How to Read a Hay Analysis

We send our hay to a lab and have it tested. Nobody makes us do that, and most growers don’t bother. Every sample we pulled this season came back under 10% sugar and starch, which is the figure that matters most if your horse is metabolic, insulin resistant, or prone to founder. Every report is below, unedited, including the one that tested worst.

If that’s what you came for, call Steve at 630-777-6179 and we’ll work out what fits your horse.

If you’d rather understand what those numbers actually mean, so you can read any hay report and not just ours, the rest of this page walks through one of them start to finish.

Most explanations of a hay test walk you down the page line by line, in the order the lab printed it. That’s the order the lab needs. It isn’t the order you need. Nobody opens a forage report wanting to learn about dry matter percentage. You open it with a particular horse in mind and have three questions about him:

Can he have this much sugar? Is there enough protein? Will he actually eat it?

Everything else on that sheet is supporting evidence for one of those three. Here’s how to get to the answers, using a real report: sample 1989368 (following), taken off our farm and sent to Analab, Agri-King’s professional agricultural testing laboratory.

We picked this one on purpose. Of everything we’ve run this season, it came back with the highest sugar of the group. A walkthrough that uses your best number teaches nothing.

Every analysis opens with a description line. On our sample 1989368 it reads Mix Hay, and that’s worth stopping on, because the hay type tells you more about what to expect than any single measurement below it.

At Galusha, the labels below are what we use to describe how much alfalfa is in a bale:

So Mix Hay is not grass hay. It’s somewhere between a third and two thirds alfalfa, and that single fact predicts a lot: higher protein, higher calcium, softer eating, and a different fit for an easy keeper than a straight grass hay would be.

Two practical notes. Not every seller uses these categories the same way. Plenty of people will call anything with a stem of alfalfa in it “alfalfa mix,” so ask what the actual proportion is rather than trusting the label. And the category is a range, not a recipe: two loads both honestly labeled Mix can sit at 35% and 65% alfalfa and feed noticeably differently.

There’s one thing the label will not tell you, and it happens to be the number most people came for. We’ll come back to it at the end.

This is the question that sends most people to a hay test in the first place, and it’s the one where the report is genuinely useful.

The number you want is ESC plus starch.

ESC, ethanol-soluble carbohydrates, is the simple sugar fraction. Glucose, fructose, sucrose. Absorbed in the small intestine, and they move blood sugar and insulin.

Starch is starch. Also digested to glucose, also moves blood sugar.

Add them and you have the fraction of the hay that behaves like sugar inside your horse.

Sample 1989368: ESC 4.83%, starch 4.43%. Total 9.27%

Under 10% is the working benchmark most veterinarians and metabolic-horse groups use for horses with insulin dysregulation or EMS. Some institutions set it slightly higher, at 10 to 12%. This hay clears either one, and it doesn’t clear the stricter one by much, which is exactly why it’s the useful example.

One shortcut if you’re reading an Analab report specifically: the line labeled Non Structural Carbohydrates is ESC plus starch, already added. It reads 9.27 on this sample. Other labs define that line differently, so on a report from somewhere else, add the two yourself rather than trusting the label.

And one term that causes trouble: ESC is not WSC. Water-soluble carbohydrates sweep in fructans along with the simple sugars. Fructans ferment further back in the gut and don’t produce the same blood sugar response, so a WSC number reads higher and scarier than the fraction you actually care about. That does not make fructans harmless: a large enough fructan load can still trigger hindgut acidosis or laminitis. It just means they aren’t what the sugar-and-starch number is measuring. If someone quotes you a sugar number, ask which one it is.

Then the part we’d rather say plainly than bury at the bottom: 9.27% clearing a general benchmark does not mean this hay is safe for your laminitic horse. It means one number cleared one threshold. Your vet knows things a forage lab never will: where his insulin sits, what he weighs, what else is in the bucket, whether he’s out on grass. Take the report to them. That’s what the report is for.

Crude protein (CP). Sample 1989368: 15.90%

For orientation: most mature horses in light work need somewhere around 8 to 10% crude protein in the total diet. Grass hays often land between 8 and 12%, though an early cutting can run well above that. Once alfalfa enters the bale the number climbs, and straight alfalfa can go past 20%.

At 30 to 70% alfalfa, 15.90% is right where a Mix should land. If you’re feeding a hard keeper, a broodmare, a growing horse, or something in real work, that’s protein you’d otherwise be buying in a bag. If you’ve got an easy keeper already carrying condition, it’s more than he needs. Not harmful. Just more than he needs.

Worth saying since we’re being straight about it: our Alfalfa Mix samples ran higher, and one came back at 19.63% CP. At 70 to 90% alfalfa that’s no surprise, but it is a lot of protein. For some horses it’s exactly right and for others it’s the wrong hay, and anybody selling you alfalfa mix who won’t tell you that is hoping you don’t ask.

One limitation of the number itself. Crude protein is measured by finding nitrogen and multiplying by 6.25. It tells you how much protein is present, not whether it’s the right protein. What matters is whether the specific amino acids your horse can’t manufacture are actually in there, and lysine is usually the first to run short. A hay can post a respectable CP and still be the wrong protein. That’s where hay stops answering the question and a ration balancer picks it up.

This is where the report starts running out of things to tell you, and almost nobody says so.

NDF (neutral detergent fiber) is the closest the lab gets. It measures the structural cell wall. As NDF climbs, horses eat less of the hay and get less out of what they do eat. As a rough guide, the low 50s and below is comfortable for most horses, the high 50s is workable, and up in the 60s you’re into hay that some horses will sort through rather than clean up.

Read that as a guide, not a verdict. Straight grass hay naturally lives in the 60s, and that isn’t a defect, it’s what grass hay is. A high NDF number means a lot of bulk per calorie, and for a horse who needs to chew for hours without taking on weight, that’s the entire point. The number tells you what kind of hay you’re holding. It doesn’t tell you whether that’s the kind your horse needs.

ADF (acid detergent fiber) is the tougher subset. It tracks digestibility, and lower is better. The 30s are good.

Sample 1989368: NDF 46.79%, ADF 31.38%. Both comfortable. This is soft, digestible hay.

The tradeoff nobody mentions

Our own samples demonstrate this better than any explanation could.

Sample 1989369 is also a Mix Hay, and it tested at 5.43% sugar and starch, far below the one we’ve been discussing, and on that single line, the obvious winner. But its NDF came back at 54.59%, its ADF at 35.46%, and its protein at 12.23%. Its digestibility is the lowest of them all.

So the hay with the better sugar number is also the coarsest, the least digestible, and the lowest in protein. That isn’t coincidence. 1989369 was cut at a more mature stage than 1989368, and maturity is what drives fiber up and protein down. The sugar came down with it.

Shop a hay test on one line and this is how you end up with hay your horse sorts through. For a horse with real metabolic trouble, 1989369 is probably still the right call and the fiber is a price worth paying. For a horse who just runs a little heavy, the softer hay at 9.27% may get eaten, hold condition, and waste less. No line on the report resolves that. You resolve it, with your vet, for your horse.

And then there’s the part the lab can’t measure at all

NDF is the beginning of the intake answer, not the answer.

Here’s why, and it’s easy to miss: the lab grinds the sample before it tests anything. Grinding destroys the single most important thing about how a hay eats, which is leaf.

Leaf is where the nutrition and the flavor live. Stems are what your horse pushes into the corner of the stall at midnight. A leafy bale and a stemmy bale off the same field in the same week can return nearly identical NDF, because once it’s powder there’s no difference between a leaf and a stem. Your horse can tell. The machine can’t.

Palatability does not appear anywhere on a hay analysis. There is no line for it. It’s the one criterion you evaluate with your own hands and your own nose, and it’s very often the criterion that decides whether hay gets eaten or wasted. A test can tell you a hay is nutritionally appropriate. It cannot tell you your horse will touch it.

So when you’re standing in front of the stack:

• Smell it. Sweet and grassy is what you want. Musty, sour, or sharp means it went up damp. Dust means trouble for any horse with an airway, and a horse will back off dusty hay well before you notice anything wrong with it.
• Squeeze a flake. If it’s soft in your hand, it’s soft in his mouth. If it scratches you, he’ll sort through it.
• Shake a flake out and look for leaf on the ground. Leaf shatter means it was raked or baled too dry, and the best part of the hay is now chaff on your barn floor.
• Check maturity, not just color. Seed heads out and stems like pencils means it was cut late. It will test as fiber and eat like fiber. Bleaching on the outer flakes of an outdoor stack is cosmetic; brown all the way through is heat or rain.
• Ask which cutting it is. First tends to be coarser and stemmier. Second is usually the leafy, balanced middle. Third is finest and richest, which isn’t automatically better. For an easy keeper it can be too much.
• Consider what he’s eating right now. The most common reason a horse refuses perfectly good hay is that it isn’t the hay he had last month. Change over across a week or more, not overnight.

Every sample we’ve run this season, sorted by sugar:

Every sample we’ve tested this season came in under 10% sugar and starch. Most came in under 6.2%. The one we spent this entire page on was the highest of the group.

Now look again at the type column

Read down the protein and NDF columns and the labels roughly behave. Alfalfa Mix clusters high in protein and low in fiber. Grass clusters the other way. Mix sits in between. Knowing the type puts you in the right neighborhood on both, though not reliably: one of our grass hays came back at 15.37% protein, higher than three of our Mix samples and well above where grass hay usually lands. An early cutting will do that.

Now read the sugar column, where the label stops helping at all. Our Mix Hay samples run from 3.44% to 9.27%. Our grass hays run from 4.36% to 7.29%. Same tags, same farm, same season, and in the case of the Mixes nearly three times the sugar between the lowest and the highest. Our sweetest grass hay tested above three of our five Mix samples, even though plenty of people will tell you grass hay is automatically the safe choice for a sugar-sensitive horse.

That’s the whole argument for testing, and we didn’t plan it, we just ran the samples and read them. The type on the tag gets you a rough expectation for protein and fiber. It tells you nothing reliable about sugar. There is no way to know the sugar in a particular load of hay except to send it to a lab, and that means a seller who hasn’t tested cannot answer the question no matter how confident they sound.

Nobody makes us do this. There’s no regulation anywhere requiring a hay grower to send a sample to a lab, and most don’t. We run the tests because we want the number, and because a number you can check beats a description you have to trust.

We feed our own horses off these same stacks.

Everything above came from samples we pulled this season, and you can download the full Analab reports here: https://galushafarm.com/files/2026/09/10-haytestresults-Sept2026.pdf. They’re dated, and that’s exactly the point. Hay changes every year, every field, every cutting. A report from this September describes this September’s hay and nothing else, which is why a seller quoting you figures from two years ago is quoting you nothing.

Which raises the obvious question. What about the hay you’re looking at right now?

We’ll test it for you. $100, and it comes off the load if you buy. Pick any hay we have on hand. We pull the cores properly, which matters more than people assume, because a handful grabbed off the end of one bale tells you about that bale and not about the stack. We core multiple bales across the stack, send the sample to Analab, and forward you the report exactly as it comes back, unedited, the same way you’ve just read ours. Results usually take about 3-4 days. The report is yours either way. If the numbers say our hay is wrong for your horse, you’ve spent a hundred dollars to avoid a considerably more expensive mistake, and you can take that report to your vet or hold it up against whatever the next seller tells you.

What a test can honestly tell you, and what it can’t. A core sample describes the stack it came from. It is not a certificate on every individual bale, and no honest grower will tell you otherwise, because hay is a crop and crops vary. So we cover that gap two other ways. You can pick your hay bale for bale out of what’s on hand before you commit to any of it, and if something turns out unusable we replace it bale for bale. The full terms are on https://galushafarm.com/hay-quality-promise/ page. Between seeing it, testing it, and being covered afterward, there isn’t much left to guess at.

Tell us what you’re aiming at and we’ll point you at it. If your vet has given you a target, whether that’s sugar and starch under a certain figure or protein in a particular band, say so when you call. We’ve grown hay and tested it every year long enough to know which of our fields and cuttings tend to land where. We won’t promise you a number before the lab returns it, and you should be wary of anyone who does. But we can usually tell you which hay to test first, and that beats testing at random.

And if what you need isn’t in the barn, we can grow it. Once we know what works for your horse, we can plan around it: species, timing, and when we cut. Hay is an annual crop, so that’s a conversation about next season rather than next week. It also means we keep your numbers on file and aim at them on purpose. A fair number of our longest-standing customers started with one question about one sample.

Call Steve Berning at 630-777-6179, extension 3

Galusha Farm — Warrenville, Illinois. We deliver throughout Chicagoland and ship beyond it.