WEBVTT 00:00:10.830 --> 00:00:13.570 Hello and welcome to the Bullvine Podcast. I'm 00:00:13.570 --> 00:00:15.890 your host, Bella. We're so excited you've joined 00:00:15.890 --> 00:00:18.449 us today for what I think is going to be a game 00:00:18.449 --> 00:00:20.949 -changing conversation for many of our listeners. 00:00:21.230 --> 00:00:24.449 And I'm Douglas, joining Bella today to dive 00:00:24.449 --> 00:00:26.690 into something that's not just changing the future 00:00:26.690 --> 00:00:30.210 of dairy farming. It's already transforming operations 00:00:30.210 --> 00:00:33.789 right now. We're talking about how machine learning 00:00:33.789 --> 00:00:36.689 and artificial intelligence are revolutionizing 00:00:36.689 --> 00:00:38.950 reproductive management on dairy farms across 00:00:38.950 --> 00:00:42.310 the country. Douglas, I know many of our listeners 00:00:42.310 --> 00:00:44.670 might be thinking, oh great, more tech talk that 00:00:44.670 --> 00:00:48.390 won't apply to my farm for years. But that's 00:00:48.390 --> 00:00:51.869 not the case here, is it? Not at all, Bella. 00:00:52.549 --> 00:00:56.060 Look, I'm going to cut right to the chase. If 00:00:56.060 --> 00:00:58.320 any listeners are still relying on visual heat 00:00:58.320 --> 00:01:01.579 detection alone in 2025, they're leaving serious 00:01:01.579 --> 00:01:06.400 money on the table. The numbers don't lie. Automated 00:01:06.400 --> 00:01:08.920 monitoring systems powered by back -propagation 00:01:08.920 --> 00:01:11.719 neural networks are delivering 21 -day pregnancy 00:01:11.719 --> 00:01:15.500 rates above 30 % in progressive herds while slashing 00:01:15.500 --> 00:01:20.040 hormone use by 75%. Those numbers are remarkable. 00:01:20.599 --> 00:01:23.500 Today we'll break down exactly how these systems 00:01:23.500 --> 00:01:26.799 work. the economic impact they're having, and 00:01:26.799 --> 00:01:29.939 most importantly, practical steps for implementation 00:01:29.939 --> 00:01:33.239 on your farm. So grab your coffee and settle 00:01:33.239 --> 00:01:35.280 in as we explore this reproductive revolution 00:01:35.280 --> 00:01:37.959 that's changing the economics of dairy farming. 00:01:40.299 --> 00:01:42.760 Douglas, before we dive into the new technology, 00:01:43.019 --> 00:01:45.859 let's talk about where most farms are today with 00:01:45.859 --> 00:01:48.299 reproductive management. What's the current landscape 00:01:48.299 --> 00:01:51.260 look like? It honestly drives me crazy, Bella. 00:01:51.950 --> 00:01:55.370 National surveys show that 51 % of dairy farms, 00:01:55.569 --> 00:01:58.489 that's more than half, still rely primarily on 00:01:58.489 --> 00:02:01.810 visual observation for heat detection. Despite 00:02:01.810 --> 00:02:04.430 overwhelming evidence, visual observation misses 00:02:04.430 --> 00:02:11.449 more than half of all standing heats. Think about 00:02:11.449 --> 00:02:14.270 it this way. Would you accept a milking system 00:02:14.270 --> 00:02:17.330 that leaves half your milk in the cow? Of course 00:02:17.330 --> 00:02:20.539 not. Yet when it comes to reproduction, we're 00:02:20.539 --> 00:02:23.219 surprisingly willing to accept massive inefficiency. 00:02:23.840 --> 00:02:26.979 I imagine most farmers understand reproduction 00:02:26.979 --> 00:02:30.319 is important, but maybe not exactly how important. 00:02:30.900 --> 00:02:33.159 Ricardo Chabel from the University of Florida 00:02:33.159 --> 00:02:36.699 puts it plainly. He says, Reproductive efficiency 00:02:36.699 --> 00:02:41.280 is a key driver on the economics of a farm. Now 00:02:41.280 --> 00:02:43.759 that might sound like stating the obvious, but 00:02:43.759 --> 00:02:46.479 here's what most people miss. Poor reproductive 00:02:46.479 --> 00:02:49.259 performance creates this nasty ripple effect 00:02:49.259 --> 00:02:52.379 through your entire operation. What kind of ripple 00:02:52.379 --> 00:02:56.120 effects are we talking about? It's not just about 00:02:56.120 --> 00:02:59.199 pregnancy rates. It affects lactation persistence, 00:02:59.759 --> 00:03:02.939 peak milk in the next lactation, lifetime production, 00:03:03.340 --> 00:03:06.719 replacement decisions. The whole economic picture 00:03:06.719 --> 00:03:09.719 gets warped. And most dairy accounting systems 00:03:09.719 --> 00:03:12.599 don't capture these costs because they don't 00:03:12.599 --> 00:03:15.509 connect production, replacement, and genetic 00:03:15.509 --> 00:03:18.610 opportunity costs. So farms might not even realize 00:03:18.610 --> 00:03:20.969 the full impact of their reproductive performance. 00:03:22.530 --> 00:03:25.770 Exactly. The impact of suboptimal reproductive 00:03:25.770 --> 00:03:28.930 performance is probably 30 to 50 percent higher 00:03:28.930 --> 00:03:33.210 than most farmers currently estimate. Let's talk 00:03:33.210 --> 00:03:36.129 dollars and cents, Douglas. How does this actually 00:03:36.129 --> 00:03:38.270 translate to the bottom line for a typical dairy 00:03:38.270 --> 00:03:41.550 operation? Want some numbers that'll make your 00:03:41.550 --> 00:03:44.819 coffee taste bitter? For a 500 -cow operation, 00:03:45.219 --> 00:03:47.939 each additional day of average days open costs 00:03:47.939 --> 00:03:52.719 you about $2 ,500 in lost profit. $2 ,500 per 00:03:52.719 --> 00:03:57.159 day. Per day, Bella. If your days open are pushing 00:03:57.159 --> 00:04:00.740 140 -plus days, and let's be honest, many herds 00:04:00.740 --> 00:04:04.240 are, you're talking about over $100 ,000 annually 00:04:04.240 --> 00:04:06.819 compared to herds hitting 110 -day averages. 00:04:07.099 --> 00:04:09.620 And that's not even counting increased culling, 00:04:09.659 --> 00:04:13.139 replacement costs. and suboptimal genetic advancement. 00:04:13.419 --> 00:04:16.240 When you put it that way, investing in better 00:04:16.240 --> 00:04:18.639 reproductive technology isn't really an expense. 00:04:19.160 --> 00:04:23.319 It's stopping a massive ongoing loss. Precisely. 00:04:23.420 --> 00:04:26.060 And here's a simple calculation any farmer can 00:04:26.060 --> 00:04:29.339 do right now to estimate their losses. Multiply 00:04:29.339 --> 00:04:33.420 your average days open beyond 110 by $5 per cow 00:04:33.420 --> 00:04:36.180 per day. That's the minimum annual profit you're 00:04:36.180 --> 00:04:39.540 leaving on the table. So let's talk about what's 00:04:39.540 --> 00:04:43.230 changed. We've had activity monitors and pedometers 00:04:43.230 --> 00:04:46.329 for years. How is this machine learning technology 00:04:46.329 --> 00:04:49.550 different? Remember when activity monitors first 00:04:49.550 --> 00:04:52.170 came out, those glorified pedometers that counted 00:04:52.170 --> 00:04:56.430 steps? That's ancient history now. Today's systems 00:04:56.430 --> 00:04:58.230 use sophisticated machine learning algorithms 00:04:58.230 --> 00:05:00.750 that transform behavioral data into insights 00:05:00.750 --> 00:05:04.009 that were unimaginable even five years ago. What 00:05:04.009 --> 00:05:06.769 kind of algorithms are we talking about? I know 00:05:06.769 --> 00:05:08.689 our listeners would appreciate understanding. 00:05:09.500 --> 00:05:13.639 what makes some systems better than others. The 00:05:13.639 --> 00:05:16.639 technical differences matter significantly. Get 00:05:16.639 --> 00:05:19.819 this. Algorithm performance metrics range from 00:05:19.819 --> 00:05:26.740 73 .3 % to 99 .4 % for sensitivity, 50 % to 85 00:05:26.740 --> 00:05:33.339 .7 % for specificity, and 72 .7 % to 95 .4 % 00:05:33.339 --> 00:05:36.139 for accuracy. The back -propagation neural network 00:05:36.139 --> 00:05:39.399 algorithm with a 0 .5 -hour time window consistently 00:05:39.399 --> 00:05:42.319 outperforms everything else for predicting estrus 00:05:42.319 --> 00:05:46.139 in dairy cows. So basically some systems are 00:05:46.139 --> 00:05:49.160 much better than others at correctly identifying 00:05:49.160 --> 00:05:53.779 when cows are in heat. Exactly. And what makes 00:05:53.779 --> 00:05:56.060 cutting -edge monitoring systems so powerful 00:05:56.060 --> 00:05:59.120 is their comprehensive data integration. They're 00:05:59.120 --> 00:06:01.699 tracking 12 distinct behavioral parameters simultaneously. 00:06:03.129 --> 00:06:05.810 how long cows stand, lie, walk, feed, and drink, 00:06:05.930 --> 00:06:08.550 how often they switch between activities, step 00:06:08.550 --> 00:06:11.370 counts, displacement, velocity, and frequencies 00:06:11.370 --> 00:06:14.870 of various behaviors. When you run all that through 00:06:14.870 --> 00:06:17.389 advanced machine learning algorithms, you get 00:06:17.389 --> 00:06:20.069 reproductive patterns that even your most experienced 00:06:20.069 --> 00:06:23.029 herdsperson couldn't detect with 24 -7 observation. 00:06:24.569 --> 00:06:27.050 It's basically like having a team of experts 00:06:27.050 --> 00:06:30.759 watching every cow. every minute of every day, 00:06:30.980 --> 00:06:33.439 but with even more precision than human eyes 00:06:33.439 --> 00:06:37.000 could provide. And they never need sleep, never 00:06:37.000 --> 00:06:40.480 get distracted, and never have a bad day. I've 00:06:40.480 --> 00:06:43.279 heard that high -producing cows can be particularly 00:06:43.279 --> 00:06:45.579 challenging when it comes to heat detection. 00:06:46.060 --> 00:06:49.540 Is that true? And does this technology help with 00:06:49.540 --> 00:06:53.079 that specific problem? Bella, you've touched 00:06:53.079 --> 00:06:55.750 on something critically important. Have you noticed 00:06:55.750 --> 00:06:57.970 your highest -producing cows are getting harder 00:06:57.970 --> 00:07:01.490 and harder to catch in heat? It's not your imagination. 00:07:01.870 --> 00:07:06.509 It's biology working against you. Really? What's 00:07:06.509 --> 00:07:08.670 happening there? Chabelle's research clearly 00:07:08.670 --> 00:07:11.170 shows that production levels dramatically affect 00:07:11.170 --> 00:07:14.509 estrus expression. When a cow has low milk production, 00:07:14.810 --> 00:07:17.670 say below 70 pounds per day, the probability 00:07:17.670 --> 00:07:21.709 of detecting estrus ranges from 70 % to 100%. 00:07:22.540 --> 00:07:25.519 For moderate producers between 70 and 90 pounds 00:07:25.519 --> 00:07:30.420 daily, that drops to 50 to 75%. High producers 00:07:30.420 --> 00:07:36.019 between 90 and 110 pounds, only 35 to 60 % detectability. 00:07:36.420 --> 00:07:40.040 And for your elite cows producing over 110 pounds 00:07:40.040 --> 00:07:43.620 daily, detection probability plummets to just 00:07:43.620 --> 00:07:47.459 20 to 40%. So traditional methods are least effective, 00:07:47.779 --> 00:07:50.360 precisely where they need to be most effective. 00:07:50.889 --> 00:07:55.350 with your best cows. You've got it. And this 00:07:55.350 --> 00:07:57.689 creates a real challenge for traditional fixed 00:07:57.689 --> 00:08:01.629 -time AI protocols too. They treat all cows identically 00:08:01.629 --> 00:08:03.730 despite dramatic differences in reproductive 00:08:03.730 --> 00:08:07.569 physiology and behavior. Douglas, it sounds like 00:08:07.569 --> 00:08:09.730 these systems don't just detect heat better. 00:08:09.949 --> 00:08:12.629 They might actually change our whole approach 00:08:12.629 --> 00:08:16.850 to reproductive management. Is that right? Can 00:08:16.850 --> 00:08:19.569 I rant for a minute? The dairy industry's one 00:08:19.569 --> 00:08:21.990 -size -fits -all approach to reproductive management 00:08:21.990 --> 00:08:26.050 is wasting millions on unnecessary hormonal interventions. 00:08:27.069 --> 00:08:29.829 We're stuck in this weird time warp where we 00:08:29.829 --> 00:08:32.230 acknowledge that cows are individuals for milk 00:08:32.230 --> 00:08:35.330 production, health, and nutrition, but then we 00:08:35.330 --> 00:08:38.710 treat them identically for reproduction. Go on. 00:08:39.269 --> 00:08:41.730 Why are we still treating high fertility cows 00:08:41.730 --> 00:08:44.029 the same as their struggling herd mates when 00:08:44.029 --> 00:08:47.460 we have the technology to tell them apart? Automated 00:08:47.460 --> 00:08:50.299 monitoring enables a fundamental shift from blanket 00:08:50.299 --> 00:08:53.139 protocols to targeted reproductive management. 00:08:54.080 --> 00:08:57.179 Instead of treating every cow the same, you use 00:08:57.179 --> 00:08:59.620 individual cow data to determine the optimal 00:08:59.620 --> 00:09:03.440 protocol for each animal. What kind of results 00:09:03.440 --> 00:09:06.100 are farms seeing with this targeted approach? 00:09:06.659 --> 00:09:09.360 The results are excellent. In cows with intense 00:09:09.360 --> 00:09:12.000 estrus, researchers reduced hormone injections 00:09:12.000 --> 00:09:16.220 from 9 to about 2 per cow. a 78 % reduction. 00:09:16.840 --> 00:09:19.419 Beyond the obvious cost savings, this approach 00:09:19.419 --> 00:09:22.240 addresses growing consumer concerns about pharmaceutical 00:09:22.240 --> 00:09:26.360 use in agriculture. Beyond hormone costs, what 00:09:26.360 --> 00:09:28.799 other economic benefits come from this targeted 00:09:28.799 --> 00:09:32.360 approach? You're also reducing labor for treatments, 00:09:32.720 --> 00:09:36.139 decreasing stress on animals from fewer handlings, 00:09:36.299 --> 00:09:39.080 and identifying problem breeders earlier for 00:09:39.080 --> 00:09:42.720 intervention or culling decisions. Most importantly, 00:09:43.240 --> 00:09:45.259 You're focusing your breeding resources on the 00:09:45.259 --> 00:09:48.200 animals most likely to conceive, which improves 00:09:48.200 --> 00:09:51.399 your overall reproductive efficiency. Let's talk 00:09:51.399 --> 00:09:53.960 about the return on investment. These systems 00:09:53.960 --> 00:09:56.840 aren't inexpensive. When do they start paying 00:09:56.840 --> 00:10:00.259 for themselves? I know what most farmers are 00:10:00.259 --> 00:10:03.279 thinking. Will automated monitoring deliver ROI 00:10:03.279 --> 00:10:06.960 on my operation? That's the right question. And 00:10:06.960 --> 00:10:10.580 the answer isn't a simple yes or no. What factors 00:10:10.580 --> 00:10:13.870 affect the payback period? A Dutch research study 00:10:13.870 --> 00:10:16.669 provides some fascinating insights. They use 00:10:16.669 --> 00:10:19.990 stochastic dynamic simulation modeling, a fancy 00:10:19.990 --> 00:10:22.470 way of saying sophisticated economic analysis, 00:10:22.769 --> 00:10:25.750 to compare visual detection with automated detection 00:10:25.750 --> 00:10:30.950 for a 130 -cow herd. What did they find? Visual 00:10:30.950 --> 00:10:35.350 detection yielded a 419 -day average calving 00:10:35.350 --> 00:10:40.289 interval and 1 ,032 ,278 kilograms of annual 00:10:40.289 --> 00:10:43.639 milk production. Automated detection reduced 00:10:43.639 --> 00:10:46.779 the calving interval to 403 days and increased 00:10:46.779 --> 00:10:54.820 annual production to 1 ,043 ,398 kg. That's an 00:10:54.820 --> 00:10:59.320 11 ,120 kg production difference, approximately 00:10:59.320 --> 00:11:03.740 85 kg per cow. Significant revenue improvement, 00:11:04.039 --> 00:11:06.840 but you must weigh it against the initial 17 00:11:06.840 --> 00:11:13.500 ,728 euro investment. roughly $136 per cow. So 00:11:13.500 --> 00:11:15.580 the payback period really depends on your starting 00:11:15.580 --> 00:11:20.580 point. Absolutely. If your estrous detection 00:11:20.580 --> 00:11:24.419 rates are below 60%, either timed AI protocols 00:11:24.419 --> 00:11:27.179 or automated monitoring can substantially improve 00:11:27.179 --> 00:11:29.600 reproductive performance and reduce cost per 00:11:29.600 --> 00:11:32.700 pregnancy. But if you're already achieving excellent 00:11:32.700 --> 00:11:36.470 estrous detection rates above 70%, The economic 00:11:36.470 --> 00:11:39.110 justification must consider additional benefits 00:11:39.110 --> 00:11:42.970 beyond heat detection. What about implementation? 00:11:43.250 --> 00:11:46.429 I imagine the technology alone doesn't guarantee 00:11:46.429 --> 00:11:48.850 success. You've hit on something critical, Bella. 00:11:48.990 --> 00:11:52.909 I've seen this countless times. Similar technologies 00:11:52.909 --> 00:11:55.250 delivering dramatically different results across 00:11:55.250 --> 00:11:58.789 operations. Why? Because implementation ultimately 00:11:58.789 --> 00:12:01.289 determines whether technology delivers transformative 00:12:01.289 --> 00:12:03.789 results or becomes an expensive disappointment. 00:12:04.799 --> 00:12:07.620 What separates successful implementations from 00:12:07.620 --> 00:12:11.200 disappointments? Several critical success factors 00:12:11.200 --> 00:12:14.200 consistently differentiate high -performing implementations. 00:12:14.419 --> 00:12:17.440 First, comprehensive staff training and buy -in. 00:12:17.620 --> 00:12:20.019 Technology alone can't improve reproduction. 00:12:20.379 --> 00:12:22.899 It requires people who understand and use the 00:12:22.899 --> 00:12:25.340 information effectively. Second, integration 00:12:25.340 --> 00:12:28.480 with existing workflows. The technology must 00:12:28.480 --> 00:12:31.039 complement rather than disrupt established management 00:12:31.039 --> 00:12:35.350 routines. Building on that, What about working 00:12:35.350 --> 00:12:37.950 with the farm's veterinarian? That's the third 00:12:37.950 --> 00:12:41.669 critical factor, veterinary collaboration. Engaging 00:12:41.669 --> 00:12:44.450 your veterinarian in system implementation dramatically 00:12:44.450 --> 00:12:48.669 improves outcomes. And fourth, continuous performance 00:12:48.669 --> 00:12:52.570 monitoring and refinement. The leading implementations 00:12:52.570 --> 00:12:55.549 establish weekly reviews of key performance indicators, 00:12:56.029 --> 00:12:58.450 monthly comparisons of system recommendations 00:12:58.450 --> 00:13:01.289 with actual outcomes, and quarterly assessments 00:13:01.289 --> 00:13:04.399 of economic impact. And I imagine setting realistic 00:13:04.399 --> 00:13:09.399 expectations is important too. Absolutely. Understanding 00:13:09.399 --> 00:13:11.840 the typical adoption curve prevents premature 00:13:11.840 --> 00:13:15.379 disappointment. Successful implementations typically 00:13:15.379 --> 00:13:18.179 see an initial adjustment period of one to two 00:13:18.179 --> 00:13:20.620 months with limited performance improvement, 00:13:20.980 --> 00:13:23.720 followed by gradual improvement over three to 00:13:23.720 --> 00:13:26.299 six months, and finally breakthrough performance 00:13:26.299 --> 00:13:29.179 once the system is fully integrated, usually 00:13:29.179 --> 00:13:32.620 around six to 12 months in. Douglas, we've covered 00:13:32.620 --> 00:13:35.899 a lot of ground today. If our listeners take 00:13:35.899 --> 00:13:38.860 away just a handful of action items, what should 00:13:38.860 --> 00:13:42.460 they be? Let's not sugarcoat it. The evidence 00:13:42.460 --> 00:13:45.759 is clear. Automated reproductive monitoring systems 00:13:45.759 --> 00:13:48.360 powered by sophisticated machine learning algorithms 00:13:48.360 --> 00:13:51.480 can fundamentally transform your operation's 00:13:51.480 --> 00:13:54.240 reproductive performance. But technology alone 00:13:54.240 --> 00:13:57.460 doesn't guarantee success. Here are five specific 00:13:57.460 --> 00:14:00.120 steps to revolutionize your reproductive performance. 00:14:01.610 --> 00:14:04.809 First, start with an honest performance assessment. 00:14:05.090 --> 00:14:07.490 Calculate your current reproductive metrics and 00:14:07.490 --> 00:14:09.950 compare them with industry benchmarks to identify 00:14:09.950 --> 00:14:13.429 your specific improvement opportunities. Second, 00:14:13.629 --> 00:14:17.289 quantify your complete economic picture. Go beyond 00:14:17.289 --> 00:14:19.950 basic reproduction costs to calculate the actual 00:14:19.950 --> 00:14:21.950 financial impact of your current performance. 00:14:22.429 --> 00:14:25.970 To estimate the minimum profit opportunity, multiply 00:14:25.970 --> 00:14:29.830 your average days open beyond $110 by $5 per 00:14:29.830 --> 00:14:33.830 cow daily. Third, select technology aligned with 00:14:33.830 --> 00:14:37.129 your specific challenges. Choose systems using 00:14:37.129 --> 00:14:39.590 backpropagation neural networks for superior 00:14:39.590 --> 00:14:41.889 performance, particularly in high -producing 00:14:41.889 --> 00:14:45.470 herds. Fourth, implement targeted reproductive 00:14:45.470 --> 00:14:48.789 protocols. Develop dual -track approaches using 00:14:48.789 --> 00:14:51.509 technology to identify animals suitable for natural 00:14:51.509 --> 00:14:54.129 service versus those requiring hormonal intervention. 00:14:55.340 --> 00:14:58.100 This targeted approach reduces hormone use by 00:14:58.100 --> 00:15:02.200 50 to 75%, improving overall performance. And 00:15:02.200 --> 00:15:04.980 finally, establish clear evaluation metrics and 00:15:04.980 --> 00:15:07.960 timelines. Set specific performance targets and 00:15:07.960 --> 00:15:11.399 evaluation points at 3, 6, and 12 months post 00:15:11.399 --> 00:15:14.139 -implementation. So to wrap up today's discussion, 00:15:14.419 --> 00:15:16.620 we've learned that machine learning technologies 00:15:16.620 --> 00:15:19.820 aren't just the future. They're transforming 00:15:19.820 --> 00:15:23.600 dairy fertility management right now. These systems 00:15:23.600 --> 00:15:26.539 can achieve 21 -day pregnancy rates above 30 00:15:26.539 --> 00:15:31.620 % while reducing hormone use by up to 75%. They're 00:15:31.620 --> 00:15:34.279 particularly valuable for high -producing cows, 00:15:34.539 --> 00:15:37.759 where traditional methods often fail. And while 00:15:37.759 --> 00:15:40.600 the technology requires an investment, the economic 00:15:40.600 --> 00:15:43.820 benefits are substantial. Each one -point improvement 00:15:43.820 --> 00:15:47.080 in the 21 -day pregnancy rate represents approximately 00:15:47.080 --> 00:15:51.940 $35 to $50 per cow annually in additional profit. 00:15:52.350 --> 00:15:54.309 The operations that will thrive through the rest 00:15:54.309 --> 00:15:56.669 of this decade will effectively combine technological 00:15:56.669 --> 00:15:59.389 capabilities with sound management fundamentals. 00:16:00.350 --> 00:16:02.929 Automated monitoring won't replace good reproductive 00:16:02.929 --> 00:16:05.350 management, but it will dramatically amplify 00:16:05.350 --> 00:16:08.110 your ability to execute your strategy with unprecedented 00:16:08.110 --> 00:16:12.029 precision. Isn't it time your reproductive management 00:16:12.029 --> 00:16:15.090 strategy evolved beyond approaches that waste 00:16:15.090 --> 00:16:17.809 money while leaving significant genetic and economic 00:16:17.809 --> 00:16:21.649 potential untapped? Your reproductive efficiency 00:16:21.649 --> 00:16:25.090 directly impacts your bottom line. And today's 00:16:25.090 --> 00:16:27.389 technology offers unprecedented opportunities 00:16:27.389 --> 00:16:30.690 to maximize that critical driver of dairy profitability. 00:16:31.250 --> 00:16:35.129 Healing. That's all for today's episode of the 00:16:35.129 --> 00:16:37.809 Bullvine Podcast. If you found this information 00:16:37.809 --> 00:16:40.389 valuable, please share it with other dairy professionals 00:16:40.389 --> 00:16:43.149 who might benefit. And don't forget to subscribe 00:16:43.149 --> 00:16:45.850 wherever you get your podcasts so you never miss 00:16:45.850 --> 00:16:50.379 an episode. Until next time, I'm Bella. And I'm 00:16:50.379 --> 00:16:51.440 Douglas. Thanks for listening.