Having genomically tested youngstock in the last year, the results from a herd genetic report show that good decisions have been made on the farm run by Caryn Webster at Browns Bay, outside Islandmagee in Co Antrim.

Caryn milks 60 pedigree Holstein cows in a medium-input, grass-based system, with the autumn- and winter-calving herd producing just under 7,500l from 2t of concentrate fed.

She is one of 20 dairy farms in NI and 56 across the UK, taking part in a DEFRA-funded Dairy Carbon Network project, led by researchers at the Agri-Food and Biosciences Institute (AFBI).

The aim of that project is to highlight practical actions that improve efficiency and, ultimately, result in lower greenhouse gas emissions on farm.

On the Webster farm, three actions have been chosen relating to improved forage production and utilisation, feed supplements and breeding for improved feed efficiency.

It was this third action which was the main focus at an on-farm event last Thursday.

In 2025, Caryn got calves and year-old heifers genomically tested for the first time.

With a big focus on grazing, she is been keen to breed cows with a smaller stature and had previously tried a few Friesian straws.

However, that wasn’t overly successful and she wants to continue with a Holstein-only herd.

“I am looking for positive components, fertility, maintenance, good teat length, chest width, lower stature.

“It is hard to find a bull with all those things – I did compromise a bit on components,” she said.

Those decisions are reflected in a Herd Genetic Report run by the Agriculture and Horticulture Development Board (AHDB). That report is available to all UK dairy farmers who milk record at least three times per lactation on ICAR-approved milk meters. According AHDB animal genetics manager, Daniel Pitt, the report allows farmers to compare against national averages.

However, the key issue is not whether a herd is top or bottom in various indexes and traits, but instead to use the report to identify strengths and weaknesses in herds.

The results put the milking herd on the Webster farm in the top 55% for Profitable Lifetime Index (£PLI), but youngstock are in the top 45%, so the herd is “heading in the right direction”, said Pitt.

On the AHDB fertility index, the milking herd is in the top 65%, but youngstock are in the top 45%. Other sub-indexes such as HealthyCow and EnviroCow also how positive improvement, although Predicted Transmitting Ability (PTA) for both butterfat and protein has slipped.

Selection

When it comes to selecting which genomically tested heifers to actually breed from, Pitt emphasised that each farm has different needs and different breeding goals, and with around 40 different traits assessed, it can be difficult to digest all the information. He suggested a “good starting point” is to use PLI, but it is still important to look beyond the headline figure, as, for example, you might have an animal with a high PLI, that is below average for fertility.

Using the Webster results, he said an option might be to initially select out heifers with PLI below 150, then look at other traits such as fertility and maintenance.

“You can keep going. The likes of TB advantage could be a final trade off in a borderline case,” he said.

The main argument made by Daniel Pitt last Thursday is that it only takes a minute to breed a cow, but it can take a lifetime to breed out wrong decisions.

His advice to dairy farmers is to use young genomic bulls as a means of accelerating genetic gain in the herd, but still exercise some caution by using a number of bulls, in case the reliability of an animal changes once milking daughters are available.

However, with free genomic testing now available through the Bovine Genetics Project run by Sustainable Ruminant Genetics (SRG) in NI, it is the genomic testing of female breeding stock where real gains can be made.

Reliability is an indicator of how accurate a particular genetic index actually is, and it can range from 20% to 99%, depending on the amount and source of information in the calculation.

The lower the reliability, the more likely it is to change as more data are added.

In non-genomically tested herds, the predicted genetic traits are essentially an average of parents, with analysis by AHDB putting the reliability of these traits for a newborn heifer at 31%, rising to 46% at year one and 50% by year two, once siblings, etc. are in production. However, a son or daughter is not the average of their parents, as different eggs and different sperm come with different genetic abilities.

Genetics

By genomically testing, the actual genetics of the animal is known, resulting in much higher reliabilities of 59% at birth, rising to 65% by year one.

“The reliability is approximately doubled in genomically tested heifers, compared to a parent average proof, allowing for more confident breeding decisions to be made earlier in the animal’s life,” said Pitt.

He used the example of the PTA for milk production (kg) in the genomic results for heifers on the Webster farm. On the parent averages, two heifers had PTA figures of +538 and +481, but after genomic testing, these figures dropped to +86 and +79kg.

“You wouldn’t want to breed from that – that is the power of genomics,” said Pitt.

Other work by AHDB has shown that herds which routinely genomically test their heifers (and use high PLI sires) have an average PLI which is £193 more per cow than a herd not doing any testing. This genetic influence on overall farm margins is estimated at around £20,000 extra per year in a 150-cow herd. Data from earlier this year showed that there are around 400,000 Holstein females in the UK now genomically tested, which equates to about 35% of Holstein females. Just over 30% of all dairy females have been tested.

“We are expecting that figure to shoot up thanks to SRG,” said Pitt.

NI dairy farmers who register for the Bovine Genetics Project and avail of free genomic testing will have access to the various selection indexes produced by the Agriculture and Horticulture Development Board (AHDB).

The main UK index is the Profitable Lifetime Index (PLI) which indicates the additional profit that a daughter of a high PLI bull is expected to earn over her lifetime, compared with one sired by a bull with a PLI of zero.

PLI is designed for year-round calving herds and comprises various traits such as milk production (yield and solids), fertility, feed efficiency and udder health.

There is also the spring calving index (SCI) and autumn calving index (ACI), which tend to put higher weightings onto the likes of milk quality and fertility, helping farmers to select heifers and bulls best suited to these systems.

Below these three main indexes, AHDB has also developed two sub-indexes which can be used to help narrow bull and heifer selection.

EnviroCow puts strong weighting onto milk production, lifespan and feed efficiency. HealthyCow weights fertility the highest and includes a number of health-related traits.

As well as changes to breeding, the two other issues being explored on the Webster farm as part of the Dairy Carbon Network mainly relate to grass utilisation and concentrate feed.

Caryn has been measuring grass weekly since 2023 as part of the AgriSearch GrassCheck programme. Despite a drought in 2025, followed by flooding in late October, the farm grew 9.38t of grass dry matter per hectare (DM/ha) last year. To date, in 2026, 7.73t has been grown.

Cows would normally stay out into November, but had to be housed on 3 October last year.

They went out again on 20 March this year. Some silage has been offered at milking this summer, given restricted growth due to dry conditions.

In 2025, there were 80 cows on the grazing platform, but a TB outbreak resulted in the loss of 19 cows and a heifer, so there are only 60 being milked this year. Caryn hopes to increase to 70 cows this winter, which “is probably a nice number to stay at,” she said.

She also hopes to make some improvements in grazing infrastructure over the winter months, including moving drinkers into the centre of paddocks.

As part of the project, a decision was taken late last winter to switch from a high-fibre 18% nut to something with more soya and wheat, with AcidBuff and Megalac added in.

Butterfat did fall a little, but yield per cow was up from 21.5 to 25l within 10 days.

“The yield alone was worth it. But we were past peak – we will try again earlier this year,” said Caryn.