If you are short of trouble, take a goat. ~Finnish Proverb
Milk test sounds way more intimidating than it actually is. The first time you start reading about it, you get hit with DHI, DHIA, DHIR, DRPC, CDCB, DCR, VT, PTA, PTI, *M, AR, Top Ten, and suddenly you’re wondering if you accidentally signed up for a government program instead of weighing goat milk.
But underneath all those acronyms, milk testing is pretty simple. You milk the goat, weigh the milk, take a sample, write things down carefully, and send the samples and paperwork where they need to go. The lab tests the sample, the data gets processed, and eventually those records can become part of your herd records, award records, and genetic evaluations.
Once you understand how the system works, milk test stops feeling like mystery paperwork and starts feeling like one of the most useful tools we have as dairy goat breeders.
Milk test is the system used to officially record how much milk a doe produces and what is in that milk. That means you’re not just guessing that a doe is a heavy milker because the bucket feels full. You’re getting actual numbers for milk weight, butterfat, protein, somatic cell count, days in milk, lactation totals, 305-day projections, and sometimes MUN, depending on the lab.
That information can help you manage your herd, make better breeding decisions, and document production for milk stars, Advanced Registry, Breed Leader, and sometimes Top Ten eligibility. It also feeds into genetic evaluations, and that part matters more than a lot of people realize.
Every production PTA, every production index, every genetic evaluation for milk, fat, and protein starts with real people in real barns weighing real milk. Your barn sheet is not just a piece of paper. It is part of the national dairy goat data system.
Let’s untangle the alphabet soup. DHI stands for Dairy Herd Improvement. That is the milk testing system. DHIA stands for Dairy Herd Improvement Association. This is usually the organization or affiliate that helps manage your testing program, paperwork, herd code, test setup, and sometimes your connection to the lab and processing center.
DHIR stands for Dairy Herd Improvement Registry. In the ADGA world, this is the official registry program that recognizes production records and awards. So, in plain English, DHI is the testing system, DHIA is the group helping run the testing process, and DHIR is the registry side where official records and awards happen.
You may also see DRPC, which stands for Dairy Records Processing Center. That is where the data gets processed after the lab and paperwork are submitted. CDCB stands for Council on Dairy Cattle Breeding. Despite the name, goat production data goes there too. CDCB uses milk test data in genetic evaluations, which eventually show up in places like ADGA Genetics.
The milk data flow looks something like this: farm to component lab to processing center to CDCB to ADGA to online databases and genetic evaluations. That means the numbers you see on ADGA Genetics do not appear out of thin air. They come from milk stands, barn sheets, samples, labs, and a whole lot of recordkeeping.
There are three big reasons I think milk test matters: management, genetics, and recognition. Some people start milk test because they want milk stars. Some start because they want better herd data. Some start because they want their herd contributing to genetic evaluations. All of those are valid reasons, but I think the real value is that milk test teaches you what your does are actually doing instead of what you think they are doing.
Milk test gives you measurable information instead of barn guesses. We all have does we think are milking well because they look good, feel good, and seem productive. But milk test will humble you. Sometimes the doe you thought was carrying the milk room is just dramatic and good at looking productive. Sometimes the quiet doe in the corner is the one doing the real work.
Milk weight tells you how much she is producing. Butterfat and protein tell you about milk quality. Somatic cell count gives you another piece of the udder health picture. Days in milk help you understand her lactation curve and persistency.
Somatic cell count, or SCC, is another piece we need to pay attention to because it can give you an early warning that something is going on with udder health. In goats, SCC can be a little trickier to interpret than in cows because goats naturally shed more cells in their milk, and counts can rise with stage of lactation, stress, heat cycles, illness, or other normal goat chaos. So I don’t treat one high SCC as a full diagnosis by itself. What I want to see are patterns. Is one doe consistently higher than the rest of the herd? Did a doe suddenly jump from her normal range? Are higher counts happening alongside changes in milk texture, production, udder feel, or behavior? SCC is useful because it helps you catch possible mastitis or inflammation before it becomes obvious in the bucket, and it gives you another reason to look more closely instead of waiting until there is a real problem.
MUN, or Milk Urea Nitrogen, is one of the reasons I specifically use Langston. Not every lab gives you MUN, and I want that information because it gives me another way to look at how efficiently my goats are using dietary protein. If MUN is consistently high across the herd, you may be feeding more protein than the goats can use well. That extra protein is not magically turning into milk. It is being wasted, and protein is usually the expensive part of the ration.
I would not chase one perfect MUN number, especially with goats. Herds vary, seasons vary, and stages of lactation vary. The useful part is watching trends in your own herd. If your herd average is consistently high or consistently low, that gives you a reason to look more closely at the ration, forage quality, energy balance, and what your higher-producing does may need.
Milk testing can help answer practical questions. Who is actually producing well? Who is holding production later into lactation? Who has strong components? Who looks good but is not paying rent? Are my higher producers getting enough feed? Is my ration working the way I think it is? Do I have udder health issues starting quietly? That is management gold.
Milk test is also the foundation of production evaluation. When you see a PTA for milk, fat, or protein, that number exists because milk records were collected, processed, and compared across animals and herds.
PTA stands for Predicted Transmitting Ability. It estimates what an animal is expected to pass to offspring for a specific trait. It is not the same thing as the animal’s own production. A doe can milk well herself but not necessarily transmit that same level of production strongly. A buck obviously does not milk at all, so his value for production has to come from his relatives, daughters, and genetic evaluation data.
This is why milk testing daughters matters so much. Bucks need daughter data. Families need records. Breeders need something more useful than “her dam was a lot of milk.” And yes, sometimes “a lot of milk” means the bucket looked full once in April. We need better than that.
Milk testing is not only about the doe in front of you. It helps us understand what families transmit. When records are collected across herds, they help build production evaluations. Those evaluations help estimate genetic merit for milk, fat, and protein. Those numbers show up as PTAs and indexes.
A PTA does not tell you what the animal personally produced. It estimates what that animal may transmit genetically. A positive Milk PTA means the animal is predicted to pass more milk production than average. A negative Milk PTA does not mean the animal is bad. It means the animal is predicted to transmit less milk than average for that trait.
PTI stands for Production Type Index. It combines production and type information into one number. There are different versions depending on whether the emphasis is more production-heavy or more type-heavy. PTA is a trait prediction, PTI is an index, and neither one replaces common sense.
You still need to look at the whole animal, the pedigree, the herd, the mammary system, the production record, and what your own breeding goals actually are. But without milk test data, we are missing a huge piece of the puzzle.
Milk testing is also how does earn official production recognition. Depending on the registry, test plan, and rules, milk testing can lead to milk stars, Advanced Registry, Breed Leader recognition, Top Ten eligibility, and production records that show up behind animals in pedigrees.
In ADGA, a doe may earn an *M when she meets the requirements for milk production, butterfat, or protein through approved testing. The number before the star shows the generation of milk star recognition in the maternal line.
For example, a 3*M doe is a third-generation star milker. That means she earned her star, her dam earned hers, and her maternal granddam earned hers. Bucks can also carry production designations through their dam or offspring. That matters because a buck’s value is not just what he looks like. It is what his daughters may do.
This is the part I wish more people talked about before they buy supplies. Your goal determines your test plan.
If you only want herd management data, you may not need the most complicated plan. If you want milk stars, you need a plan that qualifies for official recognition. If you want Top Ten eligibility, you need to be even more careful because supervision, verification testing, DCR, and paperwork all matter a lot.
Before you start, ask what you are actually trying to accomplish. Are you testing for your own management records? Are you trying to earn milk stars? Are you aiming for Advanced Registry? Are you hoping for Top Ten or Breed Leader eligibility? Do you want your data contributing to national genetic evaluations? Can you realistically find a supervisor and verification tester?
That last question is a big deal. I quit partway through last year because I could not find someone to do the verification test. So when I say “line up your VT person early,” I am not saying that to sound organized. I am saying that because I learned the annoying way.
Before official testing starts, you need to enroll correctly. For ADGA records, that means making sure you are enrolled in DHIR. Depending on the year and your goals, this may be part of ADGA Plus or a separate DHIR application. ADGA Plus can bundle DHIR, Linear Appraisal, and DNA testing services, but you still need to pay attention to the deadlines and requirements for the current year.
Then you need a DHIA or service affiliate. This is the organization that helps manage testing, paperwork, herd codes, samples, and record processing. Like I mentioned, we use Langston because I want the MUN count, and I like having that extra piece of management data in addition to milk, fat, protein, and SCC.
Your DHIA may also tell you which lab to use, where to send samples, how to submit barn sheets, what forms are required, how to handle invoices, and how your scale needs to be approved or calibrated. Do not wait until your first test day to figure that out. Milk samples and panic are not a good combination.
Your test plan determines who weighs the milk, who samples the milk, how often a supervisor is involved, and what awards your records may qualify for.
The main categories are standard supervised plans, owner-sampler plans, shared or alternate responsibility plans, and one-day milk tests.
Standard testing usually means a certified supervisor comes to your farm, weighs milk, collects samples, and signs off on the test.
Owner-sampler testing usually means you weigh and sample your own milk. This can be more flexible and more affordable, but it may not qualify for everything. It may require a verification test for certain awards.
Shared or alternate plans split responsibility between the owner and supervisor. These can be a good fit for small herds or areas where supervisors are hard to find, but you need to understand exactly what your plan allows.
A one-day milk test is usually a supervised 24-hour test, often connected to a show or special event. It can be useful for certain milk stars, but it is a snapshot, not a full lactation record. A one-day test tells you what happened that day. A 305-day record tells you the story. Both can be useful, but they are not the same thing. I am not going to plan a breeding around a handful of one day milk tests.
Depending on your test plan, you may need a supervisor. This does not have to be some mysterious professional dairy person who descends from the mountains holding a clipboard. You can often use a friend, neighbor, or another responsible person who is willing to learn the process and follow the rules.
The important part is that the supervisor cannot be invested in your herd. They should not own your animals, have a financial interest in your records, be trying to market your animals, or benefit from your does earning awards. The point of the supervisor is to provide an outside check on the test day data.
A good supervisor needs to be dependable, careful with paperwork, comfortable following instructions, and willing to show up on time. They do not need to be a goat genetics expert. They do need to understand that the integrity of the test matters.
If you are planning to test, find this person early. Do not wait until the week before your first test. And if your test plan or award goal requires a verification test, find that person early too, because the VT person usually cannot be the owner or regular supervisor.
Before test day, make sure you have what you need. Usually that means an accepted, calibrated scale, barn sheets or official test paperwork, sample vials from your lab or DHIA, labels or a marker, a dipper or ladle for sampling, a clean pail or collection container, ice packs if samples need to stay cool, shipping supplies, your doe list with correct IDs, and your supervisor lined up if your plan requires one.
Scale accuracy is key. If your scale is wrong, your milk weights are wrong. If your milk weights are wrong, the record is wrong. You need to use a scale that is accepted by your testing program, and it may need to be calibrated on a schedule.
That means “I bought a scale” is not the same as “I am ready for official milk test.” Use whatever setup works for your barn, as long as it is accurate and accepted by your testing program. Some people use hanging scales. Some use milk meters. I use a Waikato milk meter that weighs directly in the lines of my machine, but finding a place to calibrate it can be challenging.
The barn sheet is where milk test can go from “this is fine” to “why are there seventeen boxes and what does this code mean?” The barn sheet is not just a place to write milk weights. It tracks the information needed to make the record official and useful.
Depending on the form and test type, it may include herd code, test date, supervisor signature, doe identification, index or sample number, barn name, milk weights, fresh date, dry date, reproductive status, animals that left the herd, condition-affecting-record codes, fees, and sample counts.
Those condition codes matter because they explain things that may affect the test. For example, if a doe was in heat, sick, missed a milking, had a sample missed, or had some other issue that could affect her record, that needs to be recorded correctly. This is why milk test is not just about weighing milk. It is about keeping records clean enough that the data can actually be trusted later.
If you are supervising or testing, take a picture of the barn sheet before you send it in. This is such a simple habit, but it can save you later if something gets lost, damaged, delayed, or questioned. I would make this part of your routine every single test day: finish the sheet, check it, sign it if needed, take a clear picture, then mail or submit it.
Start by recording the time milking begins. Usually, you do not need the exact time for every single doe. You need the start time for the milking session.
Milk your does the way you normally milk them. Hand milk, machine milk, pipeline, jars, buckets, whatever your setup is, the key is that each doe’s milk is weighed individually and recorded clearly.
Milk should be recorded in tenths of pounds. That means 4.5 pounds, not 4 pounds 8 ounces. It means 3.2 pounds, not “a little over 3.” There is usually a place on the barn sheet for the milk weight. Put it where it belongs and write clearly. Future-you deserves readable handwriting.
Before you take the sample, mix the milk well. Butterfat does not distribute itself evenly. If the milk is not mixed, the sample may not accurately represent that doe’s milk.
Then use a clean dipper or ladle to put milk into the sample vial. Follow your lab’s instructions, but in many cases, you fill the tube partway at the first milking and add more at the second milking. The vial has preservative in it, and the milk may turn slightly orange. That is normal.
Label the sample exactly how your paperwork requires. Usually the index number matters more than the doe’s name. The doe’s name is helpful, but the number is what keeps the sample matched to the record.
At the end, gently invert the tubes so the preservative mixes well. If your plan requires a supervisor signature, do not let the supervisor leave without signing. No signature can mean the test is not complete. Ask me how I know. Actually, don’t. I remember enough. Take a photo, like above, before you send it away!
After test day, samples go to the lab and paperwork goes wherever your DHIA or processing center tells you it needs to go. Follow their instructions exactly.
Some people ship samples with ice packs, especially in warm weather. I usually put the milk samples in one plastic bag and the barn sheet in another bag so nothing gets wet. This is also where all those random ice packs from livestock supply orders suddenly become useful. Congratulations, you were not hoarding. You were preparing. Seriously, I have boxes of Jeffers ice packs!
Before anything leaves your hands, take a picture of the barn sheet. I know I already said that, but it is worth saying twice because this is the kind of tiny practical habit that prevents big headaches later.
The lab tests the samples for components like butterfat and protein, and often somatic cell count and MUN. The processing center uses the milk weights and lab results to calculate records. Then that data can move into the larger system for registry records and genetic evaluations.
DCR stands for Data Collection Rating. Think of it as a measure of how complete and reliable the testing record is. It is generated through the system based on the quality, consistency, and completeness of the record.
For higher-level awards like Top Ten, DCR matters. If your DCR is too low, the record may not qualify even if the doe produced enough milk. This is one reason consistency matters. Missing tests, irregular records, or incomplete information can weaken the usefulness of the record.
A Verification Test, often called a VT, is a supervised test used to confirm that the milk records are accurate. It is not an accusation. It is not someone showing up because they think your doe is lying. Although, honestly, my girls would totally lie about what they put in the pail, especially to each other.
A VT is a quality control step. It protects the value of official records. For ADGA DHIR, a verification test for a twice-a-day herd validates three milkings over a 24-hour period. There is a pre-milking weight-only test to set the clock, then two official test milkings. The VT has to be done by someone other than the owner or the regular supervisor.
That is why you need to plan it early. Do not wait until the end of the lactation to find a VT supervisor. Your milk test year can be going beautifully, and then the VT logistics can throw a wrench straight into the milk pail.
Once your results come back, do not just look at who milked the most and call it done. Look at the whole picture.
A doe who gives a lot of milk but crashes quickly may not be as useful as a doe who milks steadily for a long time. A doe with moderate volume but strong butterfat and protein may be very valuable, especially in a herd focused on cheese, soap, or rich milk. A doe with a rising somatic cell count may need closer attention even if she looks fine in the barn.
Useful things to watch include peak production, persistency, butterfat percentage, protein percentage, somatic cell count trends, MUN trends, days in milk, how daughters compare to dams, and how paternal sisters compare across herds when data is available. Milk test lets you stop guessing and start seeing patterns.
One of the reasons milk test matters so much is that it does not just give you a number for one day. Over time, those records build into a much bigger picture. After enough testing, you can start seeing lifetime production summaries and genetic evaluation numbers that tell you not only what a doe has done, but also what she may be likely to transmit.
For example, a final production summary might show that a doe has six lactations on file, with average standardized production of 3,621 pounds of milk, 152 pounds of fat, and 141 pounds of protein. That already tells you quite a bit. First, she is not just a one-hit wonder. Six lactations means there is real staying power there. Second, the standardized averages let you compare her production more fairly across time, rather than just looking at one raw lactation and guessing.
Then you start getting into the genetic side of the report. In this example, the doe has a Milk PTA of 254, a Fat PTA of 5.2, and a Protein PTA of 9.0. Those are not just descriptions of what she personally produced. They are estimates of what she may transmit genetically to her offspring. So a positive Milk PTA says she is expected to pass along more milk than average. A positive Protein PTA says she is expected to improve protein production as well. That is the part that matters most if you are breeding for the next generation instead of just admiring the current one.
The reliability numbers factor in too. Here, the milk, fat, and protein reliabilities are all 77. That tells me these PTA numbers carry a decent amount of confidence behind them. They are not just a guess based on pedigree alone. There is enough performance data behind them to take them seriously, even if they are not the final word forever. Reliability is always worth checking because a flashy number means less if the confidence behind it is weak.
You can also see Predicted Producing Ability, or PPA, which in this example is 638 for milk, 183 for fat, and 223 for protein. PPA is more about expected future production performance for the doe herself, based on her records and herd data, while PTA is about what she may transmit to offspring. They are related, but they are not the same thing, and it helps to read them separately.
Finally, the Percentile Rank helps put the doe in context. A percentile rank of 96 means she stands very high compared to the population. That does not mean she is perfect, and it does not replace looking at udders, feet, legs, general durability, or the rest of the goat. But it does tell you she is not average, and it gives you a quick way to see that her production profile is genuinely strong.
This is why I like milk test so much. It starts with something very hands-on and very ordinary, just weighing milk and filling tubes at the barn. But over time, those test day records become something much more useful. They become a tool for management, for culling, for selecting replacements, and for making smarter breeding decisions.
This is also where milk test records become useful when I’m choosing a buck. For example, I have a doe, Nori, whose production I want to improve in the next generation. If I am choosing a buck for her, I don’t want to stop at his picture, his pedigree, or whether I like his dam’s udder. I want to look at his production evaluation and ask what he is actually expected to transmit.
A buck production summary is especially helpful here because the buck does not milk. His numbers come from the performance of the females behind him and, when he has them, the records of his daughters. In this example, the buck has daughters in 6 herds, 10 daughters total, and 20 lactations on file. That tells me there is actual daughter data behind the evaluation, not just pedigree guessing. His reliability is 58, so I would still leave room for the numbers to move, but I would take the direction seriously.
If I am trying to improve Nori’s production, this is not the kind of production profile I would reach for first. His Milk PTA is -78, Fat PTA is -1.9, and Protein PTA is -3.0. That means he is more likely to transmit lower milk volume, lower fat pounds, and lower protein pounds than breed average. His Fat % PTA is positive at 0.15, but that does not override the negative fat and protein pounds. In plain barn terms, he may improve richness a little, but he is not likely to improve total production.
Here’s the kind of production profile I’d be more interested in for a doe like Nori. This buck has a Milk PTA of 212, Fat PTA of 9.5, and Protein PTA of 3.4, so he’s more likely to move daughters in the direction I actually want: more milk, more total butterfat, and more total protein. His reliability is 52, so I still wouldn’t treat him like a heavily proven sire, but there’s enough information here to make the pattern useful. I’d also notice that his Fat % PTA is barely positive at 0.02 and his Protein % PTA is negative at -0.19, so I wouldn’t describe him as a component-percentage improver. What I like here is the increase in pounds of milk, fat, and protein. For Nori, that’s the kind of buck I’d rather investigate further, assuming he also fits her structurally and the planned inbreeding is reasonable.
That is the piece I want breeders to understand. A buck can be beautiful, correct, well-bred, and even useful in the right herd, but still not be the right buck for a doe who needs more production. You may notice that I mention it often in my blog, but too many people choose a buck because he has a big name even if he doesn’t fit what they need to improve. Nori needs more milk so I want a buck with positive Milk PTA, and I’d also pay attention to Fat PTA and Protein PTA because I do not want to gain volume while losing the components that make the milk valuable. I also want enough reliability to feel like the pattern is meaningful, especially if I’m making a serious herd-improvement decision.
This is why milk test helps even when you’re looking at bucks. The data helps me match the buck to the doe in front of me. I’m not just asking, “Is he a good buck?” I’m asking, “Is he the right buck for what this doe needs?” For Nori, I’d be looking for a buck who improves production without giving up the udder, feet and legs, and general durability I still need in the next generation.
Milk testing is not as scary as it looks. It is not magic. It is not just for huge dairies. It is not only for people chasing awards. It is one of the best tools we have for understanding what our does are actually doing and what our herd may be transmitting. I don’t do it for stars, though stars are pretty. I do it to understand how to improve my herd.
It helps with feeding, culling, breeding, marketing, and the whole breed when the data is used in genetic evaluations. And yeah, the acronyms are a lot, but at the barn level, milk test really comes down to enrolling correctly, choosing the right plan, using an accepted scale, milking the doe, weighing the milk, mixing the milk, taking the sample, writing it down, shipping it correctly, and doing it consistently.
That is where the data starts. Next, we’ll use this foundation to look at production evaluations and CDCB data in ADGA Genetics, using a real goat as an example. Because once you understand where the numbers come from, reading them gets a whole lot easier.
