I once pulled a root crop seedling out of its cell to transplant it and watched the whole taproot fold over on itself in my hand, bent at an angle it was never going to grow straight from again. I planted it anyway, because I had already invested weeks in it, and it sulked in the ground for the rest of the season, small and stubborn, never catching up to the ones I sowed directly a few weeks later. That was the moment I started questioning a habit I had been treating as simply correct.
I thought starting everything indoors was the serious approach
For my first two full seasons of gardening, I started almost everything on my windowsill, on the theory that more of a head start was always better and that indoor starting was what people who took their garden seriously did. I filled trays with crops that, looking back, had no real need to begin their life in a small plastic cell under uneven window light. I was not wrong that indoor starting has real advantages for some plants. I was wrong that it was the better choice by default, across the board, regardless of what I was growing. That took me two seasons of mixed results to actually admit. Part of the problem was that a full tray on the windowsill felt like progress in a way an empty patch of soil did not, even when the empty patch was actually the better plan for that particular crop. It is an easy trap to fall into, mistaking visible effort for the right effort.
What went wrong with root crops
Root vegetables were where the pattern was clearest and the most disappointing. Anything that grows a single taproot straight down does not appreciate being pulled out of a small cell and settled into open ground later, because that early disturbance to the root, even a small one, tends to show up later as a stunted or oddly shaped result. I lost most of a tray this way before I connected the pattern to the transplanting itself rather than to something about the weather or the soil that season. Once I started sowing those seeds directly into the ground instead, at whatever point felt right once the soil had warmed and the last frost had genuinely passed, the difference in how evenly and confidently they grew was obvious within a few weeks.
Fast growing greens didn’t need the head start either
Certain quick growing greens were another crop I was starting indoors for no real reason. They come up fast enough from a direct sowing that the few days of head start an indoor tray gave them barely mattered by the time everything caught up outdoors, and in the meantime I was using up limited windowsill space on cells that would have done just as well seeded straight into a container or bed. Once I noticed how quickly the direct sown rows caught up to the transplanted ones, continuing to start them indoors stopped making sense as anything other than habit.
What I still start indoors, and why
I did not throw the whole approach out. Certain tender crops genuinely benefit from extra weeks of a head start before the season is warm enough to sow them outside where I live, especially ones that need a longer stretch of warm weather than my growing season comfortably offers. Those still go on the windowsill in early spring, and I still harden them off carefully before they move outside for good. The difference now is that I ask the question before I sow anything, rather than defaulting to a tray out of habit. Does this particular crop actually need the head start, and does it tolerate being moved once it has one, or am I creating a problem I will have to solve later just because starting indoors felt like the more serious choice.
How I decide now
My rough test these days is whether a crop has a root system that resents disturbance, and whether it grows fast enough outdoors that an indoor head start would not meaningfully change when it is ready. If a crop fails either of those checks in the direction of direct sowing, it goes straight into the ground once conditions are right rather than into a tray on my windowsill. I check my seed packet for how that particular variety and grower recommend it be started too, since a packet knows its own crop better than a general habit I picked up two seasons in. Between the two, my decisions about what earns a spot on the windowsill have gotten a lot narrower, and a lot more deliberate.
What changed in my results
The garden did not get bigger or more impressive once I made this change. What changed was how many plants sulked, stalled, or simply underperformed after transplanting compared to how many grew steadily from the day they came up. Direct sowing takes a bit more patience, since you are waiting on the ground and the season rather than getting a jump on things indoors, but for the crops that actually prefer it, that patience pays back within a few weeks in the form of plants that never had to recover from being moved in the first place. I wish I had trusted that two seasons earlier instead of assuming more effort indoors was automatically the better plan. I keep a note of which crops went which way each year in my notebook now, so the decision does not rely on memory alone the next time March comes around, and so I have an honest record to check when I am tempted to fall back into old habits.