GardenTrack
Join the beta

Where every date and spacing in the app comes from

Live · This works today.

Every planting date, spacing and stage the app shows you comes from one database of 152 crops and 3,174 varieties. Most of it is sourced to a named university extension publication, and you can download every crop and every variety free and check the numbers yourself.

Everything the app tells you about a plant — when to start it, how far apart to space it, what it should be doing in July — comes out of one database.

152 crops. 3,174 named varieties under them. 852 rows describing what a plant is doing at each stage of its year.

Search knows the variety names too

Type wax beans into the search box on the plant database and you get Bush Bean, with a line underneath saying which variety matched. You used to get nothing, because the search only read crop names and no crop is called a wax bean — it is a variety sitting inside bush beans. You had to already know that to find it.

So the search now reads every variety name we hold as well. Type zucchini and you land on summer squash. Type cherokee purple or black krim and you land on tomato. The crop card tells you which of its varieties you just matched, so you know why it came up.

It does not mind plurals, and it does not mind the order. Wax beans and wax bean find the same thing, and so does purple cherokee. If you type the name of a crop we hold, that crop comes first — a variety match never pushes it down the page.

Most of it is sourced. The rest says so.

4,486 of 5,412 values cite a named publication. That's 82.9%, and the number is worth less than the thing underneath it: the 926 that cite nothing each record why.

743 of them because nobody publishes that fact. Row spacing for an uncommon herb, say — we filled it from ordinary horticulture rather than leave you with a blank. 183 because a person set it and attached no citation.

So a gap is never ambiguous. You are never left wondering whether we did not check or nobody knows.

Sometimes a number needs a condition, and the condition is the important half

A few values carry an extra line under them. Carrot is the clearest example: NC State allows a pot as small as half a gallon if it is 10 inches deep, and Maryland's table asks for 2 gallons. Both are real, both name the crop, and the line under the number says so — because the thing that actually decides a carrot is the depth, and a shallow wide tray fails at either volume.

So where the source states a condition that decides whether the number actually works, it sits on the page with the number rather than in our notes. It is not a disclaimer. If a value needed defending, we would not print it.

How deep the pot needs to be is now its own row, where we have it. It used to be squeezed into the line under the volume — that is how garlic's 4–6 inches reached you at first. A depth is a different measurement from a volume and it deserved its own place, especially for root crops and bulbs, where it is the number that decides whether the pot works at all. Blueberry makes the case on its own: the page Maryland publishes about growing blueberries in containers gives no gallon figure whatsoever, only "at least 24 inches deep". We hold it for seven crops so far and the column is mostly empty on purpose — a depth only goes in when a source states one. We will not work it out from the gallons, because that means guessing how wide the pot is, and nobody published that either.

Every plant knows its family, and its relatives

Each plant page names the botanical family the crop belongs to, and lists the relatives we hold. Tomato says Solanaceae and points at pepper, potato and eggplant. Dill says Apiaceae and points at carrot, celery and parsley.

It is there because relatives share pests and diseases. That is the whole reason rotation plans are written by family: if something went through your tomatoes, the bed remembers, and a pepper is the next thing it will happily take. Knowing what counts as a relative is the part that is hard to hold in your head, so the page holds it for you.

Every family comes from two independent botanical authorities, asked separately. 103 of our 107 plant groups got the same answer from both. Four did not, and rather than quietly pick one we wrote down what each said and why we went the way we did — beets, chard, spinach, quinoa and orach are the four, and the two systems genuinely disagree about how wide their family is.

You can have the crops and the varieties

It's a free download, CC BY 4.0. Use it commercially, redistribute it, build on it — credit us and that's the license. Every crop and every variety we hold is in it: spacing, sowing depth, sun, days to maturity and what that counts from, how big it gets, hardiness zones, pot size and how deep the pot needs to be, pests and diseases, and when to start relative to your last frost.

A second file records the source of every value in the download, with the title and a link for each publication. If a number in the app looks wrong to you, that file is how you check us rather than take our word for it.

What the download leaves out

The parts that only do work inside software. The growth stages and the care actions inside them — the thing that turns a date into "tie the vines up this week". The watering rates and the temperature thresholds our alerts fire on. The frost-relative calendar those stages resolve onto.

None of that changes what you do with a trowel, and all of it is what took us eighteen months. The download page counts exactly what came out. If you want the complete record — for research, or to review our work properly — email hello@gardentrack.app and say what it is for.

When a line is missing, and what the missing line says

Most plant pages have a gap or two — a temperature nobody publishes, a spacing no extension service states for that crop. Where that happens the page says so rather than leaving a hole you have to notice: "There is no line here for the temperature frost starts to hurt it, because we have no figure to put there. We went looking and wrote down what we read — it is missing on purpose, not because nobody checked."

That sentence is deliberately about us and not about the world. We can tell you what we hold. We cannot tell you that nothing exists anywhere, and a page that said so would be claiming to have read everything.

Some gaps have an answer in them

A few plants have an empty line for a reason worth telling you, and those pages say the reason instead of the general sentence. Three of them are about running to seed:

  • Swiss chard — "Chard will not run to seed in its first summer, however hot it gets." Minnesota Extension says so outright: it does not flower until it has lived through a winter. That is not a gap; it is an answer.
  • Kale — "Nobody has published a warning sign for kale." Five university pages were read and not one names something to watch for. The answer is genuinely missing.
  • Parsley — "Parsley does run to seed, but in its second year, because that is what a two-year plant does." It happens, and watching the weather for it would tell you the wrong thing.

All three used to show the same blank. They mean three different things, and now they say which.

Common questions

The download used to have every variety. Why is it smaller?
**Because it was a superset of what our own app reads — narrowed on 25 August 2026.** It carried all 3,174 varieties at full depth, which is eighteen months of sourcing and the inputs our planner runs on, in one free file. Being open about our data and giving away all of it are two different decisions, and we had collapsed them into one. **What is in the file now:** every one of our crops, with their full care data — that part did not change — and **the 500 most-searched varieties**, with what a seed packet prints: name, type, color, size, whether it is heirloom or a hybrid, and days to maturity. **How the 500 were chosen, so you can check it:** US monthly search volume for the variety's own name, measured on 25 August 2026, ties broken alphabetically. It is a fact about what people look up rather than our opinion about what matters, and anyone can re-run it. **What is not in it:** the parsed maturity range, growth habit, determinacy, years to first harvest, pollination needs, chill hours, ripening season, height and heat level. The file lists all of that in a `withheld` section rather than leaving you to notice. **If you need it, email us** — that offer is real and it is in the file. **One thing we would not trade.** Days to maturity only ships when we can also say what it counts from — sowing or transplanting. That difference is about a month, and a number without it is a misleading number. Where we do not know the basis, we publish neither.
Why did some varieties stop showing the same word twice?
**Because they were, and it was our doing rather than the seed company's.** A variety row shows its type as a small chip beside the name, then anything else known about it as separate pills. When a source listed the type as one of the traits too, both printed — so Bergamot Orange read *Bergamot Orange · sour · sour · strongly scented*. **66 varieties did this**, spread across bok choy, celery, elderberry, flax, lemon balm, orange and others. A row now leaves out anything its own heading has already said. Nothing has been deleted from the data — the trait is still recorded exactly as the source wrote it; the row just does not say it twice. ⚠ **One thing we deliberately did not do.** A few crops store the type field as the whole list of options that crop has — a blackberry carries *erect, semi-erect, trailing, thorny, thornless* whatever kind it actually is. If we had matched against that, we would have hidden *thornless* from three blackberries that genuinely are thornless, on the strength of a field that says nothing about them. So the rule only applies where the type is a single answer about that one variety.
My dill has a big green caterpillar. Why does the app say its hornworm entry might not apply?
**Because the caterpillar on a dill is almost certainly not a hornworm, and the two look alike in the one way that matters.** A university page does list tomato hornworm as something dill is susceptible to, so we keep the entry rather than quietly dropping it. But the big green caterpillar that eats dill down to bare stalks is normally a black swallowtail larva — a parsleyworm. Wisconsin's page on it names dill, parsley, fennel, celery, caraway and carrot as its food plants, and photographs the caterpillar on dill. We carry that entry too. **Here is the trap.** Disturb a parsleyworm and it rears up and flicks out a forked orange horn from behind its head. So you are looking at a fat green caterpillar with a horn, on a plant, and 'hornworm' is the obvious word for it. If we handed you the tomato hornworm description right then, it would half-match and you would stop looking. **So the app says the description was written about a different plant family and may not describe what you are looking at.** It is not hiding the entry and it is not guessing for you. It is telling you the one thing that would otherwise be invisible: that this particular description was not written about your plant. Worth knowing before you act — a parsleyworm becomes a black swallowtail butterfly, and plenty of gardeners leave a few on purpose.
The variety list says who described it. Why does that matter?
Because most of those descriptions come from whoever is selling the seed. About seven in ten of our variety traits — drought tolerant, long storage, good in a container — are from a seed catalog describing its own product, and printing that flat would make it look like our finding rather than theirs. So the page says "Fedco Seeds describes this as" instead. The rest come from places like NC State Extension, and those say "records" rather than "describes", because a university writing up a plant and a shop writing up its stock are not the same kind of claim and should not read as though they are. Where nobody is on record at all, the row shows nothing — a trait with no author would be us making the claim, which is the one thing we are trying not to do.
Where do the app's numbers come from?
US university extension services, mostly. They're the land-grant universities that publish growing guidance for their own state and revise it as the science moves. 983 separate publications sit behind the database, and the app shows you the source on the plant's own page.
Is all of it sourced?
No, and the database says which parts are not. 4,486 of 5,412 values cite a publication. The other 926 each record a reason: 743 because nobody publishes that particular fact and we filled it from ordinary horticulture, 183 because a person set it and attached no citation. The free download carries the same record for every value in it.
Can I download it?
Yes, free, at the link above. Every crop and every variety is in it, CC BY 4.0, which means you can use it commercially, redistribute it, or build a product on it. Credit us and that's the whole license. There's a second file recording where each value came from, and a plain-text readme that explains the gaps before it explains anything else.
Is the download everything you have?
No, and the page says so with counts. Every crop and every variety is there. What is not there is the part only software uses: the growth stages and the care actions inside them, the watering rates, the heat thresholds our alerts fire on, and the frost calendar those stages sit on. The two frost temperatures — the point a crop takes damage and the point it dies — used to be held back with them, and are not any more: they went in on 12 August 2026, because deciding whether to cover a bed tonight is a thing a person does with a number, not something only software reads. If you want the complete record, email hello@gardentrack.app and ask.
A tomato variety says both a shape and a determinate or indeterminate. Why two?
Because they answer different questions and they used to share one slot. The shape — bush, vining, upright — tells you whether it needs support and how much room it takes. Determinate or indeterminate tells you whether the fruit arrives in one glut or keeps coming until frost. A tomato can be a dwarf and still be indeterminate, and when both had to fit in one field, whichever was written second won and the other was thrown away. Three varieties in our own file said 'dwarf indeterminate' in their notes and only the dwarf survived. Split into two fields on 12 August 2026: 133 rows moved to the new one, and the 133 now show no shape at all, because a shape was never recorded for them and inventing one is the same mistake pointing the other way.
Some plants say how long to start them indoors before planting out, and others count from the last frost. Why two?
Because those are two different questions and gardeners ask both. Counting back from your last frost tells you a date, and needs you to know that date. Counting forward from when you intend to plant tells you how long the seedling sits on the windowsill, and needs nothing — it is a fact about the plant. Extension services publish both kinds, and we used to be able to record only the first, so every time we found the second we wrote it into a note and lost the number. Onion, cucumber, okra, summer squash and pumpkin now carry it properly.
Every plant page names a plant family now. What is it for?
It tells you what a crop is related to, and relatives matter for one very practical reason: they share pests and diseases, which is why rotation plans group by family rather than by crop. Tomato, pepper, potato and eggplant are all one family, so the trouble one had last year is the trouble the next one inherits if you plant it in the same spot. Every family we show comes from two independent botanical authorities that were asked separately, and where the two disagreed we recorded both answers and why we picked one.
A plant page put blossom end rot under "not illnesses". What is that row for?
**It is the row where spraying will not help.** Everything in it comes from how the plant has been grown — how evenly it was watered, how cold it got, how much sun landed on the fruit — rather than from an insect or a germ that arrived and attacked it. Blossom end rot is the clearest case: the dark, sunken patch on the bottom of a tomato looks like the worst thing that could happen to it, and no fungicide touches it, because nothing has infected the plant. Until now those sat under **Diseases** on the crop page, purely because that was the only other list we had — so the page quietly sent you looking for a spray you did not need. Root knot nematodes moved the other way, into **Pests**, because they are an animal. **Where we do not know which kind a problem is, it stays exactly where it was and we say nothing about it** — an unknown shown as an illness is how a plant doing nothing wrong gets treated for something it has not got.
Why does a problem sometimes say the description may not be about my plant?
Because the same common name can belong to two completely different organisms. Brown rot on a peach and brown rot on a potato are not the same thing at all, and the page we sourced our description from was written about one of them. Where a crop sits outside the plant family a description was written for, we now say so on that crop instead of quietly handing you the wrong one. Sometimes we say plainly that nobody has checked which organism yours is — that is a real answer, and better than a confident guess.
A plant page says "Not recommended here" for one region. Is that a gap in your data?
No — it's the opposite, and telling the two apart is most of the point. On the autumn planting list, some rows show dates and some say "Not recommended here". That second one is what the extension service actually printed: they looked at that region, and the answer is don't. A blank would mean we never checked. Until 20 August 2026 we showed a blank for both, so a South Carolina Piedmont gardener reading the tomato page saw a window for the coast and no sign their own region had been considered at all. Three crops on that chart say don't — tomato, head lettuce and winter squash — and now they say it.
The plant page tells me to sow by a date. Where does that come from?
Your own first frost, minus how long the crop takes. Nothing about it is published for your town — it's arithmetic on two numbers, the frost date for the zone or ZIP you set on that page, and the crop's days to maturity from our sources. That's why it works anywhere, while the South Carolina windows above it only work in South Carolina. Two things to know: the date uses the fastest figure in the range, so it's the last date that could work rather than one that usually does, and a days-to-maturity number is a catalog figure for a good season, not a promise. Some crops say "set plants out by" instead. That's not a slip — for those our sources measure the days from setting a young plant out, not from seed, so a sowing date would be weeks wrong.
Why do some crops have more detail than others?
Because the sources do. A tomato has been written about by every extension service in the country; a less common herb has not. Rather than fill the gap with a guess and let it look like a fact, the record leaves it empty and marks why.

More on this

Shipped 2026-08-06. Last checked 2026-08-26. Something here wrong or out of date? Tell us — support docs going stale is a bug.