Carcase feedback and the future of Merino lamb

Will Barton: I really just want to go through who we are, what we do and what we see as our role as a processor in the industry. I'll talk you through our objective measurement journey, which has been really consistent with the theme of today: it's all about building and gathering data and then driving insights, which we're getting much closer to doing now. Then I'll cover what the feedback to our producers has looked like as we've been able to collect this data, which we're now in our fourth year of doing. The first couple of years were all about beating the drum and saying, hey, we're doing this thing differently, and now we're actually seeing some real insights, which is exciting.

Then I want to change gears a little bit and talk about what that means for Merino lamb, as it relates to us in our supply chain and maybe the broader industry. I'm going to do that by talking about the history of Merino in the meat industry and where it's been, then the importance of MSA pathways, which again plays into some of the key themes you've been talking about today, and then what I think the future looks like.

Who Gundagai Meat Processors are

Our family history in meat started in 1919, when my grandfather Fred started his butcher's apprenticeship in Gundagai, which is where we obviously still are. In the 50s he established his own butcher shop after doing 30 years of trading in Gundagai with other brands. Then in the 70s my father Bill and my uncle Tony built Gundagai Meat Processors. It was beef, pork and lamb when it started, and it has been refined over time. From about 2000 to 2020 we were a fee-for-service-only plant, so toll processing, and exclusively for Coles for a long period of time. Then in 2020 we made the call to recommence buying livestock and selling meat, with the launch of Gundagai Lamb. So that's the quick history.

The three main clients and brands in our business now start with Coles, in a fee-for-service capacity. They've got their standard lamb offering as well as Coles Finest, which you might see, and that only comes through our plant and uses our objective measurement technology to grade the product. We also do the processing for what is probably Australia's largest organic red meat business; we're their only processor, so we do all of their processing through our plant. And then of course we've got Gundagai Lamb, which is again graded into two products: a standard product, and GLQ5+, which is our high-end offering that I'll talk through.

This is a snapshot of what a business like ours looks like from a local community point of view. We process a million sheep and lambs a year, which is about 3.2% of the national five-year average kill in Australia. They're obviously not all Gundagai Lamb; all of those other customers are involved in that. It's about two and a half thousand semi-trailers of incoming lambs. We've got 330 employees, and we pay just over $26 million a year in wages and salaries to that workforce, which is all Gundagai based.

The processor as the pinch point

The output is about a thousand semi-trailers of meat, assuming they were all 100% full, which from a freight point of view they unfortunately aren't always, plus the shipping containers that are exported. So that's a broad view of the scale of the business. The story of why we've gone down the path of innovation, technology, data gathering and objective measurement really starts with what our role is in the supply chain. This is a big one for me, because I think we forget sometimes that there are producers, then there's a processor, and then there's a consumer.

Our role is the middleman, getting that product from the farm to the consumer. But there's also this other role of getting information back, which I think we've critically missed for a long period of time, and we're a really narrow pinch point in the industry. These numbers are really bad; I pulled a few data sources together, including the ABS, and they wouldn't pass an audit at all, but by order of magnitude they're broadly right. You've got 31,000 producers of sheep and lambs in Australia.

There are 45 processors that have AUS-MEAT accreditation. I'm not counting really small plants and all that sort of stuff; you might get to 100, or 150, if you counted them all. But there are 31,000 producers who are feeding everything through 45 processors of commercial scale, who are then distributing that product to consumers around the globe. So it's a very, very narrow pinch point that we're coming through. For us there are really two main priorities in this kind of architecture. One is to tell producers what they sent us in as much detail as we can, and I think we do that better than anyone else in the industry.

Measuring every carcase

The second thing is to tell producers what we want them to send to us, and the way that we do that is with price. If you can price what you want to receive, then producers will respond. So, the tech: DEXA. Actually, here's an idea. If we could just do a live DEXA test on all the lambs before they come to us, then producers would know if they're too fat. How long does it take to DEXA a live sheep?

Audience member: Three minutes.

Will Barton: That's not bad, including setup no doubt. So we were the first plant in Australia to have a hot-side DEXA unit. I think there are maybe three now, and there are a few cold DEXA units.

It measures lean meat yield, fat and bone percentages on our slaughter floor, which is essentially at the point of sale. We have a grid which pays more for leaner lambs, up to a point, and then when they become too lean it starts to discount. Another first for us was the MEQ probe, which we developed with MLA and MEQ during COVID. It's now the only device accredited by AUS-MEAT to predict intramuscular fat, and we use it on every carcase. We also check animal health and carcase defects across every carcase we process.

There are 20 issues here, some of which can be managed on farm, and others that can't and are just interesting for people researching, and we collect data on those too. So every lamb that's consigned to Gundagai Lamb gets a report on its weight, its level of lean meat yield, its level of intramuscular fat, and whether or not any of these 20 issues were picked up in that carcase, carcase by carcase. If we think about where we were four years ago, and where most processing plants in Australia are at the moment, this is about 12,000 lambs in those dots.

They're organised with the fattest on the left-hand side, and you can see they're the heaviest as well. I'm game enough to use the pointer; I don't even know if you can see these ones here. Then as you come to the right-hand side, they get leaner, down to a fat score one, and they get lighter, and there are only three fat score ones out of about 12,000. This is what the world looked like four years ago at our plant. Then we put DEXA in, and now they look like that: the same lambs and the same weight distribution, but now we can see in 0.1 increments what their lean meat yield is.

So it's still fatter on the left and leaner on the right, but you go from seeing this much information to that much information. The only downside, because it's all really exciting, is that it is very complicated, and so it creates more and more for you to be thinking about and managing. But the good news is that we do something like an index, which is our GLQ score, which combines everything together to give you a single number. And a lot of the things that we're chasing are actually mutually beneficial, so not everything is a this-or-that kind of equation, which I'll get to.

So with the same lambs, where we used to only be able to see fat score and weight, you now see weight, and we've got a DEXA unit that can tell us lean meat yield. Then we swap to the same data set again, and now with intramuscular fat, so we can see which lambs have got higher and lower intramuscular fat, and which ones are too fat and which are perhaps too light in the middle. If I could see what was coming next, I'd tell you, but I think the next one overlays weight. So if we bring weight back in, using colour, you can see that the dark blue are heavy lambs, over 30 kilos, and they graduate back to lighter lambs.

The most typical scenario is heavier, fatter and more IMF. But you can see that there are heavy lambs in that 25 to 30 kilo category that have got high IMF and high lean meat yield, and that's what we've been working to encourage producers to send more of. It is increasing year on year because of the way that we've structured that supply chain.

Building Gundagai Lamb and the GLQ score

That really becomes the point at which we introduced Gundagai Lamb. Because we've got this ability to connect all of these traits together by measuring them, we built Gundagai Lamb as a lamb brand that is constantly improving through an open feedback loop between producer and processor, with the intent of increasing value for the consumer. So it must be led by whether the consumer is valuing the consistency we're providing, and the intent then is that if you've got extra value for a consumer, there'll be more profit for producer and processor. We say that we're better, because we're IMF grading lambs for guaranteed eating quality. We say that we're cleaner, because we've got lean meat yield grading to reduce wasteful fat production.

We have animal health reporting to reduce disease burden over time. And we say that we're fairer, because we're building trust through transparency and paying producers for quality. That's our understanding of what both producers and consumers value. The really interesting thing about fat today, which I was fascinated by, and which is really important for us when I get to the end and tell you what we're looking for in a Merino lamb, is that fat is really important, particularly as it relates to stress in transit and glycogen reserves, and how Merino lamb performs in our system.

But the thing about lean meat yield is that there's a difference between carrying maintenance fat, or fat buffers, and flat-out over-fattening something to create marbling. Think Wagyu. We don't want to go too far, and it's all about the balance we keep between these traits as we continue to push the envelope and try to produce more of what a consumer wants to eat. That's really where the GLQ score came about. It's designed to encourage lambs with above-average marbling. We're not trying to create the Wagyu of anything; we're just trying to have an above-average eating quality experience consistently.

It's also designed to discourage over-fattening of lambs, and to reward lambs that are free of animal health conditions and carcase defects. There is a fancy algorithm that I haven't put in this deck because it's a bit over the top, but the broad brush of it is that if you've got a 1% IMF animal, it'll never make it to GLQ5+. Regardless of how lean it is, it's just never going to be something that a consumer wants to eat. As your intramuscular fat increases, say to a seven or a nine, which are pretty rare animals at this stage (seven is something we see semi-regularly with better performers, but nines we don't see very often at all), we still don't want to encourage that animal if it's under about 55% lean, because there's just too much outside fat.

So that's the balance we're trying to strike: being lean enough to be able to sell the majority of the carcase that we're buying from you, not trim it on the floor, while we're chasing that IMF. If we come back to the example we had before, it's not anywhere near as linear as this, and the person that developed the algorithm would be upset with me for simplifying it in such a way, but essentially what we're looking for is a carcase that's got... and that's moved a bit, it's not 50% lean, it's a bit further to the right now. But it's leaner and higher in intramuscular fat.

I think the beauty of being able to see the information and track the data over a long period of time is that there's a big chunk of this population that's exactly where we want it. From a producer point of view, we just need to be proactively trying to grow fewer of the ones at the bottom and on the left, and as a processor, we incentivise that by paying more for the top right, which is what we've been doing for the last three and a half years.

Balancing growth, leanness and IMF

Audience member: Could we just explore that for a moment? Is that a combination of traits we need to combine at a genetic level? I'm thinking IMF with growth, potentially, to put that combination together.

Will Barton: If that was the best case, you'd have these measures in your selection. If you want to really know it, you'd have to go directly to the traits. But I think the combination of traits we've been talking about today would naturally land there. Our challenge will be making sure we don't overcook it and go too fat, and whether we handle that with terminals, or by killing younger, or however that system evolves. There's a whole conversation there, which I'll get to, but we can have it now.

The history of most breeds, and what they've focused on for the last 20 or 30 years, really dictates where they all are now. If you think about our plant as one that slaughters 3.2% of the national kill, with about a third of that in our brand, that's about 1% of lambs that are getting this data and feedback. We're doing experiments every day by measuring everything that we process, but the rest of the industry can't see this. You have a challenge in that the two traits are antagonistic in terms of hard selection pressure: the harder you drive IMF, the less likely you are to have great levels of lean meat yield.

But what we see when people come to us for the first time with their animals is that if you're presenting a carcase that's been bred very hard for growth and muscle, so think crossbreds, composites and that sort of stuff, the odds are that you won't perform as well on this as if you're a Merino that hasn't had the same selection pressure. So there's a real pivot point now to say, where to from here? You either pull it into the bucket of things that you're looking at and just make sure you maintain it, and in a lot of senses, if the modern Merino can just hang on to it at current levels and keep progressing everything else, they're in pretty safe territory in my mind.

Whereas the other breeds probably need to say, okay, we've had great success at the growth end and with the carcase, but we've really got to start pushing hard on IMF. So in my mind, for the modern Merino it's really an IMF maintenance piece rather than a pushing piece, because it's inherently there, since it hasn't had those antagonistic traits under really hard focus.

Audience member: So how do we find the balance between leanness and driving back whole-body energy, from a farm perspective?

Will Barton: I specialise in end of life. I said I specialise in end of life, and the management of that process. Smarter people than me can probably answer that question, and I'd be reluctant to. But I think the first step is, every single time you slaughter something, understanding what happened, because it's a little bit like the chemical drum without the labels. You're a bit lost if you don't have that information.

Audience member: What proportion of your lambs are Merino?

Will Barton: Merino? Jake?

Jake: It depends on the time of year. At this time of year it's heavy Merinos, because they're getting to maturity. About 35%.

Will Barton: So 35% Merinos. We'll steer right back into Merino lamb at the end.

Audience member: But in this context, how much of that whole-body energy is in IMF? If you're driving IMF in your selection process, are you driving whole-body fat? Is that the best way to target it, or can you target whole body as well?

Another audience member: I've done the maths on that, and it's not enough. I can't do it in my head that fast, but there might only be half a kilo or something, not even that, in that pool.

Will Barton: I suppose so. And I think the piece for us is that we're still learning. We're still getting shocked by great results and shocked by bad results, and our understanding of the ingredients that drive a great outcome is still evolving. Your neighbour there, Andy, will be able to tell you about that, because he's had those experiences where you think it's going to play out this way and then it plays out that way, or you're cautious and it turns out better. So it's very much a learning process for us.

The reputation of Merino lamb

That ties into how we view Merino lambs. You've got to go back to what the history of Merino meat has been. In my mind there were two broad categories that it used to live in. One was a light, unfinished lamb, sold in a drought cycle where the prospect of keeping it for wool production was lower. The other was mutton, sold at the end of its useful life: in the case of wethers, after wool production, and in the case of ewes, after wool production and breeding. Because of those two categories, the reputation was that they had a secondary eating quality outcome to crossbred lambs, and I'll unpack a little bit why that thought was in place and how we've debunked it.

You had these references to leggy frames with low eye muscle area. When you were talking about the way your father described lambs, my father would describe lambs as silky, and I laugh and still say to Dad, what's a silky lamb? How does a silky lamb eat? Just silky? We've progressed a long way with data and information, so I can relate to that. They were also seen as difficult to process cleanly, which is still a challenge. Merino skins are a bit hard to actually remove during processing. They take a little bit more care, and they tear a little bit more easily.

And they have this reputation of being likely to have a high seed burden, because they're running on the hills and the secondary country for wool. So that's the starting place, if you like, for Merino as we re-entered buying lambs and selling meat in 2020. This is the reputation we were faced with.

Why MSA pathways matter

One of the amazing things that industry has done is the MSA pathways work. How many of you know what MSA pathways are, or have heard of them? Hands right up. If we buy a Merino lamb and it hasn't followed MSA pathways, it will not work. We've proved this to ourselves over and over again, particularly in the last 12 months, as we get better and better at understanding what impacts GLQ scoring and what impacts carcase performance at our plant.

MSA pathways were really about three things: on-farm management, pre-slaughter handling, and processing. Processing was the easiest thing first, because there are only 45 of us, so you can make processors do certain things, and the on-farm stuff is a longer-term thing. For crossbred and composite lambs, it sets a minimum daily weight gain requirement for the two weeks leading up to slaughter, a rising plane of nutrition of 100 grams per day. For Merinos there's a higher requirement: they've got to be doing 150 grams a day.

This is really important. Sucker lambs have to be over 16 kilos and weaned lambs need to be greater than 18 kilos, which is not a big deal for us, because we're targeting a 24 to 32 kilo lamb. And they have to have a minimum fat score of two, which is really about fat coverage. If you go online now to the MLA website and look up MSA pathways, this is published for everyone to know, but no one's really talking about it as much as they should.

If you send lambs to an abattoir and you sign your NVD and tick the MSA box, this is what you're saying is happening. We know from experience that most of the time when we have stuff-ups with GLQ, it's this. Farmers that have been sending us lambs at 30 or 40% GLQ5+, and getting an 80 cent per kilo premium for those animals, will send us a mob and there's 15% GLQ5+. We go, oh, something's wrong with the technology. Then Claire and Jake and I spend all day running around trying to figure it out. We call people at Murdoch University and others that have helped us put the data together: did we calibrate it properly this morning, what's wrong?

Eventually we get to six o'clock at night and go, we're going to have to call this producer and just ask them, did anything go wrong? And every single time they've had a check. Every single time, that rising plane of nutrition has somehow been impacted, whether it's a worm burden or something else that's gone on on farm, and the producer always knows. They always know. So this is really critical, and a focus for us now in educating people as they're sending lambs to us: these MSA pathways are important.

Audience member: Does the IMF drop?

Will Barton: IMF drops, yeah, almost always. And so there's a question there about which fat goes first. Who was the lady, Charlotte, that you spoke to? I reckon I've listened to that eight times, because it's really interesting the way she talks about fat deposits and fat depletion and all those sorts of things. We're learning a lot as we go about how to make sure that pre-slaughter run is as good as it can be.

There's some other stuff which you can look up: a minimum of two weeks off shears prior to slaughter; total time off feed no greater than 48 hours, with access to water whenever they're not on a truck; two weeks on the consigning property before dispatch; and a maximum time in transit of 24 hours. At processing, we've got to make sure we're looking after lairage, so not overcrowding, access to water, and a range of other stresses in terms of the way livestock flows. We've got to manage ultimate pH and the rate of decline, carcase hanging and ageing, QA systems and identification. All of those things have been fairly deeply ingrained in processing for the last 10 to 15 years.

So what does that mean? It means that after 400,000 consumer taste tests, which is what the MSA sheepmeat model is up to now, so 400,000 people who have sat down, eaten something and marked it, there's no difference in eating quality at a consumer level between Merinos and crossbred lambs, if you've followed the MSA pathway. So if we go back to the reputation of a Merino lamb, think about the circumstances in which Merino lambs used to come to market: drought cycles, light and unfinished, and Merinos at the end of their life. You're not really giving them much chance of success, or of following any of those MSA pathways.

There's a higher weight gain threshold for Merinos because they're more susceptible to stress in transport and the processing steps. Because they're more susceptible to stress, they need a bit more glycogen in reserve, as well as sufficient fat cover to safeguard the pH decline as they go through the chilling cycle, to make sure that we hit optimum eating quality outcomes. Someone smarter than me can explain why.

Audience member: I think it's the two fibre types, oxidative and glycolytic muscle fibres. There are more glycolytic muscle fibres in your carcase animals, and oxidative fibres feed off fatty acids, off the fat. I think it's the muscle type difference.

Will Barton: So is that just inherent in them right now?

Audience member: I think so.

What we look for in a Merino lamb

Will Barton: The way we manage that at GMP and Gundagai Lamb when buying a Merino lamb is this. I've yet to see, in my travels globally as I sell our product, any Australian lamb that's distinguished at market as being Merino or non-Merino. So the idea that we would go and discount a Merino lamb because a consumer is somehow going to know is a farce. The only reason the world discounts a Merino lamb (not us, because we don't discount them; in fact we probably pay more on average for a Merino lamb than a crossbred lamb, because of our system) is that they're stressed it's going to do brand damage, or it's not going to eat as well, or they're going to put it into a secondary market or cut rather than into their main product. But it's not distinguished, and so what we're doing is focusing on carcase outcomes.

In addition to the standard terms and conditions we've got for every lamb that comes to us, around wool length and all those other things we require of a normal lamb, for Merinos we say it has to be a modern Merino. Think Mumblebone, because they do perform really well in this space for us: plain bodied, with improved carcase conformation and growth. The opposite of that leggy, low eye muscle area kind of carcase (I don't want to use the word leggy, but I'll use it again) works really well. We tell our producers we're targeting 55 to 68 kilos on farm.

Our experience is that if they yield about 48%, they're going to dress 26 to 32 kilos, and that's our sweet spot in our grid. Now, if a producer says they might never get to 68 kilos, which is often the case, we would say it just has to be a minimum of 55. Our experience is that if it's less than 55, it's probably not cruising in terms of fat and growth, and if it sits at a 22 to 24 kilo carcase weight, it's going to end up not having the other traits that we're rewarding, like intramuscular fat. It's got to be really well finished, with good fat cover.

We don't really deal in fat scores, because we've got something that's vastly superior to that, which is a DEXA machine. It cost us buckets of money. But on farm, failing having your own DEXA machine, think fat score four. What that means for us is that if it's got really good cover, it's more likely to have intramuscular fat, it's more likely to be where we want it in terms of growth trajectory, and we know it's not going to cut dark or any of those other things that can happen if you get a hit to that glycogen reserve in transport.

The other really important thing is that we really need less than 50 mil of wool length, and we'd probably tell you less than 30 mil, because they can be problematic to process with long wool. They need to be properly crutched and presented in a clean fashion. It has a huge impact on our shelf life when lambs arrive that are not presented well.

What we're measuring next

This is a bit of an overview of what we've already done and what we're looking to do next. Hot standard carcase weight we've had for about 50 years under AUS-MEAT. We introduced lean meat yield with DEXA, which we give to producers. We've got IMF, which we're tracking. We've got a traceability app, so our consumers globally can scan a QR code and know how many lambs graded GLQ5+ that day and the nearest town they came from. So if we buy lambs from, say, here and Cootamundra and Gundagai, there'll be a little map that says on this day the lambs came from these towns.

And there's the animal health feedback we went through. We've also been trialling an eye muscle area camera for the last six months, so before the end of the year we'll be in a position where we're capturing eye muscle area data on about 20% of all lambs. Because we won't do it for every single lamb, we're a bit reticent to give the feedback lamb by lamb, but we might provide a benchmarked eye muscle area to say, this is where you sit percentile-wise with your eye muscle area.

That's a really important one for us, particularly as it relates to Merino, because one of the high risks for the Merino is that it won't have a great eye muscle area. Then there are the things we're still playing with, which are the next frontier. I'm really interested in the connection between soil health, plant health and animal health, and in turn what the impact of that is on human health. From a consumer point of view, we've spent a lot of time, and this was a bit of what Russ touched on, saying from a narrative point of view: here, eat this lamb, it's not bad for you, or it's not treated badly.

But what if we could say, eat this lamb proactively, this will be better for you, you can actually use it to improve your health because these are its attributes? For me that's about measuring those outcomes on plant, and there's some evidence to suggest that at a biological level we might be able to see some of those things in a carcase. Maturity is another one I'm really interested in, because now that we've got this really good measure of eating quality, the cut-off point becomes blurred.

If we turn our minds back to the yearling lamb discussion (which is way nicer than hogget; whoever named lambs that get too old hoggets should be shot from a marketing point of view), what opportunity exists there, and how can we do something better than teeth, in an objective measurement sense, inside the carcase, something like ossification? Then there's shear force. We know that high IMF doesn't necessarily mean low shear force; there are animals that have high IMF that also carry a higher shear force. There's a really good correlation between high IMF and low shear force, but we can't 100% trust it, so we'd love to have a measure of shear force on plant as well.

The other one that's got me a bit interested at the moment is profit per unit of CO2 emitted. Again, I think this is one that plays well for Merinos, in the sense that you're getting wool and meat, and if the value of that wool and that meat is recognised and set at an appropriate level, I think it'll be hard to beat a Merino for profitability in terms of carbon emitted.

That's a bit of where I've gone here. If you think about past, present and future, you've got a carcase which was considered in the past to be leggy, with low eye muscle area and perhaps underweight. Even if it was at 22 kilos and presented like a trade lamb, it probably wasn't being presented with a massive glycogen buffer, so it probably did tend to dark cut more often than not, and it probably did tend to get bad feedback from a foodie more often than not. So it got that reputation for eating quality, which I've just touched on.

Then there's value. You've got high-value wool and a high-value skin, but low-value meat at the moment. We're improving those traits with seedstock producers like Mumblebone; you're getting better outcomes in those top two with some modern Merinos, and you've got improving meat value. Where I would say Merino ends up is that you've got this constant feedback coming back to you from businesses like ours, helping you breed better animals through time, and you end up with high-value wool, a high-value skin and high-value meat, because we know that it can perform in all three buckets. So, ta-da. Questions, if you've got them.

Questions from the audience

Audience member: How many producers have you seen working with you and really adopting this and getting excited about it?

Will Barton: Because we started from scratch in 2020, we didn't have to change anyone. It wasn't as if we'd decided to take, say, a stud down a different track and were at risk of losing a heap of loyal customers. We just didn't have any farmers, so we were starting from scratch, and that was an advantage in being able to say we're going to launch a value-based marketing grid, we're going to pay on lean meat yield, and we're going to give you a bonus when there's intramuscular fat present while it's not over-fattened. Most of the people that have come on board are forward-thinking believers who are just excited to have the data and be part of something that's different.

So there's certainly that underlying it. Probably our biggest challenge is that we set our grid pricing about 20 cents under market, and the reason we do that is that if somebody sends us lambs and there's no GLQ score bonus, we don't want them incentivised to come back next week. There has to be some form of disincentive. So the base grid is 20 cents under market, and for every lamb that's GLQ5+ and up, there's 80 cents. For a producer to come on board and get back the 20 cents they're under market, they need 25% of their lambs to make the GLQ5+ category, and then they're square to market. There's risk associated with that, so as you're all managing costs and all that sort of stuff, you have to have a fairly strong sense that this is where the industry is, or should be, going.

You also have to have a sense that we're not 16 years ahead of the market, like Coke was with Zero, and not everybody shares that view. So for some of the people that have taken the leap, they've gone, never again, I've got no IMF in my lambs, that's a dumb idea, we're not doing that again. And to be honest, 99% of the supply chain can't see it and doesn't care, so there's probably logic to that. But the people that have found they've got a great article stick to us like glue.

If they get 60% of their lambs to make it, then they're 20 cents under market on the base and they're taking home a 50 cent per kilo premium. The best mob ever has been 90 or 95% of the lambs making GLQ5+. The challenge for us has been trying to help and educate producers as we progress, because often they ask why that consignment was so good and we don't have all the answers. So you've got to be of an inquiring mind, if you like, to go, well, it could have been this or it could have been that. We'll work with you and form a view, because that's where we're all going.

Yeah, I think industry has got really good at saying we introduced MSA pathways in 2007 and everyone follows them, but the lived experience is that almost nobody knows what the actual dot points in the book are. Getting MSA accreditation is probably 15 minutes online with your PIC and your LPA number: yep, I do all those things, and then you're there. Now that we're actually seeing that all of the research, decades of research, is accurate, because we can prove that it is, it's become really important to us. If you go to another abattoir, they won't realise the importance of it, because they can't see it playing out in results.

Audience member: Will, have you seen any difference between grass and grain finishing from the people that are sending you lambs?

Will Barton: It's more about net nutrition, I would say. Again, it's scary talking about this in front of people like the people in this room, because end of life is my speciality. But what we've found is that as long as the animal's nutritional needs are being met, it performs well. A lot of people say, well, how do you get a Merino out to 55 to 68 kilos when it's approaching 10, 11 or 12 months of age and is typically getting to maturity?

A lot of them have access to grain, but a lot of them, season depending, have also got cracking grazing on canola or things like that. So there are pasture-based systems that will deliver a result, seasonally dependent, and people are still experimenting. We don't have a great data set that really tells us. We have been collecting finishing nutrition for about 12 months, so we can understand what they're on, but from a scientific point of view you'd really want an energy analysis of what they're consuming, which is not something that people have.

I think the answer to that is whatever gets them to put on 150-plus grams a day for the two weeks before they come to slaughter, and to over 55 kilos with a fat score of four. That's the answer to that question. Some people will be able to do it out of the ground; others will need grain.

Audience member: If you could measure maturity, as you call it, do you still see that as an objective measure with a cut-off point, or some sort of continuum?

Will Barton: Probably a continuum. It's difficult, though, because again, people here know more about this than I do. But typically speaking, if I go back to my dad's logic (and sorry, he's a very forward-thinking, highly evolved human being, I don't want to paint him as anything else), the spring lamb being held out as the ultimate article: I still agree with that, in the absence of being able to see IMF. If you gave me a choice between eating a spring lamb and an old season lamb without any insight into what the IMF level was, I'd eat the spring lamb. But if you gave me the choice between an old season lamb with high IMF and a spring lamb with no IMF, I'd take the old season lamb with intramuscular fat.

I think the reason that exists in the spring lamb, the immature animal, is the way its muscle fibres have developed. So on that continuum of age, you probably need more and more IMF to overcome the maturity of the muscle, simplistically, and we don't really know what that looks like. We've started collecting age data as well, in months of age, and it's usually in bands like seven to eight or eight to nine months, but we don't have precise data on age.

Audience member: Are you going to introduce forward contracts?

Will Barton: We've been trialling them. We've still got spots in our August forward contract; I think it's $7.40, if you're interested. Speaking now, compared to this time last year when the lamb price was $3.85 a kilo less, it really does feel like just deserts for producers, because the pain you were feeling this time last year, we are very much feeling right now. We've been trialling select forward contracts because one of the other things we recognise is that once producers who want to try us have got over the "am I willing to take a punt on this 20 cent equation" question, most of them are really well organised and are trying to book lambs in advance.

To date we've been running a booking system where we'll book you in, say, in October, but you won't know the price until close to the time. For us, we don't have a balance sheet that supports taking a punt on a future price 90 days from now. We just don't have that grunt as a business; it's very much a startup in that sense. So we've tried a few with producers that we know perform well in GLQ5+, so that we can keep customers lined up for the end product, but it is something we'll do more and more of into the future.

The volatility that you're trying to manage is just as real for us, and it's playing out larger than life at the moment, which I think is a bit scary for the industry generally. To go from where we were at the low of September last year to a heavy lamb indicator today of $9.50 is just rubbish for everyone. It got way too low and now it's way too high. Anyway, that's a whole other conversation, probably over a beer.

Audience member: So at the moment, are you just doing a spot rate, or how far out do you go?

Will Barton: We publish a grid on a Monday for the Monday fortnight after. The longest out we've gone is three months: we did have a May, June, July, and we do have an August forward, which was really a first. It's not on the website; we just sent it to a group of producers to see how we think this will work. We've got to look at risk metrics to see how far out of whack we're going to be and all that sort of stuff.

Audience member: In the broad picture, what's the scope for the industry to move along this way? At 1%, you're probably not influencing the other 99%. Do you think you'll drag them along?

Will Barton: Look, I'll do my best. I'll do my best to tease out all the processors at LambEx, because I think one of the big things about the carcase trial is that they've done that trial on 28 producers. When are you going to publish this video? After LambEx? So 28 producers have sent lambs in for a trial, and I don't know how many lambs they've sent in each, 25, 50, 100, whatever it is. We've basically been doing that live trial every day for three years on plant, and we have really big-scale data to support it. We will keep growing our brand.

In answer to your question, though, it's a pretty rubbish decision that a producer has to make, to send a lamb to us to get feedback on their lambs. Really what should be happening is that they should be selecting Gundagai Meat Processors on the basis of how Jake and Claire and the people on our team treat them and work with them, not the fact that they can't get data anywhere else, and not a lot of people do it just for that reason. The ideal scenario is that everyone goes to every abattoir and they all get this information.

We'd drop the bottom 30%. We're already, I think, light years in front: the New Zealand IMF average is thought to be about 1.8% and the Australian was 4.2%. If everyone in industry did this tomorrow, my job selling lamb into restaurants globally would be so much easier, and we'd actually start to compete with bluefin tuna rather than being in with everybody else trying to flog lamb for the lowest common denominator. I'm absolutely gobsmacked that nobody else is doing it. I can't believe it. I've been looking over my shoulder for three years thinking, when is one of the big houses going to pop up?

But I think it's that change piece. Some of these guys are sourcing millions of lambs, and if they go to their suppliers and say, okay, we're doing it this way now, and half their lambs go straight to the next biggest house, then they've lost a big whack of their supply chain. So I think it's gently, gently, but hopefully not the Coca-Cola Zero timeframe. Jake did threaten to leave before, when I said it might take us another 15 years to get critical mass. He was upset with that.

Audience member: What about the Coles contracts? Are they saying they'd better get on board?

Will Barton: First of all, I wouldn't speak to what they're doing, but I'd be surprised if people like that aren't looking. Definitely. For the seedstock sector and the commercial producer sector, it's a bit of a hamstrung scenario not to be getting the data when it could be provided. But it's not free. It costs money for us to collect the information and to house it, with data warehousing, data cleansing and people looking at it to make sure it's accurate. So it's not a lack of willingness, as if it were free and easy. A DEXA unit is about $2 million, and the intramuscular fat probe has got an ongoing subscription in the hundreds of thousands. I don't think we should assume it's easy; the commercial reality for a lot of people just isn't as obvious yet.

Audience member: One of the biggest limitations of getting the data, from a seedstock perspective, is that you have to kill the animal. So where can you see, in the future, being able to link the data you're collecting back into the system?

Will Barton: We're not going to live DEXA animals. The things that really excite me are identifying dams to lambs, which is the tech that neXtgen Agri has been working on. It's still evolving, but in another year or two we'll be able to do it seamlessly. We do it now for trials, but really the missing link is who was Mum. We've toyed with the idea of getting ram team averages with consignments, so that we understand what the ram team average was for that group, and there are a lot of different ideas like that we kick around.

The other thing is the live device that MEQ developed in beef. The challenge for them is that they've got to have the device at the same temperature as the animal, and the oil at the same temperature as both, and wool's a problem. So then I say, well, maybe you could do all your ewes straight off shears, while you had them in for shearing, and just identify the bottom 10% of ewes for IMF. That's commercially, not at seedstock level, so you can just say that's one of our drafting criteria: we draft for age and structure and intramuscular fat performance, or something like that.

So there's a lot of thinking going on around it, but there's nothing really obvious. I think the best possible outcome is connecting the ewe to the lamb. Dad's not there, so you can't get him to hang out with all of the lambs he's parented, but having the ram team average and the ewe would be powerful. And if you think about the rate of change in artificial intelligence, machine learning and processing power generally, something we think is going to be 10 years away probably ends up being two years away.