Every seed listing carries a yield figure. “500–650 g/m² indoor.” “Excellent outdoor yield.” Growers compare those numbers, buy on them, and are then surprised.
The numbers are not usually lies. They are measurements taken under conditions that are not stated, and a measurement without its conditions is not a measurement. This post is about what the conditions are, why they can matter more than the genetics, and what you can legitimately conclude from a catalogue figure. If you came here looking for a list of “high-yield strains”, read on: the rest of this post explains why we don’t give one, and what to look at instead.
Start with the physics, because it settles most of the argument
Some of the most useful controlled data on cannabis yield comes from an indoor trial at the University of Guelph, published in Frontiers in Plant Science in 2022. It grew a high-THC cultivar (‘Meridian’) under LED light at 600, 800 and 1,000 µmol·m⁻²·s⁻¹ PPFD for 45 days under short-day (flowering) conditions.
Yield rose in proportion to light across that range. The authors ran a simple regression on the mean yields at the three light levels. It predicts approximately 51 g/m² of additional dry inflorescence for every extra 100 µmol·m⁻²·s⁻¹ at the trial’s planting density (4.6 g per plant). Inflorescence dry weight rose 1.6-fold from the lowest to the highest light level, the same factor as the increase in light.
Take that seriously for a moment. Moving from 600 to 1,000 µmol·m⁻²·s⁻¹ (same room, same cultivar, same everything else) changes yield by roughly 200 g/m² on that slope. That can be larger than the entire quoted range on many seed listings.
The limits matter just as much. This is one cultivar in one trial, and the linear relationship holds only within the tested range (600–1,000 µmol·m⁻²·s⁻¹). The trial used no supplemental CO₂, with day temperatures around 26 °C. Above roughly 1,000 µmol·m⁻²·s⁻¹, plan on CO₂ and climate control to match. An earlier trial by the same group used a different cultivar (‘Stillwater’), 12 weeks of flowering and ambient CO₂. Its yield still rose linearly, from 116 to 519 g/m² across 120–1,800 µmol·m⁻²·s⁻¹. The direction is the same, but the slope is not. Treat 51 g/m² per 100 µmol as an illustration, not a constant for your room.
Which means: the difference between a “high-yield” cultivar and an average one can be smaller than the difference between your lighting and the breeder’s. If you take a 650 g/m² cultivar and run it at 600 µmol·m⁻²·s⁻¹ under a canopy that is half filled, you will not get 650 g/m², and the seed was not the problem.
Both trials found something else worth carrying forward: cannabinoid concentration did not change significantly with light intensity. Cannabinoid output per square metre rose because biomass rose. In these trials, light moved yield but not concentration: two different levers, and only one of them is on the seed packet.
The variables a catalogue figure hides
Light intensity and photoperiod. As above. Almost never stated.
Plant density. A yield per square metre depends entirely on how many plants occupy that metre. Four large plants vegged for six weeks and one small plant vegged for two produce very different figures from the same genetics, and both are honestly reportable as “g/m².”
Vegetative time. Not part of the flowering yield figure at all, but it is one of the biggest levers most growers actually control, because it decides whether the light is used at all. A plant that fills the canopy before the switch uses all the light in the room. A plant that does not, wastes it.
Canopy area versus footprint. A trained, screened plant occupies more effective canopy than its pot footprint suggests. Whether the breeder measured the square metre of floor or the square metre of canopy changes the number and is never specified.
Training method. SCROG, SOG, topping, mainlining: each changes light interception per plant substantially.
Wet or dry. Fresh flower is roughly three-quarters water. A wet weight is not comparable to a dry weight, and “yield” in casual conversation is frequently wet.
Trim level. Whether sugar leaf is included changes the number materially. Machine-trimmed and hand-trimmed weights from the same harvest differ.
Container size and root volume. Restricted root volume caps growth regardless of everything upstream.
Whether it was an average or a best result. A catalogue figure representing the best plant of a trial and one representing the mean of twenty are both “the yield,” and the difference between them is large.
Why grower-to-grower comparison is nearly meaningless
Somebody posts 700 g/m². Somebody else posts 350 g/m² on the same cultivar. Neither is lying.
Consider what has to match for those numbers to be comparable: PPFD at canopy, hours of light, spectrum, CO₂ concentration, day and night temperature, VPD, root zone volume, substrate, feed programme and EC, plant count, vegetative duration, training method, flowering duration, dry-down method, final moisture content, and trim standard.
If any one of those differs, the comparison is not a comparison of genetics. It is a comparison of two different experiments that share a cultivar name. And given how much commercial seed segregates, the two growers may not even have had the same plant.
The one comparison that is informative is a comparison you run yourself: two cultivars, same room, same cycle, same treatment. That tells you which performs better for you. It tells you nothing about which will perform better for anyone else, which is fine, because you are not growing for anyone else.
What grower experience says about high-yield genetics
Breeders and growers have long made some general observations about high-yield genetics. These are practitioner observations, not trial data, but they are worth stating because they describe types of plant rather than specific products:
High-yield cultivars are typically indica-dominant, with dense structure, dense flowers, and a short flowering period of about 7 to 9 weeks. The short cycle is much of the point: yield per year, rather than per cycle, is what determines a commercial operation’s output.
Sativa-dominant lines can yield excellently but generally need longer, which shifts the annual arithmetic.
Individual females of the same cultivar vary. This is the most practically important of these observations, and it is consistent with the fact that many commercial seed lines still segregate. The practical answer is still correct: if consistency across runs matters to you, start from seed (regular or feminized), grow out several plants, select a mother and clone her. Every plant from that mother has the same genotype. Every plant from a seed packet does not. Regular seed mainly matters if you intend to breed.
Autoflowering lines generally yield less per cycle than photoperiod lines, largely because a fixed, short life cycle caps the total light a plant can intercept. See the topic Autoflowering Cannabis (topic 1326) for more on this.
What we are not going to print
Older grower guides contain long lists of named cultivars presented as the highest-yielding available: a dozen breeders and dozens of product names, many of them a decade or more old.
We are not reproducing those lists, for three reasons. Product lines change and several of the named ones no longer exist in the form described. Years of breeding have moved the field. And the figures attached to them were catalogue figures, which is the exact thing this post is about not trusting.
Similarly, we are not printing a table of “expected yields” by cultivar type. We could not source one from a controlled multi-cultivar trial, and a plausible-looking table without that backing would be the worst thing in the article.
What to actually do with a catalogue yield figure
Treat it as an upper bound obtained under favourable conditions. Not a target, not a promise.
Compare figures only within one breeder’s catalogue. A single breeder is probably measuring consistently across their own range, so their relative rankings carry information even if their absolute numbers do not transfer to your room.
Ask for the conditions. A breeder who will tell you the PPFD, the plant density and the vegetative time behind their figure is a breeder whose number means something. Most will not answer. That is itself informative.
Fix your own limiting factor first. If you are at 500 µmol·m⁻²·s⁻¹ with a canopy that is 60% filled, the highest-yielding genetics in the world will not fix that. The light data above gives you a rough idea of how much you are leaving on the floor.
Keep your own records. Your own yield figure, for your own room, from your own last three runs, is the only number in this whole subject that reliably predicts anything.
The short version
Yield is mostly light, canopy management and vegetative time, and vegetative time decides how much of the light the canopy actually catches. Genetics set a ceiling and a flowering duration; they do not set a number. A catalogue figure is a measurement whose conditions have been removed, and the missing conditions often matter more than the figure.
The best cultivar for yield is the one that finishes fast enough to give you more cycles per year, fills your specific canopy under your specific light, and does not rot in your specific climate. No catalogue can tell you which one that is.
Sources for the light–yield relationship: Llewellyn D, Golem S, Foley E, Dinka S, Jones AMP, Zheng Y (2022). “Indoor grown cannabis yield increased proportionally with light intensity, but ultraviolet radiation did not affect yield or cannabinoid content.” Frontiers in Plant Science 13:974018. doi:10.3389/fpls.2022.974018. Rodriguez-Morrison V, Llewellyn D, Zheng Y (2021). “Cannabis Yield, Potency, and Leaf Photosynthesis Respond Differently to Increasing Light Levels in an Indoor Environment.” Frontiers in Plant Science 12:646020. doi:10.3389/fpls.2021.646020. The section on high-yield genetics is practitioner observation, not trial data.
In Thailand, cultivation must follow current controlled-herb and licensing rules. In countries where cannabis is banned, this article is for information only.