Robot start-up
"Nine months in and we're already planning to expand."
The farm: Over 120 heifers. 2 robots. New start-up.
Where we started
A robot start-up. Retrofit shed, new robots, new herd, and no history to work from. No old records, no "this is how we've always done it". Every part of the system had to be set up from day one: the ration, the robot feed tables, fresh cow protocols and routines.
Fresh-calved heifers going straight onto robots isn't as hard as you might think. They've never seen a parlour, so they've got no bad habits to unlearn and they take to the box fast. The challenge was making sure everything around them was right, so they could milk to their potential from the start. A start-up has to earn its keep quickly.
Why it matters
A robot herd is only as good as the system behind it. Get the diet at the feed fence, the robot feed and the robot settings working together and cows visit freely, milk well and look after themselves. Get it wrong and you're fetching cows, chucking cake at cows and chasing problems.
First-calvers also need feeding differently from cows. They're still growing while they milk. They peak lower but they should hold that peak well. Feed them for that and they'll pay you back across this lactation and the next. Push them too hard for early litres and you'll lose it in fertility and condition.
What we did
The first job was getting everyone round the same table (farm team, vets and AMS team) and making sure we were all on the same page for the journey. On a start-up the nutrition, the robot settings and the health protocols all affect each other, so it made sense to build them together rather than one at a time.
The ration was built around the robots from day one. The farm's leadership team set the target of using as much forage and home-grown feed as possible, with a low feed rate to drive margin. My target was a maximum of 6kg through the robot, which is unusually low for a free access system.
Robot data drove the decisions from the start. Visits, refusals, rumination and yields are reviewed weekly, and the ration and settings were adjusted as the herd settled.
The result
The heifers took to the robots and the start-up went smoothly. The first 73 heifers calved in November and December 2025, achieving 3.4 visits and 36.4kg of milk off 5.6kg of compound through the robot. The next 50 heifers calved in March 2026 and achieved similar results.
Since the start-up we have introduced Grazeway gates and trialled an A/B grazing system, with plans to move to A/B/C grazing in 2027. Fertility and cow health have been excellent, with the majority back in calf to first service. After a positive first nine months, the farm has the confidence to expand in 2027.
If you're in the same position
- Get your vet, AMS team and nutritionist talking before the robots go in, not after the first problem
- Build the PMR for the robot. The gap between barrier and box is what drives visits at start-up
- Watch the fetch list from day one. It's your early warning
- Make sure fresh heifers visit from the start. Early visiting behaviour is how they carry on
- Don't judge the first month too harshly. Judge the trend
In their words
"Starting robots with a shed full of heifers was a big step. Dylan knows robots and he knows nutrition. What made the difference was how he worked with our vet and the AMS team. No egos. Everyone pulling the same way. The switch went smoothly, and after nine months we're confident enough to expand in 2027."
Farm owner, Pembrokeshire