When Is Cannabis Ready to Harvest?
The short answer: when the resin glands on your buds have gone from clear to milky white, with a small fraction just starting to turn amber. Everything else on this page is about how to see that accurately, because almost every early harvest comes from looking at the wrong part of the plant.
Harvest timing is the highest-leverage decision left in your grow. You cannot fix it afterwards. Cut a week early and you lose yield and potency that were still being laid down. Cut two weeks late and the cannabinoids you spent four months building have already begun breaking down. The window is real, it is roughly seven to ten days wide for most strains, and it is findable with a tool that costs about ten dollars.
First, the words you need
If you are growing for the first time, a handful of terms do most of the work on this page, and most guides use them without ever defining them.
Trichomes are the tiny mushroom-shaped resin glands that coat the flowers. They look like a frost or a dusting of sugar to the naked eye. Each one has a stalk and a bulbous head, and the head is where cannabinoids and terpenes are actually produced and stored. When people say a plant is "frosty," they mean it has a lot of trichomes.
Pistils are the thin hairs sticking out of the buds. They start white and wiry and darken to orange or brown as the flower matures. They are the plant's pollen-catching structures, and their job is finished long before the resin is.
Calyxes are the small teardrop-shaped pods that the pistils emerge from. Stack enough calyxes together and you have a bud. This matters because the calyxes are the part you are actually harvesting, so the trichomes on the calyxes are the ones that decide your timing.
Sugar leaves are the small, slim leaves poking out from between the calyxes, frosted with resin. Distinguish them from the big seven-fingered fan leaves lower down. Sugar leaves matter more than they look like they should, because their trichomes are the easiest ones to see and the most misleading. That is the subject of half this page.
The Three Harvest Cues, in Order of Reliability
There are three signals people use to time a harvest. They are not equally good, and treating them as equally good is why so many growers cut early. Use them in this order.
If you would rather watch a full run first, here is a trained plant taken all the way to the cut, then come back for the timing detail the video does not have room to cover.
▶ Watch the full harvest tutorial on YouTube
| Cue | What it tells you | Reliability | How to use it |
|---|---|---|---|
| Strain flowering window | Roughly which week to start paying attention | Planning only | Start checking trichomes 2 weeks before the breeder's stated finish |
| Pistil color | That flowering is broadly finished | Rough - can be wrong by 2 weeks | Use as a prompt to get the loupe out, never as the decision |
| Trichome color | The actual chemical state of the resin | Decisive | This is the cue you cut on |
Cue 1: The strain's flowering window (a planning tool, not a timer)
Every cultivar has a rough flowering period. Indica-dominant photoperiod strains commonly finish in 8 to 9 weeks of flower, sativa-dominant strains often take 10 to 12, and autoflowers run 9 to 11 weeks from seed regardless of light schedule. Your seed bank's listing is usually the most accurate source, because they are growing what they sell.
Treat that number as the date you start checking, not the date you cut. Breeder-stated flowering times are measured under the breeder's conditions, and yours are different. Light intensity, temperature, pot size and how hard the plant was trained all shift the finish by days in either direction. Our week-by-week flowering guide covers what should be happening at each stage if you want to sanity-check where your plant is.
Cue 2: Pistils (useful, but routinely over-trusted)
As a flower matures, its white pistils darken and curl inward. The common guidance is to harvest when 90% to 95% have turned brown, and as a prompt that is reasonable.
As a decision, it is unreliable, and it is worth knowing why. Pistils brown in response to pollination and to age, but they also brown from handling, from low humidity, and from being brushed against. A plant that has been moved, defoliated or bumped will show darker pistils than its resin justifies. We have had plants at 95% brown pistils whose trichomes were still visibly clear, a full ten days out. If you cut those, you lose a meaningful amount of both yield and potency for no reason.
So: when the pistils darken, that is your cue to get the loupe out. It is not your cue to get the scissors out.
Cue 3: Trichomes (the one you actually cut on)
Trichome heads change color as the cannabinoids inside them mature, and that color change tracks the chemistry closely enough to time a harvest by. This is the only cue that reports on the thing you actually care about, which is what is in the resin.
It is also the only cue that requires you to do it properly to get a useful answer. The next section is the part most guides skip.
How to Read Trichomes Properly
Reading trichomes is not hard, but it is easy to do in a way that gives you a confidently wrong answer. Three things decide whether your reading means anything: what you look through, where you look, and what light you look under.
What you need
A jeweller's loupe at 30x to 60x magnification is enough, costs under fifteen dollars, and is what we use day to day. A clip-on phone macro lens or a cheap USB microscope also works and has the advantage that you can photograph the same spot across several days and compare properly, which is far more reliable than comparing two memories.
Below about 30x you cannot separate a milky head from a clear one with confidence. Above 60x your depth of field gets so shallow that holding steady on a living plant becomes the limiting factor. Somewhere around 60x is the practical sweet spot.
Where to look - and the mistake almost everyone makes
This is the single most valuable paragraph on this page, so here it is plainly: sugar-leaf trichomes are not bud trichomes, and they ripen roughly a week ahead.
Sugar leaves are the small leaves poking out of the buds. They are frosted, they are flat, they hold still, and they sit right on the surface where a loupe can reach them easily. Calyx trichomes are on the round, tightly packed body of the bud itself, which is awkward to light and awkward to focus on. So when someone new to growing checks trichomes, they almost always end up reading sugar leaves without realizing there is a difference.
Sugar leaves consistently run ahead. If you time your harvest off sugar-leaf trichomes showing 10% amber, the calyxes underneath are typically still nearly all milky with barely any amber, and you are cutting five to ten days early. In our own grows this is the most common single reason a harvest comes in light and underwhelming, and it is completely invisible unless someone tells you to look at a different part of the plant.
So: read the calyxes. Push the sugar leaves aside if you have to. If the only trichomes you can get a clean view of are on a leaf, you have not taken a reading yet.
A sampling protocol you can repeat
One look at one bud is an anecdote. Ripeness is genuinely uneven across a plant, so a single sample tells you about that spot and nothing else. What you want is a small routine you run the same way each time, so that day-to-day changes mean something.
- Sample three to five sites per plant. Two from top colas, one from the middle of the canopy, one from a lower bud. Take them from the same physical spots each time - pick a landmark branch and stick with it.
- Read calyxes, not sugar leaves. Every time. If your view is of a flat leaf surface, reposition.
- Use consistent light. Trichome color reads very differently under a warm-white LED than under daylight, and clear can pass for milky under a strong grow light. Take your readings under the same light source each session, ideally with the grow light off and a neutral white phone torch instead.
- Estimate in thirds, not percentages. Nobody can eyeball 87%. Ask instead: mostly clear, mostly milky, or milky with visible amber. That resolution is enough to make the decision and it is far more repeatable between sessions.
- Check every day or two once you are close, starting about two weeks before the breeder's stated finish. The window moves faster at the end than people expect.
Decide on your best flowers, not your smallest. Lower popcorn buds on an untrained plant frequently never catch up, and waiting for them costs you ripeness on the colas that actually make up your yield.
Clear, milky, amber: what each one means
Two pieces of vocabulary first. CBGA is the precursor compound the plant builds first and then converts into THCA or CBDA. CBN is what THC turns into as it oxidises, and it is markedly less psychoactive and more sedating. Trichome color is a rough visual readout of where the resin sits along that path.
| Trichome head | What is happening chemically | Effect if harvested here | Action |
|---|---|---|---|
| Clear / transparent | Still largely precursor compounds, conversion incomplete | Weak, thin, low yield | Wait. Aim for under 10% to 15% clear at harvest |
| Milky / cloudy white | Peak THC and CBD content | Strongest, most energetic | This is your target. 80% to 90% milky |
| Amber / gold | THC oxidising into CBN | Heavier, sedating, less potent overall | 5% to 10% is ideal. Past 20% you are losing THC |
| Gray / shriveled | Degraded and physically damaged | Harsh, badly diminished | Overdue. Cut now |
What Amber Trichomes Actually Mean
There is a widely repeated idea that amber trichomes are a dial you turn for a heavier high - that more amber gives you a stronger, more relaxing product. That framing is backwards in a way worth correcting, because it leads people to deliberately wait too long.
Amber is not a separate cannabinoid being produced. Amber is THC being destroyed. As THC oxidises it converts to CBN, which is substantially less psychoactive and more sedating.1 So an amber-heavy harvest is not a stronger harvest that happens to feel different. It is a weaker harvest whose remaining effect profile has shifted toward sedation.
The degradation itself is well characterized in the storage literature. THC content in stored cannabis declines in the range of roughly 16% to 25% per year at room temperature, with the loss accelerating under light and heat exposure.23 On the plant, in a lit grow room in the final weeks, that same oxidation is already running - which is precisely what you are watching when trichomes turn gold.
None of that makes a sedative harvest wrong. If you want something that helps you sleep, letting amber climb toward 20% is a legitimate choice and we would not argue with it. Just make it knowing what the trade is: you are paying THC for sedation, not gaining a new kind of strength.
Harvest windows by what you actually want
All three of these are legitimate harvests off the same plant. The difference is where in the window you cut, and what you are willing to trade for it.
| You want | Cut at | Trade-off |
|---|---|---|
| Clear, energetic, daytime | 85% to 90% milky, 0% to 5% amber | Narrow window, easy to miss by a day or two |
| Balanced (our default) | 80% to 90% milky, 5% to 10% amber | The widest, most forgiving window |
| Heavy, sedative, evening | 70% to 80% milky, 15% to 20% amber | Measurably less total THC |
When to Harvest Autoflowers
Autoflowering plants begin flowering on age rather than on a light-schedule change, usually somewhere around week 3 to week 5 from germination. That means you cannot count "weeks of flower" the way you would with a photoperiod plant, because there was never a flip to count from. Count from seed instead.
Most autoflowers finish between 9 and 11 weeks from seed, and some larger cultivars run to 13. The three cues above all apply unchanged - flowering window, pistils, trichomes, in that order of trust. What changes is your margin for error. The whole life cycle is compressed, so a week of drift is a much larger fraction of the grow. For where harvest sits at the end of that arc, the full seed-to-harvest timeline maps every stage and how long it takes. Start checking trichomes around week 7 from seed rather than waiting for pistils to brown.
Autoflowers also tend to ripen less evenly than photoperiods, largely because they get less time to build a level canopy. It is common for the top cola to be ready while lower buds still show white pistils. When that happens, harvest the top and leave the lower growth another three to five days with the light still on. The exposed lower buds will finish quickly once the canopy above them is gone.
Because autos cannot afford recovery time, they should be trained differently as well - LST first, and at most a single early topping. Our topping guide covers where that line sits.
Signs Harvest Is One to Two Weeks Away
Beyond trichomes and pistils, a finishing plant tells you it is finishing in several ways at once. None of these are decisive on their own, but together they are a good confirmation that your trichome reading is pointing the right way.
Calyxes swell. The teardrop pods that make up the bud visibly fatten in the final week or two as the plant puts its last energy into them. Buds get noticeably denser and rounder rather than longer.
Fan leaves yellow from the bottom up. The plant pulls mobile nutrients, nitrogen especially, out of older leaves and moves them into the flowers. Gradual yellowing in the last two weeks is normal and healthy. Yellowing that starts much earlier than that is a different problem - our deficiency diagnosis guide will tell you which.
Water uptake drops. Pots stay heavy noticeably longer than they did mid-flower. The plant is winding down its metabolism.
Aroma peaks and then changes. Terpenes are produced in the same glandular trichomes as the cannabinoids and accumulate through late flower,6 so the smell is at its strongest in the final stretch. Many growers notice it going from sharp and green to rounder and sweeter as the plant finishes.
Foxtailing. Some plants push new white pistils from the tops of already-mature buds, producing a spiky, tower-like growth. In some cultivars this is simply genetic. In others it is a stress response to too much light or too much heat at the canopy. If you see foxtails and your calyx trichomes are already 80% milky, harvest rather than wait - the plant is not going to improve from here.
Why Trained Plants Ripen More Evenly
This is the part of harvest timing that gets decided months before harvest, and it is worth understanding because it is the difference between one clean cut and a staggered, fiddly harvest.
On an untrained plant, the apical tip dominates. You get one tall central cola sitting close to the light and a skirt of smaller buds below it receiving a fraction of the intensity. Light falls off with the square of distance, so a bud sitting 40 cm (16 in) below the canopy is not receiving slightly less light than the top - it is receiving dramatically less. Those lower buds develop more slowly and ripen later, sometimes by two weeks or more.
A level canopy removes most of that spread. When every bud site sits at roughly the same distance from the light, they receive comparable intensity, develop at comparable rates, and arrive at comparable ripeness. That means one harvest instead of two, and it means the "decide on your best flowers" rule costs you far less, because your worst flowers are much closer to your best.
This is the practical payoff of topping plus continuous low-stress training through veg, using something like LST clips to hold branches horizontal while they set. The mechanism is covered properly in our canopy development science article. By flowering the training itself is finished, and support is all that is left - heavy colas in late flower want tying up rather than bending, which is what garden ties like the BudHuggers are for.
The Last 48 Hours Before You Cut
Three practices are near-universally recommended in the final days. They are not equally well supported, and it is worth being clear about which is which.
Pre-harvest defoliation: worth doing
Removing the large fan leaves a day or two before cutting makes the plant much faster to trim, improves airflow through the buds while they dry, and lowers mold risk. This is the one genuinely practical step of the three, and the full method is in our defoliation guide.
Flushing: no demonstrated benefit, and a possible yield cost
Flushing means feeding plain pH-balanced water for the last one to two weeks, on the theory that it pulls residual mineral salts out of the plant and gives a cleaner-burning, smoother product. It is close to universal advice. The controlled evidence does not support it.
A 2024 trial across five cannabis cultivars compared plants fed to harvest against plants given only water in the final phase. Cannabinoid concentrations differed in just 30% of samples and terpenoid concentrations in only 3%, with no meaningful effect on mineral accumulation, biomass or the terpene profile. Flushed plants did yellow more and lose chlorophyll, which looks like the process working, but bud quality did not improve.4
A 2025 Clemson study went further and found a cost. Testing flush durations from zero to three or four weeks, even a one-week flush changed inflorescence mass and tissue cannabinoid concentration in both cultivars, and the authors caution that whatever growers are hoping to gain from flushing may come at the expense of dry mass and total cannabinoid yield.5
Both caveats are worth stating: the 2024 work used medical cultivars and the 2025 work used high-CBD cultivars, so neither is a perfect stand-in for a high-THC home grow. But the direction is consistent, and there is no controlled trial we are aware of showing that flushing improves smoothness, flavor or ash color.
Our position: if you have been feeding heavily, a short 5 to 7 day flush is defensible and unlikely to hurt much. Long flushes of two weeks or more are the ones to avoid, since that is where the yield cost shows up. What flushing should never be is the ritual you perform instead of getting your timing right - the yellowing you see is the plant running out of nitrogen, not impurities leaving the buds. If you are unsure when to stop feeding before harvest, our feeding guide covers the taper.
48 hours of darkness: no good evidence
The claim is that a day or three of total darkness before the cut triggers a final burst of resin. We have not found peer-reviewed support for it, and the proposed mechanism does not have much behind it. It very probably does not harm the plant. We do not do it, and we would not spend two days of flowering light on it. If you want to try it, 48 hours is the usual figure.
How to Cut the Plant
Once the trichomes say go, the physical harvest is quick. There are two approaches, and your ambient humidity mostly decides which one you want.
Method 1: Whole plant
Cut the main stem near its base with pruning shears or a serrated knife and hang the entire plant. The stem holds a substantial amount of water, which slows the whole dry down and gives you a gentler, more even moisture loss - and slower drying preserves terpenes.
The trade-offs are space and risk. A whole plant needs real hanging room, and because the interior of a dense plant dries slowest, whole-plant hanging is only safe when your ambient humidity is comfortably below 60%. Above that, the middle of the plant can develop mold before it dries.
Method 2: Branch by branch
Cut individual branches and hang or rack them separately. It takes longer and gives up some of the slow-dry benefit, but it fits in far less space and it is much safer in humid conditions because air can reach every surface.
Choose branch-by-branch when your humidity is above 60%, when the plant is too large or wide to hang whole, or when a trellis net makes whole-plant removal impractical - the situation covered in our SCROG guide.
Tools worth having ready
- A 30x to 60x jeweller's loupe, or a phone macro lens
- Sharp, clean pruning shears for stems and trimming scissors for finer work
- Disposable gloves - resin gets on everything, and clean hands make the trimming stage far less unpleasant
- Hangers, a line or a drying rack, ready before you cut rather than after
- Garden ties for hanging branches and for supporting colas in the final weeks
Straight After the Cut
Drying begins the moment the plant comes down, and it is where a well-timed harvest gets won or lost. Get the plant into its drying space immediately rather than leaving it in a warm, bright room while you tidy up.
The short version: dark room, 60°F to 68°F (15°C to 20°C), 55% to 65% relative humidity, gentle indirect airflow, 7 to 14 days. Never aim a fan at the buds. Buds are dry when a thin stem snaps rather than bends.
If you have seen 45% to 55% RH recommended elsewhere, that guidance is dated and it dries too aggressively, stripping terpenes for no gain. Our complete drying guide covers the 60/60 rule, the common drying mistakes and the full process properly - that article is the one to follow from here. After drying comes trimming and then curing, which is where the quality you protected at harvest finally shows up.
Common Harvest-Timing Mistakes
Reading sugar leaves instead of calyxes. The big one, covered above, and worth repeating because it costs more yield than anything else on this list.
Cutting on pistils alone. Pistils brown for reasons that have nothing to do with resin maturity. They are a prompt, not a verdict.
Judging trichomes under the grow light. Warm-spectrum LEDs make clear heads look milky and milky heads look amber-ish. Kill the light, use a neutral torch, and take every reading the same way.
Waiting for every bud to be ready. Lower popcorn buds on an untrained plant may simply never catch up. Decide on your best flowers.
Chasing amber for strength. Amber is THC that has already broken down. If you want sedation, that is a real choice - just do not mistake it for potency.
Harvesting with nowhere to dry. A plant hanging in an uncontrolled room at 75% humidity can develop mold in under 48 hours. Botrytis cinerea, the fungus behind bud rot, is one of the most damaging pathogens in cannabis production and thrives in exactly those conditions.7 Set the drying space up before you cut, not after.
Frequently Asked Questions
How do I know when my cannabis is ready to harvest?
Check trichomes on the calyxes with a 30x to 60x loupe. Harvest when 80% to 90% of the heads are milky white and 5% to 10% have turned amber. Use the strain's flowering window to know when to start checking, and brown pistils as a prompt to look, but make the decision on trichome color alone.
What is the most common harvest timing mistake?
Reading trichomes on sugar leaves rather than on the calyxes. Sugar-leaf trichomes are easier to see and ripen roughly a week earlier, so timing a harvest by them typically means cutting five to ten days early. Always push the leaves aside and read the body of the bud itself.
Do more amber trichomes mean stronger weed?
No, the opposite. Amber means THC has oxidised into CBN, which is less psychoactive and more sedating. An amber-heavy harvest has less total THC. Waiting for amber is a valid choice if you want a sedative effect, but it trades potency for that effect rather than adding strength.
When should I harvest autoflowers?
Most autoflowers are ready 9 to 11 weeks from seed, with some larger cultivars running to 13. Because autos flower on age rather than a light flip, count from germination rather than from a flip date. Start checking trichomes around week 7 from seed, since the compressed life cycle leaves less margin for error.
Can I harvest part of the plant and leave the rest?
Yes, and it is often the right call when ripening is uneven. Cut the top colas once they are ready and leave the lower buds under the light another three to five days. With the canopy above them removed they receive far more light and will finish quickly.
What magnification do I need to see trichomes?
30x to 60x. Below 30x you cannot reliably tell a clear head from a milky one. Above 60x the depth of field becomes so shallow that holding steady on a living plant is the limiting factor. A jeweller's loupe or a clip-on phone macro lens both work well.
What happens if I harvest cannabis too early?
Trichomes are still largely clear, meaning conversion to THC and CBD is incomplete. You get lower potency, lighter and less dense buds, and less total yield, since the plant lays down a meaningful amount of both weight and resin in the final two weeks.
What happens if I harvest cannabis too late?
THC continues oxidising into CBN, so potency falls and the effect becomes heavy and sedating. Past roughly 20% amber you are giving up THC. Very late harvests also carry more bud rot risk, since dense mature buds hold moisture.
Should I flush my plants before harvest?
The controlled evidence does not support it. A 2024 trial across five cultivars found flushing changed cannabinoid levels in only 30% of samples and terpenoids in 3%, with no gain in bud quality. A 2025 Clemson study found even a one-week flush reduced inflorescence mass, meaning flushing may cost yield. A short 5 to 7 day flush after heavy feeding is defensible; two-week flushes are the ones to avoid.
Does 48 hours of darkness before harvest increase potency?
There is no peer-reviewed evidence that it does. The practice is widespread but the proposed mechanism is not well supported. It probably does not harm the plant. We do not do it, and we would rather not give up two days of flowering light for it.
Why are my lower buds not ready when the tops are?
Light intensity falls off sharply with distance, so buds well below the canopy develop more slowly and ripen later. A level canopy built with topping and continuous low-stress training largely removes the gap, which is why trained plants tend to finish in a single cut.
Do I need to check trichomes on every plant?
Yes, if the plants are different cultivars, since flowering times vary widely. Even clones of the same plant in the same tent can differ by several days depending on their position under the light. Sample three to five sites per plant rather than assuming the room is uniform.
My pistils are all brown but the trichomes are still clear. What do I do?
Wait, and trust the trichomes. Pistils brown from age, handling, low humidity and physical contact as well as from maturity, so they routinely run ahead of the resin. It is common to see 95% brown pistils on a plant that still needs another week or more.
How long does the harvest window stay open?
Roughly seven to ten days for most strains between hitting the balanced target and drifting past it. That is why the checking interval matters more than any single reading. Once you are within two weeks of the breeder's stated finish, look every day or two.
References
- Lindholst, C. (2010). Long term stability of cannabis resin and cannabis extracts. Australian Journal of Forensic Sciences, 42(3), 181-190. https://www.tandfonline.com/doi/abs/10.1080/00450610903258144
- Trofin, I. G., Dabija, G., Vaireanu, D. I., & Filipescu, L. (2012). Long-term storage and cannabis oil stability. Revista de Chimie, 63(3), 293-297. https://revistadechimie.ro/Articles.asp?ID=2851
- Zamengo, L., Bettin, C., Badocco, D., Di Marco, V., Miolo, G., & Frison, G. (2019). The role of time and storage conditions on the composition of hashish and marijuana samples. Forensic Science International, 298, 131-137. https://www.sciencedirect.com/science/article/abs/pii/S0379073819300994
- Saloner, A., Sade, Y., & Bernstein, N. (2024). To flush or not to flush: Does flushing the growing media affect cannabinoid and terpenoid production in cannabis? Industrial Crops and Products, 220, 119479. https://www.sciencedirect.com/science/article/abs/pii/S0926669024011348
- Alden, M. J., & Faust, J. E. (2025). Exploring the legacy practice of flushing in controlled-environment production of high-CBD cannabis (Cannabis sativa). HortScience, 60(10), 1818-1825. https://journals.ashs.org/view/journals/hortsci/60/10/article-p1818.xml
- Booth, J. K., & Bohlmann, J. (2019). Terpenes in Cannabis sativa - From plant genome to humans. Plant Science, 284, 67-72. https://www.sciencedirect.com/science/article/pii/S0168945219301487
- Punja, Z. K. (2021). Emerging diseases of Cannabis sativa and sustainable management. Pest Management Science, 77(9), 3857-3870. https://onlinelibrary.wiley.com/doi/10.1002/ps.6307
About the Author
Henrique Dias is the CEO and co-founder of BudTrainer. He is a mechanical engineer who also holds a graduate certificate in Commercial Cannabis Production from Niagara College in Ontario, Canada - the first college-level cannabis cultivation program in North America. Before founding BudTrainer, Henrique worked with Health Canada licensed cannabis producers, where he gained hands-on experience in cultivation, processing, and post-harvest management of cannabis at commercial scale. He started BudTrainer to bring that same level of craft to home growers through better-designed gardening tools and clear, science-backed education.
DISCLAIMER: Everything taught and sold by BudTrainer® is to be used strictly for legal purposes. We do not endorse the production of illegal substances and it is your duty to ensure that you are complying with the law. The words "hemp," "cannabis," "weed," and "marijuana" are used interchangeably to refer to the same plant (legal hemp with less than 0.3% THC) for the purposes of this lesson.
