Current Progressive Dairy digital edition

Managing fresh cows in a robot barn

Laura Flory for Progressive Dairyman Published on 21 January 2016

Oh, fresh cows, why are you so hard to figure out? One of the things we have constantly battled is learning the best way to use all of the numbers from the robotic system to manage fresh cows optimally.

The biggest trouble we have had is learning what is normal. In our previous facility, we had zero information regarding cow health unless we collected it ourselves. If a cow happened to be fresh around test day, we got fat and protein, but that was about it. Now we have so much information; we are striving to make new parameters for points of intervention and are always looking for trends that help us make decisions. So here is what I have seen in our experience.



Rumination seems to be the driver of everything. If a cow’s rumination goes any way but straight up in the first seven days, she has a major problem. In the same token, appearance isn’t everything. There are numerous cows that I have judged as “fine” upon visual observation that tell a different story on paper, and if I don’t act until they look sick, they are going to be really sick. I learned this pretty quickly, thank goodness. The best thing about this parameter is you can know for sure if a cow is responding to a treatment or not to know the best way to move forward. This has saved endless hours of debating about what to do next. A regular day truly does involve making a list of which cows are on treatment and if their rumination is going up or down.

Secondly, trust the fat and protein. We now have data on every cow every day for fat and protein. If the fat is going up and she’s 14 days in milk, she probably has ketosis. I finally learned to stop wondering and just go check her the first day I see the trend. This one surprisingly does not seem to have a tremendous amount of variability from day to day unless the cow has an issue. If I check her and she’s fine, I’m not out much time, but if I miss it for two days, I probably just cost the cow, and myself, a lot of trouble.

As a whole, our incidence of ketosis is almost unchanged, where we did expect it to go down. In my opinion, this is due to two things we didn’t plan well enough for. One being that finding the cows is now 10 times easier, therefore making us more likely to treat it for the problem it really has, And second being the fact that our production has gone up. If they milk more, they are more likely to get ketosis. I guess you can’t have your cake and eat it too. We are still hoping there is room for improvement in the future as we learn how to manage the animals better in this system.

The third key indicator for me is how often the cow is visiting the robot. This is only possible to monitor on cows that have already been through a lactation in the robot barn, but those are usually the culprits with fresh cow problems anyway. If a cow is at 10 days in milk and has been averaging five milkings per day and you see that trend down to 2.5 or 3, there is most likely something brewing.

On the flip side of the coin, if you see every fresh cow start to milk four times a day and have 30 refusals – aka times she comes to milk too early – there may be a nutritional problem. We struggled early on with ketosis around day eight or nine, and when we went to look for the problem, we realized that almost classically on days five through six those cows would have a ton of refusals. After some head scratching, we realized they were simply hungry and we did not have our feed table adjusted appropriately.


Here’s to a bright future of fresh cow health and continued learning!  PD

Laura Flory
  • Laura Flory

  • Dairy Producer
  • Hillside Farm