Microgreens Yield Per Tray: A 12-Variety Pre-Registered Experiment

Most microgreens articles tell you the theoretical yield range — "a 1020 tray produces 8–14 ounces" — without specifying which variety, what setup, what density, or where the numbers came from. I've never been satisfied with that. The variation between varieties is huge, the variation between grow conditions is bigger still, and someone planning a microgreens setup deserves real numbers from a real tray.

So I'm running the experiment. Twelve varieties, same trays, same medium, same lights, same temperature, sown the same week.

This piece is what I'd call a pre-registered experiment: I'm publishing the methodology and my expectations before I have all the measurements, then updating each row with the real numbers as each tray comes off over the next 3–4 weeks. The article's lastUpdated date will bump every time fresh data lands.

The expected ranges below come from two sources: published seed density and harvest data from suppliers (True Leaf Market and Bootstrap Farmer both publish reasonably honest ranges), and my own prior grows across the same varieties since 2019. They're not guesses — but they're also not the same thing as a single clean measurement on a calibrated scale. The point of the experiment is to produce those measurements.

What I'm reporting

For each variety I'm recording:

• Seed used (grams) — pre-soak weight where applicable, dry weight otherwise
• Days to harvest — from sow to scissor
• Wet harvest weight (grams) — cut above the medium, weighed immediately
• Yield ratio — wet harvest divided by seed used; the more useful "is this variety efficient?" number

What I'm not reporting (yet): dry weights, nutritional density per gram, or cost-per-gram (seed cost varies too much by source). Those are follow-up pieces.

The numbers

Expected ranges below. Values shown as ranges because seed density choices and small temperature variations meaningfully move the numbers. The middle of each range is what I expect from a careful single trial.

Variety	Days to harvest	Seed used (g)	Wet harvest (g)	Yield ratio
BroccoliDense canopy. Mild flavour. The reliable workhorse.	10–12	12–25	160–220	—
Radish (Daikon)Among the fastest. Spicy bite.	8–10	25–40	200–280	—
Radish (China Rose)Pink-stemmed. Visually striking, same speed as Daikon.	8–10	25–40	200–280	—
Pea shoots (Dun)Pre-soaked. Big seed, big yield. My personal favourite.	10–14	100–150	280–380	—
Sunflower (regular)Pre-soaked. Crunchy stems. Hull-removal is the work step.	8–12	100–150	280–420	—
Sunflower (black oil)Pre-soaked. Slightly milder than regular sunflower.	8–12	100–150	280–400	—
KaleSlow start, dense finish. Tastes like kale, surprisingly.	10–14	12–25	150–220	—
Mustard (yellow)Hot. A small handful goes far.	8–12	18–30	150–220	—
WheatgrassPre-soaked. Grown for juicing, not eating. Yield is huge but seed-cost is real.	8–12	200–300	400–700	—
Cress (garden)Mucilaginous seed — see the basil/cress notes elsewhere on the site.	7–12	15–25	150–200	—
FenugreekBitter, savoury. The research-backed pick. Separate article on this one.	9–12	20–30	140–200	—
AmaranthSlowest. Smallest. Heat-loving — slow at sub-20°C ambient.	14–21	8–15	80–140	—

A few things to flag in this table before the measurement step:

• Wheatgrass is the outlier on yield. A 400–700g harvest sounds huge until you notice it took 200–300g of seed to get there — the yield ratio is closer to 2× rather than the 3–5× of the more efficient varieties.
• The big-seed varieties (peas, sunflower, wheatgrass) all need pre-soaking. The seed weights above are pre-soaked weights, which is the convention published by most suppliers.
• Amaranth is the trickiest. Tiny seed, slow growth, picky about temperature. I'm including it because it's a varietal nobody else publishes real data on, but it's also the one most likely to surprise me.

Three patterns I expect to see in the data

Before the measurements come in, here's what I'm betting:

1. Big-seed varieties win on absolute yield, lose on efficiency. Pea shoots and sunflower will sit at the top of the absolute-grams column — the seeds are large and the plants pack on water during growth. But the yield ratio (wet harvest divided by seed used) will probably be lower than the small-seed varieties. Both can be the right pick depending on whether you optimize for tray output or seed economy.

2. The 8–12 day band is the commercial sweet spot. Broccoli, kale, radish, mustard, cress — all reliably between 8 and 12 days, predictable density, no exotic care needed. If you ever sell microgreens, you live in this band most weeks.

3. Wheatgrass is misclassified. Most yield tables include it alongside the others, but it's a different category — grown for juicing, harvested at a different growth stage, and the seed weight dominates the unit economics. Including it in the comparison is useful for completeness, but readers should read its row as "not directly comparable."

The single number I'm most uncertain about: amaranth. Slow, tiny-seed, and finicky at sub-20°C. I might end up with a thin tray after three weeks. If I do, that's the actual finding — and worth reporting honestly.

Why this matters if you're thinking about selling

For anyone considering microgreen farming as a business, the yield ratio is the most underrated number. Restaurants and farmers' markets price microgreens at $20–40 per pound retail, which is $44–88 per kilogram. If you can get 300g of wet pea shoots from a tray that used $0.40 of seed, the unit economics are good. If you only get 150g, suddenly the seed cost is a noticeable line item.

I'll come back to that math in detail when the numbers are in, but the framing matters: yield per tray is not just curiosity — it's the foundation of every other calculation in commercial microgreens.

Method notes (for anyone trying to reproduce)

• All trays are standard 1020s (about 25 × 50 cm / 10 × 20 inches), with drainage holes.
• Medium: 50/50 mix of peat-based seed-starting mix and coir, ~3 cm deep. Wet to "just no drip" before sowing.
• Density: each variety sown at the density typically recommended by reputable seed suppliers (e.g. True Leaf Market, Sprout People). I'm not optimizing density here — that's a separate experiment.
• Light: full-spectrum LED at approximately 150 PPFD, 16 hours on. Distance from soil ~25 cm.
• Watering: bottom-watering only, every other day for the first week, daily once cotyledons fully open.
• Temperature: ambient room 18–20°C day, 16–18°C night.
• Harvest moment: when most plants in the tray show fully-open cotyledons. For some varieties (broccoli, kale) I wait until the first true-leaf hint is visible, which is the conventional definition of "microgreen."
• Weighing: cut above the medium, no shaking off water, on a kitchen scale immediately after cutting. Wet weight only.

Reproducing this exactly is the only way to compare cleanly. If your tray size, light intensity, or ambient temperature is meaningfully different, your numbers will be too.

Honest limitations

• Single trial per variety. A proper study would run each variety at least three times and report averages with standard deviation. I'm planning to extend this to triplicates over the next few months. Treat the current numbers as a single data point.
• Density not yet optimized. I used "recommended" densities, but optimal density varies by season, lighting, and humidity. This is a follow-up experiment I'm planning.
• No flavour or visual quality scoring. A 350g sunflower tray isn't automatically "better" than a 280g one if half of it is leggy and pale.

I'll add the triplicates and the flavour/visual scoring as separate updates rather than rewriting this piece — the goal is to keep this article honest about what it is (a single careful weigh-in) and not pretend it's more.

Related on the site

• Best Growing Mediums for Microgreens — why the 50/50 mix in the method notes
• Best Lighting for Microgreens — the LED setup behind these yields
• pH Water for Microgreens — the water side of the equation
• Is Microgreen Farming Profitable? — what these yield numbers turn into commercially

This piece is a living document. Numbers will be filled in as the trays are weighed, and I'll bump the lastUpdated date when they are.