No one can watch every cow every day. As dairy herds grow and labor remains constrained, that creates a practical gap between what is happening in the herd and what the farm team has time to observe.
Automated monitoring can help close that gap by creating consistent observations of individual cows and highlighting changes that may warrant a closer look. The technology does not replace visual observation, hands-on examination or veterinary judgment. Its role is to help farm teams decide where those resources should be focused.
“The problem that we’re trying to solve is that no one can watch every cow every day,” says Nicole Mozeliak, CEO of Herd-i US.
For Mozeliak, who comes to dairy from an operations background, the challenge is fundamentally one of capacity. Locomotion and body condition are useful indicators of herd health, she says, but farms do not always have someone available to monitor every cow consistently.
From Monitoring Every Cow to Finding Exceptions
Automated camera systems can observe cows daily as they move through the parlor. In the Herd-i system, an in-parlor camera captures cows as they exit milking, while existing cow identification links observations to individual animals. The system provides automated locomotion and body condition measurements along with video and historical information. Thus creating a way to identify cows whose measurements fall outside parameters established by the farm.
Rather than asking employees to review every cow and every measurement, automated monitoring can generate a watch list of animals that warrant a closer look. That can be particularly useful for subtle changes. An obviously lame cow is likely to attract attention. A cow showing a less noticeable change in locomotion may not.
“The precision that it provides is in terms of ‘let’s give you a shorter list of places for you to go look,’” Mozeliak says. “Nothing will replace actual hands on a cow, but it gives you a place to look.”
The goal is to streamline herd checks and make observations more targeted.
Turning Automated Alerts Into Action
Once a cow appears on a watch list, the farm team still has to determine what the finding means.
Mozeliak describes a workflow that starts with reviewing the cow’s record and associated video. The team can then determine whether the finding warrants immediate attention or should be incorporated into an upcoming visit from the veterinarian, nutritionist or hoof trimmer.
A change in locomotion, for example, may prompt a closer hoof-care assessment. A change in body condition may lead to a discussion about nutrition or other factors affecting the cow. The appropriate response depends on what the team finds when it evaluates the animal.
“It’s more about the data delivery than it is about giving them guidance on what action they should take,” Mozeliak says. “They know what action they should take.”
That creates a more targeted starting point for herd health work. A veterinarian arriving for a herd visit may already have a list of cows that warrant examination, allowing more time to be spent determining what is happening with those animals.
Adapting Monitoring to the Dairy System
The way automated monitoring is used also depends on the production system around it.
Mozeliak says Herd-i’s experience in New Zealand provided a starting point, but the U.S. dairy market presents different operating conditions. New Zealand farms often use pasture-based systems and different milking schedules, while U.S. dairies may have larger herds, freestall systems and more frequent milking.
Those differences can affect how information fits into the farm’s daily workflow. With less downtime between milkings, quickly identifying which cows need attention can become particularly important. A dashboard or watch list that can be reviewed as part of the existing routine may be more useful than adding another process for employees to manage.
For Mozeliak, adapting the technology to the U.S. market is therefore not simply about bringing the same system to a different country. It is about fitting the information into the way each dairy already operates.
Why BCS and Locomotion Can Be More Useful Together
One potential advantage of automated monitoring is the ability to consider multiple observations from the same cow.
Body condition and locomotion provide different information, and Mozeliak says looking at them together may lead to a different conclusion than looking at either measurement alone. She cautions that the relationship should not be overstated. The company has not established how frequently the two changes occur together, and either one could contribute to the other or both could reflect a separate underlying issue.
For the farm team, the value is in having more information to investigate.
A cow showing changes in both locomotion and body condition may give the team more context than a cow showing a change in only one measurement. It gives the team another reason to examine the cow and determine what is happening.
Using Exception Reporting to Reduce Data Overload
Continuous monitoring can create its own problem if every measurement becomes another piece of information employees are expected to review.
“Sometimes less is more, especially when it comes to data,” Mozeliak says.
Exception reporting is one way to make continuous monitoring more manageable. Farms can establish thresholds around the measurements they want to monitor and determine how long a cow must remain outside those parameters before appearing on a watch list. This reduces a large volume of observations to a smaller group that deserves attention.
Those thresholds can also help bring different members of the herd health team into the same conversation. Producers, veterinarians, nutritionists and hoof trimmers can review the same information and determine which cows need to be addressed immediately and which can wait for a scheduled visit.
In that sense, the technology becomes less about generating more data and more about delivering useful information to the person who can act on it.
Where AI Fits Into Herd Health
As artificial intelligence becomes more common in dairy technology, Mozeliak sees a similar division of responsibility.
AI can help collect and analyze information, but producers, veterinarians and other members of the herd health team still need to interpret what they are seeing and decide what to do. That distinction is particularly important when considering predictive technology.
“Earlier information is worth a great deal to a farmer. However, predictive could be dangerous,” Mozeliak says.
A prediction is not the same as an observation. Acting on a problem that has not actually developed could create unnecessary work or intervention. Mozeliak sees more immediate value in identifying changes early enough for the farm team to investigate them.
The technology can help identify where attention is needed. The people caring for the herd decide what that information means.
Uncrowding a Crowded Plate
The broader opportunity for automated monitoring may be less about replacing individual tasks and more about giving farm teams additional capacity. Dairy teams already make frequent decisions under time and labor constraints. Asking the same people to observe every cow, sort through every measurement and determine which changes are meaningful is not always realistic.
“Honestly, I think it’s uncrowding a really crowded plate right now,” Mozeliak says.
Automated monitoring cannot replace the veterinarian walking into the pen, the hoof trimmer examining a foot or the producer knowing what is normal for a particular cow. What it can do is help those people decide where to start. Automated body condition scoring and locomotion monitoring can help turn continuous observation into a more targeted herd-health workflow, putting human time and clinical judgment where they are needed most.
To hear more from Nicole Mozeliak, as well as Dr. Joaquin Azocar, veterinarian with DeLaval, on the benefits of automated body condition and locomotion scoring, tune in to the latest episode of “The Bovine Vet Podcast":
