in Reno. Before this, the shop ran on spreadsheets, a QuickBooks tab, three supplier websites and a whiteboard nobody updated. So I built InkTracker to run my shop. Then I spent seven months trying to break it on real jobs. This post is the bill for that.
I'm writing it because I keep meeting shop owners who are building their own. A spreadsheet that got out of hand, a weekend of vibecoding, a cousin who knows some web stuff. I get it. I was that person. I want to show you what the road looks like past the first weekend, so you can decide with your eyes open.
The numbers
I didn't keep a timesheet, so I measured it after the fact from the commit history and the session logs on my machine. Time counts when two events are under an hour apart, plus a quarter hour for each sitting. That is a conservative way to count it.
- First day of work: April 11, 2026.
- Days worked: 162 of the 205 since.
- Sittings: 461.
- Hours: 1,117 so far, counting anything under two hours apart as one stretch. The strict count, with hour gaps, is 881. The logs from the early months are mostly gone, so the real number is higher, not lower.
- October alone: 328 hours. The longest day was 20.6 hours.
- Weekends: 134 hours.
That is a full-time job stacked on top of running a print shop. And most of it was not building features. It was finding out what a feature does wrong on a real job, and fixing it.
The first weekend is the easy part
A quote screen is a weekend. Anyone messing around with some AI bot can get a page that takes a quantity and a color count and shows a price. That is the part people show off. Here is what came after mine.
- The quote that was right, until the broker. A contract customer with their own pricing, a margin on top, and a customer-supplied garment with no cost. Every one of those is a different quote, and your spreadsheet has one formula.
- The tax-exempt customer. Then the one in a county with a different rate. Then the one who paid, and the books said something else.
- QuickBooks disagreeing with you. We found that QuickBooks Online ignores the tax code you send on some invoices and honors an explicit total instead. That took weeks on a sandbox company to prove. Nobody tells you that on a weekend.
- The supplier order that was real. I placed a "test" order with a supplier through the integration. It was not a test. The blanks showed up.
- The blank price that moved. The quote went out on Monday's price. The supplier raised it on Wednesday. The customer approved on Friday. Your spreadsheet still says Monday. Now the quote pulls the live supplier price when you build it, and I watch the catalog for changes.
- The refund on a surcharged card. Charging a card fee is legal in most states and banned in a few, capped in one, and when you refund the sale you have to refund the fee with it. Our refund flow tells you the amount now. It didn't at first.
- The tap that didn't register. On the shop floor, on an iPad, with ink on your thumb. The button was fine in a browser. It was dead on the device. Three days.
- The quote that stayed in Draft. A customer clicked Send while we were deploying an update. The send never landed. Now sends survive a deploy. Found it the hard way.
- The two days no one could make a quote. I changed one database default. It took every new quote down for two days before I caught it. That one still bothers me.
None of those were features. All of them were hours. And every one was found on a real job, with a real customer waiting, which is the only place these things show up.
What you are actually signing up for
If you build your own, the first version will work. It will work for the jobs you had in mind when you built it. The hours start when a job shows up that you didn't. And the thing about a print shop is that a job you didn't have in mind shows up every week.
You will also be the only person who can fix it, at the moment it breaks, which is usually the moment you are busiest. I had the luxury of making InkTracker my second full-time job. Most owners I talk to don't want a second full-time job. They want the first one to run better.
The honest version
I'm not saying you can't build it. I'm saying I already did, on a real shop, for 1,117 hours and counting, and I'm still finding things. Your hours are worth more on the press, on the phone, or at home.
What that buys you
Every fix above is in the product now. Live blank costs from the suppliers on the quote, a purchase order that knows which sizes are ordered, tax that matches the books, a card fee that follows the rules in your state, a shop floor view that works with ink on your thumb. Plus the things I haven't hit yet, which I'll hit before you do, because I run my shop on it every day.
The Free plan has no monthly bill. If you are deciding between building your own and trying this, the honest test is a week of real jobs. That is about what the first weekend of building your own would have cost you anyway.