Halter
Halter 是品类领先的虚拟围栏平台,已跑出真实牛群规模的采用;但 $2B 估值能否站住,还要看非公开经营经济性
Halter 已经搭出可信的品类领先畜牧运营平台,客户与产品规模都有真实证据;但在 $2B 估值上,公开证据仍只支持继续研究,直到私有单位经济、治理和动物福利耐久性被充分承保。
封面要素
公司概况
Halter 是一家源自 New Zealand 的私营牛业科技公司,围绕太阳能 GPS 项圈、连接能力和软件,搭建畜牧运营系统,让奶牛场主和肉牛牧场主能远程围栏、调群和监测牛群。公开证据支持它在 New Zealand、Australia 和 United States 已有实质运营牵引:服务超过 2,000 名农场主和牧场主,售出一百万条项圈,并在 2026 借助直连卫星连接,把远程肉牛场景继续铺开。同一份公开材料也显示,相比当前估值,治理可见度、经审计经营经济性和长期独立动物福利证据仍不完整。
- 创始人
- Craig Piggott, Max Olson
- 创立地点
- New Zealand
- 总部
- Auckland, New Zealand
- 产品
- Halter 销售太阳能 GPS 项圈、连接服务和移动软件,用于奶牛与肉牛经营中的虚拟围栏、牛群调动、牧场管理和动物监测。
- 客户
- New Zealand 奶牛场主和美国肉牛牧场主,以及 New Zealand、Australia 和 United States 更广泛的放牧型牛只经营者。
- 商业模式
- 以订阅为主导的硬件加软件系统,围绕按动物计价销售,并把基础设施和上手服务嵌入部署流程。
- 阶段
- Late-stage private / Series E
- 融资情况
- 2026 年 3 月 E 轮融资 $220M,估值 $2B;公开报道显示 2025 年 D 轮估值约 $1B,但累计融资总额的精确披露仍不完整。
执行摘要
主要优势
- 2,000+ 客户、售出 1M 项圈、多市场部署,给出真实规模证据。
- 产品覆盖虚拟围栏、转场、牧场管理和监测,集成度高。
- 投资人支持强,2026 年 3 月 Series E 带来新资金。
主要风险
- $2B 估值跑在 ARR、利润率、留存和股权结构条款的公开证据之前。
- 长周期独立证据仍有限,动物福利与监管正当性还带条件。
- 国际扩张、卫星 rollout 和 200+ 招聘计划抬高执行风险。
未决问题
- 经审计 ARR、收入结构、毛利率,以及按 cohort 拆分的留存。
- 股权结构表、优先股堆叠、稀释和瀑布分配细节。
- 生产环境中的独立长周期动物福利、事故率和投诉数据。
目录
01公司概况
1.1 身份定位、产品系统与地理覆盖
以私营农业科技公司标准看,Halter 现在的定位异常清晰:官方 2026 年材料称它是奶牛与肉牛农场的操作系统,底层是太阳能、GPS 项圈、连接层和移动应用,农场主可以远程围栏、调群和监测牛群。这个定位很关键。Halter 卖的不是狭窄的单点传感器,而是试图拿下牧草分配、动物引导、健康观察和用工协调的日常运营闭环。公司 2026 年 3 月 E 轮材料和 2026 年 4 月卫星发布,都支撑一个判断:Halter 已经越过 New Zealand 试点阶段,成为真正多市场平台。公开来源一致显示,Halter 覆盖 New Zealand、Australia 和 United States;总部在 Auckland,Australia 业务以 Melbourne 为基地,美国业务在 Colorado 扎根。规模口径已经大到足以支撑本报告后续判断:Halter 称其服务超过 2,000 名农场主和牧场主,售出一百万条项圈,并且现在能通过直连卫星连接,把虚拟围栏延伸到更偏远的肉牛地形。[CO001, CO003, CO004, CO005, CO006, CO007]
| 指标 | 数值 / 状态 | 截至 | 置信度 | 备注 / 缺口 |
|---|---|---|---|---|
| 创立时间 | 2016 | 2016-01-01 | 高 | 公司、投资人和媒体来源反复提到公司创立年份。 |
| 总部 | Auckland,新西兰 | 2026-03-24 | 高 | Series E 材料确认 Auckland 为总部所在地。 |
| 澳大利亚基地 | Melbourne | 2026-03-24 | 高 | Series E 和卫星发布材料均提到 Melbourne 运营。 |
| 美国基地 | Colorado / Boulder | 2026-04-28 | 高 | 美国业务扎根 Colorado;部分来源具体写到 Boulder。 |
| 当前阶段 | 后期私营公司 / Series E 轮 | 2026-03-24 | 高 | Series E 后,公司估值为 $2B,仍为私营公司。 |
| 最新估值 | $2.0B USD | 2026-03-24 | 高 | Series E 官方材料和 Business Wire 相互印证该估值。 |
| 最新轮次 | $220M USD / NZ$377M Series E 轮 | 2026-03-24 | 高 | 各媒体币种呈现不同,但金额一致。 |
| 上一估值标记 | $1.0B USD / NZ$1.65B | 2025-06-24 | 中 | Series D 报道使用不同币种,但 USD 估值一致。 |
| 客户 | 2,000+ 名农场主和牧场主 | 2026-03-24 | 高 | Series E 和后续卫星材料反复出现这一数字。 |
| 已售项圈 | 1,000,000+ | 2026-03-24 | 高 | 多个 2026 年报道来源反复出现。 |
| 员工数 | 400+ | 2026-03-24 | 中 | 官方 Series E 材料给出最清晰的当前员工数。 |
| 招聘计划 | 220+ 个开放岗位 | 2026-03-25 | 高 | Rural News 和 Farmers Weekly 重复了这波即时招聘推进。 |
| 产品范围 | 虚拟围栏、转群、健康、繁殖、牧场管理 | 2026-04-28 | 高 | 产品和发布页面都明确给出这一范围。 |
| 美国进展标记 | 自 2024 年发布以来铺设 60,000 miles 虚拟围栏 | 2026-03-24 | 高 | 官方美国扩张指标也在澳大利亚报道中写作 ~100,000 km。 |
融资、估值和足迹行以 2026 年 Series E 轮 / 卫星发布周期为当前口径;这里尽量不用 null,因为找到了公开数字,但若干数值仍依赖公司发布披露,而不是经审计文件。
[CO001, CO003, CO004, CO005, CO006, CO008]Halter 将项圈、农场软件、客户采用、投资人支持和监管保障串成一个运营模型。
[CO003, CO004, CO005, CO006, CO007, CO008]最新公开可支撑的融资、规模和运营指标压缩视图。
历史融资沿用公司对各轮的币种表述;员工数仍是公司发布的披露,不是来自监管文件的数字。
[CO008, CO009, CO010, CO012, CO013, CO020]1.2 创始人、管理梯队与治理可见度
创始人延续性仍是 Halter 故事的重要部分。Craig Piggott 仍是公开可见的领导者,外部投资人与媒体来源也继续围绕他的奶牛场成长经历、Rocket Lab 背景,以及以产品驱动数字化牛群管理的野心来讲述公司。也有足够证据把 Max Olson 视为真正的联合创始人和首任 CTO,尽管他已经不是当前运营高管。Halter 的招聘页面显示,公司已经在创始人层之下搭出一定管理厚度,重点展示了产品工程、市场、战略和运营等职能的内部晋升负责人。不过,相对于估值和增长速度,公开治理可见度仍偏弱。已审阅来源没有披露常设董事会名单、委员会结构、当前持股比例或投资人控制条款。做尽调时,这意味着公司看起来仍由创始人带队,管理纵深在改善,但治理还不够透明;外部投资人仅凭公开材料,还无法画出正式制衡、权利安排或接班韧性。[CO001, CO002, CO014, CO015, CO016, CO017]
| 人员 | 当前或已知角色 | 公开背景 | 职能覆盖 / 创始人-市场匹配 | 依赖 / 尽调备注 |
|---|---|---|---|---|
| Craig Piggott | 创始人兼 CEO | 在 Waikato 奶牛场长大,此前在 Rocket Lab 工作。 | 撑起核心产品愿景、融资,以及在农场主和投资人面前的商业化可信度。 | 关键人依赖高;大多数公开公司叙事仍围绕他展开。 |
| Max Olson | 联合创始人;前 CTO;现任职于 Vessev | 招聘页和 Vessev 简介均将他列为 Halter 联合创始人和首任 CTO。 | 重要证据,说明 Halter 从一开始就有很深的技术创始基因。 | 他目前在 Halter 的正式影响力未公开披露。 |
| Andy McLaren | 产品与工程副总裁 | 2017 年以机械工程毕业生身份加入,后来进入高管团队。 | 在创始人之下承接硬科技和工程执行深度。 | 管理梯队在加深,但组织架构深度仍是选择性披露。 |
| Helen Moore | 营销与增长副总裁 | 曾任职 Xero;2022 年加入 Halter,并多次晋升。 | 表明公司国际扩张时具备商业化扩张纪律。 | 需要更清楚拆分营销与销售领导职责。 |
| Andrew Fraser | 总裁 | 2025 年 BLM 合作公告中以总裁身份被引用。 | 为美国扩张故事提供资深商业领导力。 | 角色范围和汇报线未在公开材料中完整解释。 |
列举范围涵盖公开来源提到的创始人和最可见的现任高管;审阅材料中未找到公开董事会、委员会或所有权利安排表。
[CO001, CO002, CO014, CO015, CO016, CO017]1.3 融资历史、投资人基础与规模信号
Halter 近年的资本历史显示,公司正从强势区域创业公司走向全球关注的成长期平台。2025 年 6 月 D 轮把公司推到约 $1B 的美元估值,2026 年 3 月 E 轮又把估值抬到 $2B。这几轮投资人的重叠很重要:Bond、Bessemer、DCVC、Blackbird、NewView、Promus、Icehouse 和 Ubiquity 仍然可见,Founders Fund 也在早期投资后回归领投 2026 年轮次。公开来源还支持融资背后已有实质运营规模:超过 2,000 名客户、一百万条项圈、400 多名员工,以及 220 多人的招聘计划。不过,免费公开记录没有完全披露精确累计融资额,因为早期轮次多见于公司摘要和数据库式报道,而不是一条清晰的官方时间线。这可以处理,但并不理想。因此,资本故事读起来可信、强劲且赞助方质量高,但围绕历史实际募资额、轮次间稀释和当前投资人治理权利,仍留下尽调工作。[CO018, CO019, CO020, CO021, CO022, CO023]
| 利益相关方 | 角色 | 经济 / 战略重要性 | 最可见轮次 | 尽调备注 |
|---|---|---|---|---|
| Founders Fund | 领投方和长期支持者 | 领投 Series E 轮,也被称为早期 Series A 支持者。 | 2017 年 Series A;2026 年 Series E | 一线机构信念的重要信号,但当前持股比例未披露。 |
| BOND | 成长轮投资人 | 领投 2025 年 Series D,并留在 Series E 财团。 | 2025 年 Series D;2026 年 Series E | 帮助验证 Halter 从新西兰主导的初创公司跃迁为全球成长资产。 |
| Blackbird | 持续参投的风险投资人 | 在 Series D 和 Series E 均可见,并与 ANZ 生态支持相连。 | 2025 年 Series D;2026 年 Series E | 需要当前按比例跟投权和董事会影响力细节。 |
| Bessemer Venture Partners | 持续参投的风险投资人 | 出现在后续两轮,并据其他报道曾领投 Series C。 | 2025 年 Series D;2026 年 Series E | 各轮具体出资金额未公开。 |
| DCVC | 持续参投的深科技投资人 | 表明其对硬科技和数据密集型基础设施有信心。 | 2025 年 Series D;2026 年 Series E | 当前治理权利未公开。 |
| NewView Capital | Series D 新投资人,Series E 参与者 | 为财团增加后期跨轮资本。 | 2025 年 Series D;2026 年 Series E | 是估值支撑的有用信号,但持股规模未披露。 |
| Promus Ventures | 持续参投投资人 | 围绕国际扩张反复出现的支持财团成员。 | 2025 年 Series D;2026 年 Series E | 公开材料无法读出持股情况。 |
| Icehouse Ventures | 早期新西兰投资人 | 增加本地连续性和早期支持历史。 | 种子 / 早期轮;2025-2026 年持续支持 | 投资人页面提供有用历史,但没有当前经济条款。 |
| Ubiquity Ventures | 持续参投投资人 | 作为连续性资本的一部分出现在后续轮次财团中。 | 2025 年 Series D;2026 年 Series E | 公开权利和当前持股仍属私有信息。 |
表格强调支持估值连续性的重复投资人模式;历史融资款项和持股比例并未完全公开。
[CO018, CO019, CO020, CO021, CO022, CO023]1.4 里程碑、认可、动物福利证据与开放风险
从 2025 到 2026 的里程碑足以同时证明产品市场匹配和外部审视升温。Halter 先后完成 D 轮和规模更大的 E 轮,同时拿出 Bureau of Land Management 合作、TIME Best Inventions 认可、New Zealand 奶牛场独立公开的 ROI 研究,以及面向肉牛经营的直连卫星发布等运营标记。这些是真实证明点,不只是创始人口号。同一组来源也说明,尽调不能停在增长数字标题。Halter 自身的动物福利材料很详细,也引用独立的 Tasmanian 研究,但 New Zealand Veterinary Association 仍称长期证据有限,强有力的保障、培训和监管监督仍必不可少。Dairy Australia 同样把合法化表述为近期且有条件,而不是永久尘埃落定。客户经济性看起来有希望,但并不轻:第三方农场财务评论仍把高级 Halter 定价放在约每头牛每月 NZ$9.90,还要叠加额外基础设施和工作流改变要求。结论是,公司动能异常强,但治理披露、成本负担和动物福利监管暴露面仍需要主动承保,不能被动接受。[CO025, CO026, CO027, CO028, CO029, CO030]
| 日期 | 事件 | 类型 | 金额 / 估值 / 状态 | 参与方 | 含义 |
|---|---|---|---|---|---|
| 2016-01-01 | 公司在新西兰创立 | 创立 | 已创立 | Craig Piggott;Max Olson | 将公司定位为由创始人主导的新西兰深科技公司。 |
| 2017-01-01 | Founders Fund 首次在 Series A 投资 | 融资 | 早期机构支持 | Founders Fund | 显示一家大型美国风投机构很早就建立了高信念。 |
| 2024-01-01 | Halter 进入美国市场 | 扩张 | 进入美国 | Halter 美国团队和早期牧场主 | 为后来美国虚拟围栏和合作指标打底。 |
| 2025-06-24 | 宣布 Series D 融资 | 融资 | $165M NZD / ~$100M USD;估值 ~$1B USD | BOND、NewView、Blackbird、Bessemer、DCVC、Ubiquity、Promus、Icehouse(投资方) | 确立 Halter 独角兽身份,并为美国扩张提供资金。 |
| 2025-08-15 | 宣布 BLM 与 Foundation 合作 | 合作 | 为牧场主提供 $2.7M 支持 | Halter、BLM、Foundation for America’s Public Lands(合作方) | 将产品叙事从私人牧场延伸到公有土地管护。 |
| 2025-10-09 | TIME Best Inventions 特别提及 | 产品 | 认可 | TIME | 外部验证表明,产品在农业之外也开始被看懂。 |
| 2026-01-05 | 突出澳大利亚监管放开 | 监管 | 虚拟围栏在六个奶业州全部合法 | Dairy Australia;州监管机构 | 移除了澳大利亚的一个重要采用约束。 |
| 2026-03-24 | 宣布 Series E 融资 | 融资 | $220M USD / NZ$377M,估值 $2B USD | Founders Fund 及既有财团 | 估值翻倍,并为更广泛的全球扩张提供资金。 |
| 2026-04-28 | 面向肉牛业务推出直连卫星 | 产品 | 推出卫星连接项圈 | Halter;High Lonesome Ranch 等早期牧场主 | 将可触达肉牛市场扩展到基站曾是瓶颈的地形。 |
| 2026-05-31 | 独立 ROI 结果仍是核心商业证明 | 规模 | 10 个农场 PBT 平均提升 13.2% | AgFirst;Transform Agri;Halter 农场主 | 说明商业案例依托农场层面的生产力提升,而不只是新奇感。 |
年级别里程碑在来源支持年份但未提供更精确公开发布日期时使用 1 月 1 日。
[CO001, CO018, CO020, CO023, CO025, CO026]Halter 从新西兰创立,到全球扩张、获得认可并推出卫星赋能肉牛产品的公开时间线。
仅有年份的里程碑在图中使用 1 月 1 日;经审查来源支持年份,但不支持更精确事件日期。
[CO001, CO002, CO007, CO018, CO019, CO020]1.5 图表
02市场分析
2.1 市场边界、纳入口径与现状替代方案
Halter 正在进攻的市场,比“所有畜牧科技”窄得多,又比单一项圈 SKU 宽得多。公司的产品页面和 2026 年融资材料,把 Halter 定位为放牧型牛只运营系统,组合了虚拟围栏、调群、牧场规划和动物监测。因此,纳入支出包括项圈、软件、连接硬件或卫星回传,以及帮助农场主分配饲草、移动牛群、监测健康或繁殖的工作流层。不纳入的支出包括永久边界围栏、围栏圈、棚舍、挤奶机器人、通用 EID 耳标和室内育肥场软件。现状替代方案仍很强:牧场主可以继续用实体围栏、临时铁丝、牧羊犬、摩托车、四轮摩托和人工。WireMesh、Farm Progress 和 UC Davis 来源都强化同一点:虚拟围栏是管理叠加层,而不是彻底替代铁丝围栏。这个判断对测算规模很重要。Halter 不是要替代每一美元牛业科技支出,而是只打远程放牧控制、更频繁转场、环境隔离区和节省人工能产生足够回报的那部分。[CM001, CM002, CM003, CM015, CM016, CM017]
| 细分 / 类别 | 纳入支出 | 排除支出 | 买方 / 付款方 | 对 Halter 的意义 |
|---|---|---|---|---|
| 牧场放牧型奶牛业务 | 项圈、App、牧草分配、发情 / 健康工作流、连接 | 挤奶设备、挤奶厅自动化、牛舍软件 | 农场主 / 运营公司 | 新西兰和澳大利亚的当前核心市场。 |
| 牧场放牧型肉牛牧场 | 项圈、虚拟围栏、转群、禁入区、远程覆盖 | 饲育场软件、牲畜拍卖场工具、围场处理系统 | 牧场主 / 牧场运营实体 | 快速增长的市场,尤其是在美国和卫星覆盖地形中。 |
| 适应性 / 轮牧管理 | 软件定义围场、牧草记录、频繁转场 | 永久边界围栏和畜栏 | 业主兼管理者 / 放牧负责人 | 核心管理用例,劳动力节省和牧草收益最快显现。 |
| 动物监测层 | 位置、行为、健康、繁殖、告警 | 独立耳标或仅产犊传感器 | 业主兼管理者 / 畜群经理 | 重要扩张向量,但单独不是完整投资逻辑。 |
| 环境 / 土地管护用例 | 河岸隔离、野火燃料隔离带、公有土地管理 | 通用可持续发展报告软件 | 牧场运营方,部分情况下加上机构 / 地主 | 在美国公有土地和敏感区域叙事中很重要。 |
| 现状替代方案 | 靠狗、摩托车、卷线器和实体围栏的人工放牧管理 | None | 农场主 / 牧场主 | 最大竞争对手往往是沿用旧运营模式。 |
纳入与排除的支出受产品页和行业解释材料的证据约束;本表定义市场边界,而不是声称每一美元纳入支出都能立即触达。
[CM001, CM002, CM003, CM015, CM016, CM017]2.2 现有地理市场的多重测算视角
对 Halter 来说,可信的市场视角来自牛只存栏量,而不是通用软件 TAM 幻灯片。New Zealand 官方统计显示,截至 2025 年 6 月,奶牛为 5.75 million 头、肉牛为 3.833 million 头;LIC 则描述了一个约 10,000 个牛群、略低于 5 million 头奶牛的奶业市场。在 United States,USDA NASS 报告称,截至 2026 年 1 月,农场中有 86.2 million 头牛和犊牛,其中包括 27.6 million 头肉母牛和 9.57 million 头奶牛。在 Australia,MLA 的 2026 年 3 月行业预测把全国牛群维持在略高于 31 million 头。这些数据支持多重市场视角。窄口径运营视角只看 New Zealand 牛只、Australia 牛群和美国肉母牛,核心机会已经超过 68 million 头。更宽的上限若纳入所有美国牛和犊牛,当前地理市场的动物基础超过 126 million 头。相较之下,Halter 的一百万条已安装项圈意味着仍有充足跑道,且尚未计入 UK、Ireland、Canada 或 South America。[CM004, CM005, CM006, CM007, CM008, CM009]
| 视角 | 地理 / 范围 | 数值 | 单位 | 方法 / 来源 | 置信度 | 限制 |
|---|---|---|---|---|---|---|
| 新西兰奶牛 | 新西兰 | 5.75 | M head | Stats NZ 2025 年 6 月农业生产统计 | 高 | 动物数量,不是软件预算。 |
| 新西兰肉牛 | 新西兰 | 3.833 | M head | Stats NZ 2025 年 6 月农业生产统计 | 高 | 动物数量;并非所有养殖场都以相同经济性进行牧场管理。 |
| 新西兰奶牛群 | 新西兰 | 10 | K herds | LIC 奶业摘要 | 中 | 四舍五入的官方行业估计,而非颗粒度客户预算数据集。 |
| 美国肉牛母牛 | 美国 | 27.6 | M head | USDA NASS 2026 年 1 月牛群报告 | 高 | 狭义运营视角;排除其他牛类。 |
| 美国牛和犊牛 | 美国 | 86.2 | M head | USDA NASS 2026 年 1 月牛群报告 | 高 | 宽口径上限,而非现实可服务市场。 |
| 澳大利亚牛群 | 澳大利亚 | 31.052 | M head | MLA 2026 年 3 月预测 | 高 | 预测值且覆盖全部牛群;并非所有细分都同样适配 Halter。 |
| 当前核心地理牛群池 | 新西兰牛 + 澳大利亚牛群 + 美国肉牛母牛 | 68.235 | M head | 官方动物数量的分析性加总 | 中 | 各国定义混杂;仍是头数而非预算 TAM。 |
| 当前地理范围的更宽上限 | 新西兰牛 + 澳大利亚牛群 + 美国牛和犊牛 | 126.835 | M 头 | 官方动物数量的分析加总 | 中 | 上限口径,不是支出预测。 |
| Halter 存量部署 | 当前披露的全球部署 | 1 | M 个项圈 | Halter 2026 年融资披露 | 高 | 项圈数不等于去重付费农场数,也不等于年化收入。 |
所有值都使用存栏动物或畜群数量,因为公开证据没有把 Halter 精确产品组合对应的纯软件支出 TAM 单独拆出来。分析加总标为中等置信度,并在限制部分解释。
[CM004, CM005, CM006, CM007, CM008, CM009]在证据受限下,从广义牛只存栏上限到 Halter 当前披露装机基础的市场金字塔。
总可用市场(TAM)和可服务市场(SAM)采用官方动物数量。可获取市场(SOM)是分析估计,因为没有公开数据集能单独识别现在会购买 Halter 全套方案的运营主体。
[CM005, CM006, CM007, CM009, CM010, CM011]按狭义与广义口径,把可寻址牛群从低到高折算为百万头。
所有数值单位均为百万头。低值采用更窄的可服务口径,例如仅看奶牛或仅看肉牛母牛;高值采用更宽的官方牛只统计。
[CM005, CM006, CM007, CM009, CM010, CM011]2.3 买方、用户、付费方与采用路径
各地区的购买中心相当一致。经济买方通常是农场主、牧场主或运营主体;日常用户是农场经理或畜群团队;技术验证者往往负责放牧计划、动物处理和合规。奶牛和肉牛工作流差异足够大,不能忽略。Halter 的奶牛页面强调牧草分配、发情与健康监测,以及降低日常工作量;肉牛页面则强调远程覆盖、隔离区,以及更高效地利用粗放或陡峭土地。采用路径也比软件更运营化:生产者先测试经济性和覆盖能力,再在封闭环境训练动物,在真实牧场证明可靠性,最后才扩展到多牧区或整座牧场。供应商格局也印证这一点。Gallagher、Vence、Nofence 和 Halter 都要求买方围绕基站、订阅捆绑或蜂窝依赖做架构选择;Datamars、Cowlar 和 Moocall 等相邻供应商,则仍更专注于监测,而不是完全软件定义的放牧控制。[CM013, CM014, CM015, CM016, CM018, CM019]
| 细分市场 | 买方 | 使用者 | 付款方 | 工作流 / 预算负责人 | 采用触发点 |
|---|---|---|---|---|---|
| 新西兰 / 澳大利亚奶牛场 | 农场主或总经理 | 农场经理和挤奶团队 | 运营公司 | 牧草、动物健康和劳动力预算 | 需要更精细分配牧草、检测发情,并减少日常工作量。 |
| 美国肉牛牧场 | 牧场主或经营者 | 牧场经理 / 牲畜团队 | 牧场实体 | 放牧、劳动力和基础设施预算 | 需要减少围栏用工,利用崎岖牧地,并远程管理更大面积。 |
| 公共土地放牧经营者 | 有机构对接方的牧场经营者 | 牧场管理者 / 现场团队 | 牧场实体,有时有补助支持 | 生产效率、生态管护和土地准入预算 | 需要在保持土地生产力的同时保护敏感区域。 |
| 高强度轮牧者 | 业主兼经理 | 放牧规划员和现场团队 | 农场实体 | 牧草和围栏预算 | 频繁转场和禁牧区用铁丝围栏会太耗人力。 |
| 仅采购监测功能的畜牧客户 | 农场主 | 畜群经理 | 农场实体 | 动物健康预算 | 想要发情、产犊或疾病预警,但暂不全面采用虚拟围栏。 |
买方—使用者—付款方角色是定性划分,但与公司产品页、竞品页面和农技推广式市场解释材料一致。
[CM013, CM014, CM015, CM016, CM019, CM020]定性矩阵说明在 Halter 核心运营场景中谁购买、谁使用、谁出钱,以及架构选择如何影响采用。
评级是基于产品页面、农业推广体系的采用报道和竞争对手架构对比得出的定性评估。
[CM013, CM014, CM015, CM016, CM019, CM020]从问题识别到规模化部署的运营采用路径。
[CM017, CM020, CM021, CM024, CM027, CM028]2.4 增长驱动、采用约束与市场缺口
最有说服力的需求驱动,是劳动力、放牧精度和土地管理灵活性。Arizona extension、Farm Progress、WireMesh 和 UC Davis 都指向同一逻辑:虚拟围栏能节省时间,减少搭建或移动临时围栏的需求,并帮助经营者更精确地使用牧草、河岸隔离区、防火带或公共土地放牧。Australia 的监管放开也移除了一个重要采用障碍。但约束同样真实。Arizona 首年虚拟围栏成本可能约为每头牛 $175 到 $400,持续年度成本仍有分量。Farm Progress 引述 Halter 美国定价为每条项圈每年 $72,另加一次性塔台成本;New Zealand 咨询内容仍把高级 Halter 定价放在约每头牛每月 NZ$9.90。连接、电池、训练纪律、边界围栏和动物福利审视,仍是闸门因素。因此,市场机会很大且真实,但近期可服务市场窄于总牛只数,仍取决于生产者实际算账能否成立。[CM020, CM021, CM022, CM023, CM024, CM025]
| 驱动因素 / 约束 | 方向 | 时点 | 影响 | 尽调问题 |
|---|---|---|---|---|
| 劳动力短缺、生产者老龄化 | 利好采用 | 当前 | 能搬围栏、查畜群的人手变少,自动化价值就上升。 | 按畜群类型和地形核实实际节省的用工。 |
| 牧草生产力和轮牧 ROI | 利好采用 | 当前 | 频繁转场和更高饲草利用率,是最清晰的经济抓手。 | 要求按客户队列给出实际饲草和产出变化。 |
| 环境和生态管护用例 | 利好采用 | 当前至中期 | 河岸缓冲区禁牧、野火燃料削减和公共土地使用,把相关性从农场便利性扩展出去。 | 核实哪些用例能转化为真实预算权。 |
| 澳大利亚监管合法化 | 利好采用 | 近期 | 法律障碍移除,更大范围奶业部署打开。 | 跟踪更宽泛的合法性是否转化为快速商业采用。 |
| 高前期成本和经常性成本 | 不利采用 | 当前 | 硬件、订阅和支持成本仍可能让家庭规模经营者的回本很勉强。 | 在假设快速渗透前,按畜群规模和地形测算回本。 |
| 连接、电池和训练要求 | 不利采用 | 当前 | 运营摩擦仍然不小,偏远地形和冬季条件尤其明显。 | 索要正常运行时间、训练天数和换电间隔的性能数据。 |
| 仍需边界实体围栏 | 不利采用 | 当前 | 虚拟围栏不能完全替代铁丝围栏或畜栏。 | 按用例厘清残余实体基础设施负担。 |
| 动物福利审查和证据限制 | 不利采用 | 当前 | 长期独立证据和清晰防护措施,仍是建立信任和通过监管的关键。 | 收集独立肉牛研究和长周期动物福利研究。 |
表中同时列出正向和负向采用因素;时间范围是基于当前证据包的定性判断,不是概率预测模型。
[CM020, CM021, CM022, CM023, CM024, CM025]2.5 图表
03竞争格局
3.1 格局、竞品类别与最重要的供应商
Halter 解决的买方问题很具体:软件定义的放牧控制,加上动物和牧场可见性,并且要以适用于商业肉牛和奶牛经营的方式交付。因此,竞争集比“其他牛项圈”更宽,但比所有农场软件更窄。独立的 2025-2026 品类评测汇聚到一小组直接虚拟围栏供应商——Halter、Gallagher eShepherd、Merck Vence 和 Nofence;Corral 也在一份学术推广综述中以可供生产者选择的方案出现。围绕这个直接集合,还有第二圈相邻监测工具,例如 Datamars、Cowlar 和 Moocall;以及 Drover、Collie、Ranchflow 等新进入者,它们从耳标、AI 监测或虚拟放牧角度争夺同一笔预算。战略重点在于,Halter 不只是和其他 GPS 项圈供应商赛跑。它还必须阻止买方选择更窄的监测产品、继续用铁丝和人工拖延采用,或等待基础设施更轻的新进入者成熟。[CP001, CP002, CP003, CP012, CP013, CP014]
| 竞争对手 / 类别 | 品类 | 规模 / 融资信号 | 目标细分市场 | 差异化 | 限制 |
|---|---|---|---|---|---|
| Halter | 直接平台 | 官方披露:售出 1M 个项圈;2,000+ 名农场主和牧场主 | 新西兰 / 澳大利亚奶牛;偏远肉牛牧场 | 围绕围栏、转场、牧草和动物监测,提出更宽的操作系统叙事;肉牛场景使用卫星 | 实际定价和续约行为仍未公开 |
| Gallagher eShepherd | 直接平台 / 现有巨头背书 | 由大型围栏品牌 Gallagher 商业化 | 广阔放牧和肉牛经营 | 太阳能 GPS 颈带;蜂窝或 LoRa 选项;现有品牌信任面 | 比无需基站的模型更重基础设施 |
| Merck Vence | 直接平台 / 现有巨头背书 | Merck Animal Health 旗下 | 大型牧场和草地牛只经营 | 基站架构面向大面积土地设计 | 基站抬高安装复杂度和资本需求 |
| Nofence | 直接平台 / 独立专业厂商 | 公开产品页更强调产品结构,而不是规模 | 肉牛,以及干奶牛和后备母牛奶业工作流 | 无需基站;5 年质保;订阅模式 | 公开清单价在已审阅来源中仍不清楚 |
| Corral | 可能进入者 / 专业厂商 | 可见早期伙伴折扣;更广泛规模未公开 | 母牛-犊牛经营 | 聚焦母牛-犊牛,并以客户支持切入 | 已验证部署规模在已审阅来源中未公开 |
| Drover | 可能进入者 / 耳标型 | 已审阅来源没有公开部署数量 | 寻求更轻基础设施的牧场主 | 用耳标替代项圈;无需基站;声称无电击 | 商业牵引和定价深度的公开信息仍薄 |
| Collie | 可能进入者 / 欧洲主导 | 官网强调结果主张,而不是已部署设备规模 | 牧草奶业和放牧畜群 | 虚拟驱赶加围栏;主打省工 | 证据多为公司发布且以欧洲为中心 |
| 相邻监测技术栈 | 相邻替代品 | Datamars、Cowlar 和 Moocall 各自销售范围更窄的动物数据工作流 | 以监测为主的买方 | 用更窄范围满足健康、行为或产犊结果 | 不能覆盖完整放牧控制广度 |
| 现状 / 内部自建 | 替代品 | 除现有工具外无需软件支出 | 成本敏感或低变更经营者 | 实体铁丝、便携围栏和人工劳动更熟悉 | 牺牲远程控制、更丰富数据和更快地块调整 |
这是基于已审阅来源包构建的局部竞争格局。融资或规模没有公开可见时,表中直接说明,而不是猜测。
[CP001, CP002, CP003, CP015, CP016, CP017]基于证据的直接替代与新兴替代序位图,用 x 轴衡量产品广度,用 y 轴衡量轻基础设施部署。
坐标轴是 1-10 的序位判断,依据抓取的产品页面和独立对比。y 值越高,部署摩擦越轻,并不代表每个用例都全面更优。
[CP004, CP006, CP008, CP010, CP015, CP016]3.2 能力宽度、基础设施选择与公开定价信号
能力和基础设施设计,才是真正拉开直接竞争对手差距的地方。Halter 最强的产品宽度故事,仍是它在奶牛和肉牛两端的拆分定位:奶牛方案覆盖牧场、调群和动物健康工作流,肉牛方案现在则重押偏远地形的直连卫星覆盖。Gallagher 和 Vence 要求买方围绕蜂窝、LoRa 或基站接受更明确的基础设施选择;Nofence 则推动无基站模式,配合订阅经济性和五年质保。独立来源强化了一个判断:这些系统不是可互换模块。买方不能假设项圈、软件和训练协议可以在供应商之间干净替换,这让架构成为持久切换成本的一部分。公开定价信号也不均衡。Halter 当前的公开价格代理仍主要来自第三方评论;eShepherd 有一个学术比较点;Nofence 披露订阅结构比披露美元标价更清楚;早期进入者大多宣传试点或折扣,而不是稳定商业价格表。[CP004, CP005, CP006, CP007, CP008, CP009]
| 购买标准 | Halter | eShepherd | Vence | Nofence | 相邻监测 |
|---|---|---|---|---|---|
| 虚拟围栏和转场 | 是;奶牛和肉牛 | 是 | 是 | 是 | 否 |
| 牧草规划工作流 | 是 | 公开证据有限 | 公开证据有限 | 公开证据有限 | 否 |
| 发情 / 健康监测 | 奶牛方案支持 | 已审阅页面中不突出 | 运动和位置监测 | 不是主功能 | 是,通常是核心方案 |
| 无基站覆盖偏远地形 | 当前肉牛卫星方案支持 | 否;基站或蜂窝选项 | 否;以基站为核心 | 是;无需基站 | 通常不是完整围栏方案 |
| 母牛-犊牛专门化 | 有部分肉牛支持 | 是,面向广阔放牧 | 是,面向草地牛只 | 不太明确 | Corral 明确聚焦母牛-犊牛 |
| 公开定价清晰度 | 只有部分第三方代理值 | 一个学术比较点 | 仅品类比较 | 合同模式公开,美元清单价不清楚 | 多为报价驱动或产品驱动,净价不透明 |
标为有限或部分的单元格,反映公开页面和独立评论能支撑的程度。未知商业深度保持明确,不从沉默中推断。
[CP004, CP005, CP006, CP007, CP008, CP009]| 供应商 | 价格 / 单位 / 合同模式 | 包含能力 | 折扣 / 未知项 | 影响 |
|---|---|---|---|---|
| Halter | 第三方代理值约 NZ$9.90 / 头牛 / 月 | 虚拟围栏、转场和更宽的操作系统工作流 | 实际折扣、硬件条款和续约行为未公开 | 价值叙事更宽,但公开价格透明度仍弱 |
| Gallagher eShepherd | Rangelands 比较列出每个基站 $4,500、每个项圈每年 $66 的成本 | 虚拟围栏,配太阳能颈带以及蜂窝或 LoRa 设置 | 客户特定安装和支持条款仍不清楚 | 架构可适配复杂地形,但前期基础设施支出是显性的 |
| Merck Vence | Arizona 成本研究称 Vence 和 Halter 大致相当;品类评审显示基站经济性很关键 | 虚拟围栏、放牧控制、位置监测 | 已审阅免费来源没有找到清晰公开清单价 | 买方经济性可能取决于土地面积和基站数量 |
| Nofence | 第一年包含应用访问、支持和更新;之后按月或按年订阅 | 无基站虚拟围栏,5 年质保 | 抓取的牛只页面未显示美元清单价 | 安装摩擦更低,即使清单美元价不透明,也能帮助试点 |
| Corral | 早期伙伴享折扣价和专属支持 | 面向母牛-犊牛经营者的虚拟围栏 | 没有标准公开价目表 | 更像早期商业化 GTM,而非成熟目录定价 |
| 品类成本区间 | Powerflex 称项圈价格为 $300-$500,另有硬件和 $30-$100 年订阅费 | 随供应商架构而变 | 区间是品类层面,不是谈判报价 | 相对于现有围栏和人工,成本仍是真实门槛 |
公开定价零散。此表只记录已抓取的清单价、比较表或合同结构信号,并把谈判后的经济性明确列为未知。
[CP023, CP024, CP025, CP026, CP027, CP028]按类别比较 Halter 在已评审替代方案中看起来最强和最暴露的环节。
正向、中性、警示和负向标签是对已抓取公开证据的定性摘要。未知单元格刻意以谨慎表示,而不是用猜测填满。
[CP005, CP007, CP009, CP011, CP019, CP031]3.3 切换成本、现状替代方案与分销力量
对任何虚拟围栏供应商来说,最大的竞争对手往往仍是少做一点。实体围栏、便携铁丝围栏和人工调群,在劳动力可得且经营者想避免高额前期项圈和基础设施成本的地方,仍是可信选择。这一点很重要,因为买方不必只在 Halter、Gallagher、Vence 和 Nofence 之间选择;他们也可以推迟整个品类。当买方真正参与时,切换成本并不低。训练动物、培训员工、学习每家供应商的地图逻辑,以及围绕不可互操作组件运行,都会制造摩擦。分销力量在各家之间也不同。Gallagher 和 Merck 可以把虚拟围栏挂到更大的地面销售和信任界面上;Nofence 和 Halter 则在更简单的部署故事比既有渠道宽度更能打动买方时获胜。相邻监测供应商承诺健康或产犊结果,却不要求买方投入完整的软件定义放牧栈,从而给预算分配增加压力。[CP030, CP031, CP032, CP033, CP034, CP035]
3.4 护城河耐久性、进入者压力与最难的竞争问题
Halter 的护城河真实存在,但更像运营护城河,而不是绝对护城河。在已审阅集合中,公司拥有最清晰的公开装机基础证明点,也有差异化的远程覆盖叙事,并且比多数同行更强调运营系统。同时,这个领域不太可能保持静止。Dairy Australia 明确预期监管开放后会出现新进入者;Corral、Drover、Collie 和 Ranchflow 也显示,竞争压力可以来自专门的母牛-犊牛工作流、耳标形态、虚拟放牧或 AI 主导的监测,而不只是直接复制 Halter 的项圈栈。最难的承保问题,在公开来源中仍未解决:真实净定价、续约行为、各竞争对手的客户集中度,以及较新进入者经过验证的部署规模。这些缺口很重要,因为它们决定 Halter 看似领先的地位,究竟能否在经济上持久,还是只是一个基础设施变轻、功能主张趋同后会被压缩的品类第一波。[CP037, CP038, CP039, CP040, CP042, CP043]
| 护城河主张 | 威胁 | 严重性 | 缓释措施 / 尽调问题 |
|---|---|---|---|
| 装机基数证明和产品广度 | 新进入者仍可切入子工作流,如耳标、AI 监测或虚拟驱赶 | 中高 | 索要队列级证据,证明广度确实提升留存和扩张,而不只是宣传册范围 |
| 远程部署灵活性 | 如果买方接受更多基础设施,Gallagher、Vence 和其他对手也能竞争 | 中 | 按地形类型和连接画像映射赢率,不要假设一种架构通吃 |
| 动物训练和运营诀窍 | 全品类成本可见后,现有围栏和人工仍显得更便宜 | 高 | 索要实施手册、训练完成率,以及按畜群类型拆分的价值实现时间 |
| 信任和动物福利姿态 | 即便单个供应商执行很好,品类层面的审查也会拖慢采购或合法化 | 高 | 审查动物福利事件、监管沟通和客户保护措施,不要依赖营销材料 |
| 定价权 | 公开定价不透明,难判断功能广度是否转化为净收入质量 | 高 | 拿到实际价格瀑布、硬件补贴和续约 / 流失数据 |
| 渠道和采购准入 | Gallagher 和 Merck 可把信任和现场销售关系打包进竞争性投标 | 中高 | 审查采购赢输记录,以及哪些买方选择专业厂商、哪些选择现有巨头 |
严重性是基于对切换、采购和毛利耐久性影响的判断。该登记表是把已抓取证据转化为尽调问题,不声称量化损失结果。
[CP031, CP032, CP033, CP035, CP036, CP037]一组紧凑的公开竞争标记,最能解释为什么 Halter 现在看起来领先,但并非追不上。
这张记分卡混合了已披露指标和硬产品事实。它不是市场份额模型或投资评分。
[CP009, CP023, CP028, CP038, CP040]3.5 图表
04财务情况
4.1 收入模式、变现方式与公开定价真正说明了什么
Halter 的公开材料指向的是混合模式,而不是单一干净的 SaaS 费用。产品按系统销售:项圈、连接、虚拟围栏、牧场和牛群工作流,以及越来越多面向肉牛或顾问的工具,把使用深度推到简单围栏之外。在变现上,独立报道比公司自身更明确。AgFunder 描述的是按牛逐月计费的订阅业务,再叠加一次性塔台基础设施;iStart 和 CMK 则提供了公开价格代理,大致集中在按动物逐月收费。这给出了可用的方向,尽管官网仍没有公布稳定费率表。更合适的读法是,Halter 通过动物级经常性使用来变现,同时也背负会随市场、牛群规模和连接路径变化的硬件与部署经济性。采用一旦粘住,这在财务上很有吸引力;但仅凭公开定价证据,不能把公司建模成低摩擦的纯软件业务。[CI001, CI002, CI005, CI006, CI007, CI008]
| 收入流 | 机制 | 单位 | 当前值 / 状态 | 质量 | 尽调问题 |
|---|---|---|---|---|---|
| 经常性动物订阅 | 按动物每月收费,覆盖虚拟围栏和管理工作流 | USD 或 NZD / 头动物 / 月 | 公开代理值约 US$5-$8 和 NZ$9.90,但无官方价目表 | 中 | 按市场、畜群规模和产品组合获取当前价目表 |
| 项圈硬件 | 与操作系统绑定的智能项圈部署 | 每个项圈 / 每头动物 | 收入组合中有隐含,但未单独披露 | 中低 | 索要硬件 ASP、补贴政策和更换节奏 |
| 塔台或基础设施费用 | 基于塔台的部署收取一次性基础设施费用 | 每座塔台 / 站点 | AgFunder 报道一次性塔台费用;卫星更新可能降低部分肉牛场景需求 | 中低 | 索要传统部署与卫星部署经济性和硬件毛利率 |
| 高阶产品扩展 | 动物健康、繁殖、牧草和顾问工具叠加在核心系统上 | 套餐 / 模块 | 产品广度在扩展,但模块变现未披露 | 低 | 索要附加率和模块级定价 |
| 服务 / 上线支持 | 客户成功、上线推进和行为改变支持 | 项目制 / 打包 | 运营上显然重要,但未披露为收入线 | 低 | 要求提供实施费、支持人员配比和服务毛利率 |
公开证据足以识别收入机制,但还不足以量化收入结构。各行区分已经可见的内容和仍属私有的信息。
[CI001, CI005, CI006, CI007, CI008, CI009]| 信号 | 价格 / 单位 / 合同 | 标价 vs 实际价格 | 来源质量 | 含义 |
|---|---|---|---|---|
| iStart 代理值 | US$5-$8 / animal / month | 独立媒体代理值,并非官方价目表 | 中 | 支持按动物计费的经常性变现,但无法说明净实际价格 |
| AgFunder 代理值 | $6-$10 / cow / month,另有一次性基础设施费 | 基于市场评论的独立估计 | 中 | 指向订阅加硬件或场地经济性的混合模式 |
| CMK 代理值 | NZ$9.90 / cow / month | 咨询式市场评论 | 中 | 将新西兰定价放在高端可穿戴设备区间内 |
| DTN 卫星更新 | 项圈和订阅价格几乎与此前相同,但不再需要塔 | 独立更新;不是完整价格表 | 中 | 卫星可能改善客户 ROI,但看不到明显标价上行 |
| 官方定价披露 | 在已审阅来源中未找到持续有效的官方公开价目表 | 不可得 | 对缺失有高确定性 | 私下谈判和按市场设定的条款可能实质性影响价格 |
本表只记录已抓取的公开价格信号。实际成交价、折扣和续约提价仍属私有,可能与标价或文章代理值存在实质差异。
[CI006, CI007, CI008, CI009, CI026, CI027]项圈、连接和工作流使用看起来如何转化为经常性收入与扩张价值。
由于 Halter 不公布收入瀑布,这座桥是定性的。它区分经常性和非经常性机制,而不是量化每一项。
[CI001, CI004, CI005, CI006, CI007, CI009]4.2 牵引、收入质量信号与公开来源允许的单位经济代理
最强的公开财务证据不是 P&L,而是牵引标记和客户结果证明的组合。Halter 的 E 轮公告给出最佳规模锚点:一百万条项圈售出、超过 2,000 名客户、400 多名员工。这些数字只能支持粗略运营代理,但粗略代理也有用:披露的平均每客户项圈数足够大,意味着是真正的牛群级部署,而不是纯试点活动;客户数相对员工数也说明,产品周围仍需要相当多人力。收入质量更多由运营结果支撑,而不是会计披露。十家农场研究和顾问材料显示,利润、牧场、牛奶和劳动力都有改善;产品一旦落地,就像任务关键系统。缺失的是投资人通常用来把价值证明转成单位经济性的那组数据——队列留存、获客成本(CAC)、回本周期和净收入留存。公开证据因此只能方向性支持粘性,还不能给出数值证明。[CI010, CI011, CI015, CI016, CI018, CI019]
| 指标 | 数值 / 状态 | 置信度 | 重要性 | 尽调要求 |
|---|---|---|---|---|
| 已披露客户数 | 2,000+ | 高 | 为粗略规模和平均牧群渗透率分母提供锚点 | 确认活跃付费账户数与历史累计 logo 的差异 |
| 已披露售出项圈数 | 1,000,000+ | 高 | 显示部署深度,并支持经常性收入潜力判断 | 拆分活跃项圈、流失项圈和更换项圈 |
| 平均每客户项圈数 | ~500 | 中 | 指向真实牧群部署,而不只是试点 | 按牧群规模和地域提供分布 |
| 每员工服务客户数 | ~5 | 中 | 粗略代理运营杠杆和服务强度 | 提供客户成功和现场服务人员结构 |
| ARR 估计 | US$70M-$100M(第三方估计) | 低 | 已审阅材料中唯一公开的经常性收入估计 | 提供经审计 ARR,以及经常性与非经常性收入桥表 |
| 农场利润提升 | 十个农场研究显示约 13%-13.2% | 高 | 支持价值主张和续约逻辑 | 提供更大样本 ROI 队列,以及按细分市场划分的回本期 |
| 毛利率 | 未披露 | 低 | 判断硬件支撑型经常性经济性时,这是关键指标 | 提供硬件、服务和软件毛利率拆分 |
| CAC / 回本期 / 留存 | 未披露 | 低 | 承销需要核心效率指标 | 提供 CAC、回本期、NRR、流失率和销售周期数据 |
私有指标缺失时,本表记录最强的公开代理值和具体尽调要求。估计项已明确标注。
[CI010, CI011, CI015, CI018, CI019, CI025]从获客到农场结果的公开单位经济逻辑,并明确标出主要非公开阻断项。
这是一座公开证据桥,不是内部队列模型。CAC、毛利率和留存输入缺失,故意展示而非猜测。
[CI004, CI015, CI016, CI018, CI019, CI020]公开数字锚点与低置信估算区间会塑造财务判断,但不能定论。
这张图的标签混合多种单位;单位嵌在各标签里。它用于展示公开估算区间和确定性缺口,而不是单一可比尺度。
[CI007, CI008, CI012, CI025, CI035]4.3 成本结构、资本强度与 E 轮后的资本充足性
成本结构是 Halter 最明显偏离标准垂直 SaaS 故事的地方。面向肉牛的工程、项圈、连接、客户铺开和持续产品开发,都意味着毛利率和现金转化会受硬件与服务层塑形,而不只是软件使用塑形。卫星扩张可能通过移除塔台要求改善客户经济性,但不会神奇地把公司变成轻资产;它只是把运营模式转向更广覆盖,同时继续承受产品和伙伴复杂度。可比公开文件即便不直接披露 Halter 数字,也有助于建立框架。Merck 的 SEC 文件显示,一家规模化动物健康平台仍有可见销售成本、SG&A 和 R&D 层,这比纯云基准更有方向性可比性。最新一轮融资明显改善了资本充足性,也给管理层留下地理扩张和大举招聘的空间。但由于现金余额、烧钱速度、现金跑道、债务和毛利率仍未披露,外部投资人只能得出一个结论:Halter 资金更充足了;至于按当前扩张速度能撑多久,仍无法精确判断。[CI003, CI012, CI013, CI014, CI021, CI022]
| 项目 | 当前数值 / 状态 | 置信度 | 重要性 | 尽调要求 |
|---|---|---|---|---|
| 最新融资 | US$220M / NZ$377M Series E(融资轮) | 高 | 刷新资产负债表容量,并为扩张提供资金 | 确认扣除费用后的净募集资金和任何指定用途 |
| 手头现金 | 公开未披露 | 低 | 核心现金跑道输入仍缺失 | 提供季末现金和受限现金 |
| 月度烧钱 | 公开未披露 | 低 | 需要用它把融资规模换算成现金跑道 | 提供过去六个月的经营性现金消耗 |
| 现金跑道月数 | 公开未披露 | 低 | 决定下一轮融资时点和融资风险 | 提供基准情景和下行情景下的现金跑道假设 |
| 计划资金用途 | 扩张、新产品和 200+ 名新增招聘 | 高 | 说明本轮是增长资本,而非纯防御性流动性 | 按地域、产品和招聘提供预算拆分 |
| 下一轮触发条件 | 公开未说明 | 低 | 对估值和稀释规划很关键 | 提供下一轮融资或盈亏平衡的内部里程碑 |
| 债务 / 项目融资义务 | 未发现公开披露 | 低 | 如果硬件扩张靠外部融资,债务可能实质性改变风险 | 提供债务表、契约、租赁和供应商融资 |
本表把公开记录实际披露的内容与其遗漏的关键流动性输入拆开。缺少披露本身就是尽调发现,不是中性占位符。
[CI012, CI013, CI029, CI030, CI035, CI042]公开证据指向真实现金需求的地方,以及披露在完整投资承销前断裂的地方。
语气标签概括公开证据质量和可能现金影响。它们不是经审计的支出比例。
[CI012, CI013, CI022, CI029, CI035, CI041]4.4 财务判断、主要风险与承保缺失证据
对 Halter 做可支撑的公开市场式财务判断,必须把业务质量和披露质量拆开。业务质量看起来有希望:公司卖进日常农场工作流,刚拿到增长资本,也至少有一些第三方定价信号和 ROI 证明。披露质量弱得多。收入结构不公开,毛利率不公开,现金和烧钱速度不公开,留存或回本指标也不公开。这意味着投资人可以承保需求和战略动能,但还不能承保完整资本效率。主要反向信号并不是业务缺乏牵引,而是虚拟围栏仍有可见按头成本和动物福利服务负担;如果客户支持需求居高不下,两者都可能放慢采用或压缩经济性。实际含义很简单:Halter 可能是一门优秀生意,但投资人要负责任地声称相信其利润率路径或现金跑道耐久性,仍需要管理层访问和非公开财务材料。[CI025, CI026, CI031, CI032, CI033, CI034]
| 缺失的私有指标 | 影响 | 重要性 | 精确尽调路径 |
|---|---|---|---|
| 经审计 ARR / 收入 | 重大 | 第三方 ARR 估计不足以支撑估值或现金跑道分析 | 要求提供经审计收入桥表,并拆分经常性与非经常性收入 |
| 按收入线划分的毛利率 | 重大 | 硬件支撑型经常性业务可能需求端很好看,但毛利端很弱 | 要求提供硬件、服务和软件毛利率瀑布图 |
| 现金 / 烧钱 / 现金跑道 | 阻断项 | 仅有最新融资规模,看不出扩张还能烧多久 | 要求提供月度现金流、烧钱历史和董事会现金跑道模型 |
| CAC / 回本期 / NRR / 流失 | 重大 | 没有队列效率和留存,无法完整判断收入质量 | 要求按奶牛 vs 肉牛、地域和产品包提供队列表 |
| 债务 / 租赁 / 表外义务 | 重大 | 隐性融资可能实质性改变下行风险 | 要求提供债务表、租赁承诺和供应商融资条款 |
| 硬件、订阅和服务收入结构 | 重大 | 收入结构决定毛利率路径,也决定业务究竟有多可扩展 | 要求管理层 P&L 按收入流和地域分段 |
这些是从有吸引力的公开叙事推进到可承销财务模型时,最低限度缺失的指标。
[CI024, CI026, CI027, CI028, CI029, CI030]4.5 图表
05产品与技术
5.1 产品系统、用户工作流与公开表面真正证明了什么
核心产品问题是,Halter 卖的是项圈、牧场管理应用,还是完整操作系统。公开材料更支持第三种解读,而不是前两种。Halter 自身的技术表面,把项圈硬件、连接、移动端和网页端工作流,以及机器智能打包成一个系统,覆盖奶牛和肉牛场景中的日常调群、牧草分配、繁育支持和远程监测。奶牛产品围绕高强度、高频次农场管理展开,包括近实时调群和发情 / 健康更新;肉牛产品则围绕崎岖偏远土地、更低基础设施负担和卫星覆盖展开。这个区别很重要,因为它说明 Halter 搭建的是针对工作流的产品逻辑,而不是把同一条通用项圈套上不同营销话术再卖。围绕 2026 年卫星铺开的独立报道,也强化了一个判断:肉牛不是奶牛产品的复制版,而是一套部署架构不同、运营约束和买方价值驱动不同的独立方案。[CE001, CE002, CE003, CE004, CE005, CE006]
| 模块 / 资产 | 主要用户 | 工作流角色 | 成熟度 / 状态 | 差异化 | 尽调缺口 |
|---|---|---|---|---|---|
| 智能项圈 | 农场主 / 牧场主 | 引导牛群、采集遥测数据,并用动物端硬件锚定整个系统 | 已在奶牛和肉牛场景上线 | 太阳能形态加提示传递逻辑,是产品身份的核心 | 无公开 MTBF、电池更换节奏或硬件故障数据 |
| 塔式连接 | 奶牛场运营者 | 在没有移动信号的农场范围内连接项圈 | 已在奶牛场覆盖范围上线 | 自供电的农场覆盖层避开了对电信运营商的依赖 | 无公开覆盖密度、冗余或故障历史 |
| 卫星连接 | 偏远肉牛运营者 | 在没有塔或蜂窝覆盖的崎岖土地上扩展项圈连接 | 2026 年面向肉牛推出 | 轻基础设施部署打开了公共和租赁土地使用场景 | 无公开卫星延迟、电池权衡或单位经济性披露 |
| 移动 / Web 应用 | 农场主 / 牧场主 / 员工 | 创建转场、围栏、任务和监控视图 | 已上线,并有活跃的 2026 年更新 | 牧群和牧场的工作流控制与监控集中在同一界面 | 无公开 API 或集成文档 |
| Intelligence / Cowgorithm / 数据平台 | 运营者和 Halter 内部模型 | 将遥测数据转成引导、警报、牧草估计和决策支持 | 已上线,但部分不透明 | 公司声称拥有大型行为数据语料,并以机器学习定位 | 无公开模型准确率、误报或数据治理细节 |
各行把可见的已上线模块,与仍主要依赖供应商材料支撑的架构或数据主张拆开。
[CE001, CE003, CE004, CE005, CE008, CE009]| 用户任务 | 当前工作流 / 痛点 | Halter 解决方案 | 可衡量收益信号 | 限制 |
|---|---|---|---|---|
| 奶牛场每日换围 | 人工移动围栏,反复驱赶牲畜 | App 驱动的虚拟分区和远程转场 | Halter 称奶牛转场约 90 秒内响应 | 公开记录中没有独立工时动作基准 |
| 牧草分配和利用 | 走场看草,饲料可见性滞后 | 实时围场数据加颗粒度更细的牧草更新 | 公开材料把牧草利用率定位为核心 ROI 杠杆 | 无公开跨农场校准或错误率披露 |
| 繁育和发情管理 | 配种管理依赖大量观察,记录割裂 | 产品工作流提供发情列表、发情活动和估计妊娠状态 | 路线图和模块页显示,繁育支持是一等功能 | 无公开误报率或按品种划分的表现数据 |
| 崎岖土地上的远程肉牛放牧 | 需要实体围栏、临时电围线,或地形难以进入 | 接入卫星的项圈,不需要建塔 | 独立报道把本次发布与超大牧场地形联系起来 | 无公开吞吐量、电池或月度连接经济性 |
| 公共 / 租赁土地放牧 | 建围栏可能被禁止,或经济上不划算 | 项圈加卫星让临时放牧控制无需永久基础设施也能实现 | 官方定位和 DTN 报道在这一用例上相互一致 | 监管和动物福利接受度仍随司法辖区和利益相关方而变 |
收益信号混合了公司声称的工作流收益和独立报道;它们不是受控 ROI 研究。
[CE002, CE005, CE020, CE021, CE022, CE029]从公开材料看,Halter 的技术栈由动物硬件到云端智能和用户工作流,分成四个相连层。
Halter 未公布正式参考架构,层边界由分析师根据公开材料划分。
[CE001, CE008, CE009, CE010, CE011, CE016]公开工作流始于应用里的放牧意图,止于牛群移动、监测和管理洞察。
[CE002, CE005, CE020, CE021, CE022]5.2 架构、数据飞轮与开发者信号
Halter 的产品主张只有在技术栈看起来真实、持续维护且有差异化时才有意义。这里最强的公开证据,不是公开 API 手册或工程架构图,而是详细官方产品语言、专利标记、应用发布信号和可见的软件招聘内容组合。Halter 称其系统通过云平台和机器学习模型,把项圈遥测转成可行动的工作流;公司还量化了数据流和历史行为语料库,暗示这不只是薄薄一层控制应用。专利标记页面和独立受让人索引,也显示公司围绕植被覆盖建模、牲畜引导和无线中继进行主动保护。同时,软件工程师画像和移动应用表面说明,组织仍在持续发软件,而不是围绕硬件 SKU 冻住。不过,公开表面对第三方实施细节仍明显单薄:有软件使用政策语言,但仍没有公开 API 或集成文档,无法让买方检查系统能多深地接入相邻牧场或奶业软件。[CE007, CE008, CE009, CE010, CE011, CE013]
| 层级 / 组件 | 角色 | 证据锚点 | 依赖 | 关键风险 |
|---|---|---|---|---|
| 项圈硬件 | 采集遥测数据并传递方向提示 | 技术页面描述太阳能、耐用性和提示传递 | 硬件制造、电池服务、现场支持 | 公开故障率和服务周期不透明 |
| 塔或卫星网络 | 在动物和应用之间传输遥测数据与指令 | 官方技术页加 DTN 发布报道 | 奶牛场需要场内塔安装;肉牛走 Starlink 路径 | 覆盖、延迟和电池权衡未公开量化 |
| App 界面 | 将遥测数据转成转场、警报和任务管理等动作 | 应用商店加路线图页面 | 移动 OS 支持和内部发布节奏 | 无公开基于角色的权限或集成细节 |
| 云数据平台 | 将遥测数据处理成实时运营洞察 | 官方描述每分钟 6,000 个数据点和云端处理 | 数据基础设施和存储经济性 | 无公开正常运行时间、数据留存或韧性指标 |
| 机器学习 / IP 层 | 牧草覆盖估计、引导逻辑和行为解读 | Cowgorithm 主张加专利标记页和受让人索引 | 模型训练数据和专有 know-how | 无公开模型验证包或客户可见的准确率阈值 |
本表停留在层级层面,因为 Halter 未发布低层架构图或开发者实施指南。
[CE008, CE009, CE010, CE011, CE013, CE014]Halter 交付产品,要靠硬件、连接、动物福利正当性和自研软件 / 数据层协同运转。
[CE005, CE014, CE015, CE016, CE023, CE027]5.3 信任、动物福利与监管控制真实存在,但仍部分由 Halter 自己书写
与许多早期农业科技公司相比,Halter 已经发布了更多治理和动物福利材料。公司维护结构化动物福利章程,公布 Veterinary Advisory Board,阐述方向性提示训练逻辑,并解释其如何处理各市场的立法合规。这些控制有用,尤其是因为虚拟围栏采用仍取决于动物福利正当性,而不只是 ROI。独立来源也支持 Australia 监管推进,以及 2026 年合法使用扩展到奶牛场主。但信任图景还没有完全闭合。批评性评论仍然存在,包括动物福利怀疑,以及 RSPCA 在 UK 采取限制性立场等例子。Halter 自身动物健康页面引用了非环境性跛行减少等轶事收益,但长期独立证据基础仍薄于公司的产品成熟度主张。实际看,信任 / 控制栈已经足够可信,可以支持继续承保商业进展;但还不足以消除围绕动物福利结果、安全保障和现场可靠性透明度的尽调要求。[CE012, CE023, CE024, CE025, CE026, CE027]
| 控制 / 信号 | 状态 | 范围 | 支撑内容 | 缺口 |
|---|---|---|---|---|
| Animal Welfare Charter | 已发布 | 公司层面的福利叙事和控制 | 显示 Halter 把福利作为正式产品治理议题 | 仍是供应商撰写,而非独立保证 |
| Veterinary Advisory Board | 已发布 | 福利框架内具名的监督结构 | 提升动物福利治理的可信度 | 未发布章程会议记录或权限边界 |
| 训练方法 | 已发布 | 可预测、可控的提示逻辑和联想学习框架 | 支持“训练是系统化流程,而非临时做法”的主张 | 无公开实时表现或压力结果仪表盘 |
| 澳大利亚监管姿态 | 2026 年为正面 | 公司总结加 Dairy Australia 报道 | 支持虚拟围栏目前在澳大利亚各司法辖区对牛群合法 | 投资人法律顾问仍需要司法辖区层面的法律包 |
| 软件 / 地理空间政策 | 已发布 | 农村专业人士使用软件和影像数据 | 显示围绕数据和软件使用有一定合同纪律 | 无公开安全认证或隐私审计材料 |
控制是真实存在的,但最强的公开信任信号仍是政策和治理页面,而不是第三方保证报告。
[CE012, CE023, CE025, CE026, CE027, CE028]公开证据显示核心工作流栈已经成熟上线,但外部保证和集成方面的成熟度证据薄得多。
这是证据质量矩阵,不是内部产品记分卡。
[CE018, CE019, CE027, CE031, CE037, CE038]5.4 成熟度在现有工作流覆盖上最强,在外部技术透明度上最弱
路线图证据足够显示这是一个活产品,而不是冻结的 2025 年叙事。Halter 的公开更新包括更细颗粒的牧场数据、繁育管理功能、活跃移动应用修订,以及至少一个仍标记为即将推出的功能;这正是真实产品团队应有的组合。因此,成熟度图景在用户可见工作流上最强:调群、牧场转移、繁育支持,以及新的卫星赋能肉牛部署。较弱之处,是企业买方和后期投资人最终会追问的项目:正常运行时间历史、事故处理、电池服务节奏、安全审计,以及面向合作伙伴或农村专业人士的实施工具。这些缺口不会推翻产品故事,但会限制外部审阅者对可靠性和集成深度的打分信心。结果是,产品章节读起来商业上真实、技术上有差异化,但对外工具化程度仍未达到成熟基础设施软件供应商标准。[CE017, CE018, CE019, CE029, CE033, CE034]
| 日期 / 阶段 | 功能 / 里程碑 | 状态 | 含义 | 来源 |
|---|---|---|---|---|
| 2026-04 | 面向肉牛项圈的直连卫星连接 | 已推出 | 2026 年最显眼的架构变化,也是最清晰的肉牛专用产品切入口 | Halter 技术页;DTN |
| 2026 年当前 | 颗粒度更细的牧草数据 | 新增功能页面仍在宣传的上线更新 | 显示 Halter 仍在打磨牧场管理工作流 | Halter 新功能 |
| 2026 当前 | 发情检测和繁育管理套件 | 已上线工作流,并在产品界面宣传 | 表明 Halter 正从围栏约束拓展到繁殖决策支持 | Halter 新功能;改善配种结果 |
| 2026 当前 | 固定在农场地图上的任务分派,支持照片 / 评论 | 即将推出 | 说明路线图里仍有运营协同功能在推进 | Halter 新功能 |
| 2026-05 访问 | 移动 App 发布节奏 | 已上线并持续更新 | 应用商店更新说明,硬件叙事之外软件也在持续维护 | Google Play;Apple App Store(应用商店) |
该路线图来自公开发布页面,并非私有产品路线图;「即将推出」项目应视为意图,而不是已承诺交付。
[CE017, CE018, CE019, CE029, CE033, CE034]5.5 图表
06客户情况
6.1 可见客户分群、买方逻辑与公开客户簿真正覆盖了什么
Halter 的客户簿并非向所有方向都很宽,但足以建立真实市场足迹。可见模式是一套双引擎客户基础:一端是 New Zealand 奶牛场,另一端是美国肉牛牧场。这与上一章建立的产品拆分相匹配,也解释了为什么客户故事会按分群强调不同价值语言。奶牛材料偏向劳动力、牧场、繁育和存栏单位生产力;美国牧场主材料则强调避免建围栏、承载能力、公共土地灵活性,以及冬季或崎岖地形管理。两类场景的经济买方很可能都是业主经营者或牧场 / 农场决策者,因为公开故事反复回到运营 ROI,而不是下游加工商或消费者结果。缺失的是分群级变现深度:公开材料足以显示客户异质性和真实需求,但不足以排序哪个分群拥有最佳留存或收入质量。[CU001, CU002, CU003, CU014, CU015, CU016]
| 分群 | 买方 / 用户 / 付款方 | 使用场景 | 规模信号 | 收入 / 战略价值 | 缺口 |
|---|---|---|---|---|---|
| 业主经营型新西兰奶牛场 | 买方和付款方似乎是农场主 / 经理;用户是农场运营人员和员工 | 牧草分配、减少用工、繁育支持和牛群监测 | 已命名农场规模为 440、580、733 和 1,545 头奶牛 | 可能构成核心经常性收入基础,绑定日常工作流深度 | 未公开单个奶牛场 ARR 或续约画像 |
| 美国肉牛牧场主 / 经理 | 买方和付款方似乎是牧场主 / 经理;用户包括牧场工人和运营人员 | 虚拟围栏、提升载畜量、冬季放牧、公地灵活性和避免实体围栏 | 已命名案例覆盖 100、160、300、400 和 900 头,另有大面积牧场 | 是美国增长和卫星用例的重要战略分群 | 未公开不同牧场类型的实际回本或合同期限 |
| 大型偏远牧场 | 牧场管理层和运营团队 | 借助卫星在广阔地形上放牧 | High Lonesome 225,000 英亩案例 | 基础设施轻量化扩张的重要参考分群 | 只有少数大型牧场公开参考案例 |
| 草场密集型新西兰丘陵牧场经营者 | 农场主经营者 | 牲畜单位翻倍、用更少精力产出更多牛奶、以草场驱动管理 | 重点展示 Waikato 和丘陵地区案例 | 说明 Halter 能卖的不只是替代围栏,还有结果叙事 | 公开证据仍偏向最佳案例农场 |
| 公地 / 伙伴介导账户 | 牧场主加保护组织或公地对手方 | 在难以或不宜修建永久围栏的区域做准入和放牧管理 | Rio Grande 和 NSW 倡议案例 | 形成类似渠道的扩张路径和政策杠杆 | 商业条款和可复制性未公开 |
该表分群的是可见公开证据样本,不是 Halter 的完整客户账本。
[CU001, CU002, CU003, CU014, CU015, CU021]Halter 的客户旅程似乎从放牧痛点走向实时部署,再走向更广的运营绑定,而不是一次性硬件销售。
[CU001, CU016, CU021, CU036, CU039, CU040]6.2 采用轨迹和具名生产证明强于留存故事
最强的客户证据不是应用商店评分或风投文章,而是一系列具体生产标记的累积。Halter 的公开材料和第三方报道如今显示,美国足迹已经远远超过少数牧场主证言,包括 11,000 英里虚拟围栏、24 个州的 250 个牧场主合作关系,以及在超大牧场面积上的早期卫星部署。具名证明集也很具体。美国一侧,材料包括围绕承载能力提升、避免围栏支出、放养率改善和节省工时的公开故事。New Zealand 一侧,证明集包括拥有数百到超过 1,500 头牛的可见奶牛场,以及研究支持的 ROI 叙事。关键判断是,这大多是生产证明,而不是试点证明。这些故事读起来像正在运行的账户,但仍没有回答其中有多少账户会扩张、留存多久,或公开样本相对整体客户基础有多具代表性。[CU004, CU005, CU006, CU007, CU008, CU009]
| 指标 | 数值 | 日期 | 来源 | 置信度 | 含义 | 缺失分母 |
|---|---|---|---|---|---|---|
| 在管牛只 | 美国、新西兰和澳大利亚共 200,000 头牛 | 2024-08-13 | Halter 美国发布稿 | 中 | 2025–2026 年加速前的早期装机基础信号 | 未按地域、流失状态或付费账户数拆分 |
| 美国虚拟围栏里程 | >11,000 miles | 2025-11-13 | BusinessWire | 中 | 说明不是试点,而是真实牧场条件下已有规模化使用 | 未说明每个牧场仍有多少英里处于活跃状态 |
| 隐含避免围栏支出 | 按每英里 $20K 估算约 $220M | 2025-11-13 | BusinessWire | 中 | 给牧场买方提供有用的客户价值叙事 | 用的是基准假设,不是客户实际回本 |
| 美国牧场主合作 | 覆盖 24 个州的 250 名牧场主;全美 300+ 个牧场已上线 | 2026 | Animal AgTech 访谈 | 中 | 美国采用广度和参考密度信号很强 | 访谈没有给出账户年限或付费 / 试用拆分 |
| 运营者 App 信号 | Google Play 下载量 5K+;59 个 iOS 评分给出 4.6/5;AppBrain 称 Android 下载量约 6.2k | 2026-05-31 访问 | Apple / Google Play / AppBrain | 高 | 确认用户端界面已上线,产品仍在持续维护 | App 信号不等于买方留存或账户级扩张 |
这些行混合了公司、媒体和应用市场来源;它们能显示采用广度,但队列耐久性仍不透明。
[CU004, CU005, CU006, CU007, CU025, CU026]| 客户 | 分群 | 部署 / 用例 | 生产使用 / 试点 | 结果 | 局限 |
|---|---|---|---|---|---|
| Grady Grissom (Colorado) | 美国肉牛 | 牧场已上线,用于放牧和载畜量管理 | 生产使用 | 案例称载畜量提升 20% | 只有官方案例页;没有独立验证或合同细节 |
| Ron Jespersen (Nebraska) | 美国肉牛 | 牧场已上线,叙事对标实体围栏建设 | 生产使用 | 案例称避免 $20K 围栏建设支出 | 只有官方案例页;无使用年限、花费或续约细节 |
| Lauren Sizemore / 加州牧场主 | 美国肉牛 | 牧场已上线,绑定避免围栏和节省用工 | 生产使用 | 案例卡片称节省数百小时;视频标题写到 $210K 和 7 英里围栏 | 案例格式混杂,但分发仍由 Halter 控制 |
| Emerald View Farm (Canterbury) | 新西兰奶牛 | 1,545 头奶牛的大型奶牛场运营使用 | 生产使用 | 仅规模就说明 Halter 能在大型奶牛场落地 | 页面上的结果指标不如美国牧场案例量化 |
| Lizzy & Cameron Te Brake (Waikato,客户案例) | 新西兰奶牛 | 奶牛场已上线,叙事围绕草场和用工 | 生产使用 | 案例包强调更多牛奶、更少精力和丘陵地区生产力 | 定性证据;没有留存或花费指标 |
这是公开样本,不是完整客户名单。相对更广的采用主张,Halter 只披露了经过筛选的少量具名账户。
[CU002, CU003, CU009, CU010, CU011, CU012]具名生产使用和可量化成果同时出现时,公开证据最强;但所有账户的持续性可见度仍然偏弱。
[CU009, CU010, CU011, CU014, CU027, CU028]6.3 耐久性、满意度与公开留存可见度的边界
公开耐久性证据,是 Halter 客户章节中最薄的一部分。可用代理存在:iOS 和 Android 应用表面确认产品仍在活跃分发和维护,应用商店也显示有非微不足道的用户互动。但这些代理不等同于买方留存、续约质量或账户级满意度。官方客户证明集也高度依赖现有使用,以及暗示持续价值的结果,这在方向上是正面的。问题是,公开客户簿止步于合同期限、续约率、NRR、GRR,甚至连基本流失数字都没有。这很重要,因为硬件加软件的现场产品可以赢下亮眼的首批部署,却在后续扩张或支持负担上表现不佳。实际结论是,Halter 有足够公开证据证明活跃客户使用,也有足够应用信号证明运营界面仍然鲜活;但没有直接尽调材料,就不足以承保耐久性。[CU025, CU026, CU027, CU028, CU029, CU038]
| 指标 / 代理信号 | 数值 | 分群 | 置信度 | 尽调要求 |
|---|---|---|---|---|
| NRR / GRR / 客户流失 | 全部客户 | 低 | 要求提供队列留存、客户流失,以及按奶牛 / 肉牛和地域拆分的扩张数据 | |
| 合同期限 / 续约期限 | 全部客户 | 低 | 要求提供标准合同期限、续约节奏和提前取消条款 | |
| iOS App 评分 | 59 个评分给出 4.6/5 | 运营者 / 用户代理信号 | 高 | 按农民 / 牧场主用户类型拆分评分,再对照账户留存 |
| Google Play 分发信号 | 5K+ 次下载 | 运营者 / 用户代理信号 | 高 | 厘清月活跃用户、付费账户和每个账户的多用户席位 |
| AppBrain Android 活跃度 | 总下载量约 6.2k;过去 30 天 460;暂无评分 | 运营者 / 用户代理信号 | 中 | 提供 WAU/MAU、流失用户,以及按活跃客户账户拆分的席位活跃度 |
应用市场和 App Store 信号是有用的运营者代理指标,但无法解释买方耐久性或收入留存。
[CU025, CU026, CU029, CU038]| 缺失指标 | 缺口为何重要 | 最佳公开代理信号 | 代理信号不能证明什么 | 精确尽调要求 |
|---|---|---|---|---|
| 续约 / 队列留存 | 耐久性是客户质量尚未解开的核心问题 | 具名生产使用案例和 App 评分 | 不能证明队列会续约或扩张 | 要求提供 NRR/GRR、队列曲线和账户级续约历史 |
| 客户集中度 | 少数牧场或农场可能贡献了过高收入占比 | 24 个州或 11,000 英里等广度代理信号 | 不显示单账户花费或头部客户敞口 | 要求按花费和部署头数列出前 20 大账户 |
| 采购漏斗健康度 | 田间产品可能卡在演示到上线之间 | 入门案例和免费一年促销 | 不能量化转化或实施负担 | 要求按分群提供线索到上线的漏斗指标和价值实现时间 |
| 已实现账户经济性 | 分群战略价值取决于实际回本,不是轶事 | 围栏成本轶事和生产力案例 | 不给出按队列拆分的实际利润率、回本或净扩张 | 要求按原型提供实际回本和 ROI |
该表把公开证据样本转成精确尽调要求,用来承销耐久性和集中度。
[CU006, CU021, CU022, CU029, CU030, CU038]6.4 扩张循环存在,但成本、动物福利和集中度盲点也存在
Halter 的公开客户动线看起来具备扩张性。公司并不把自己呈现为一次性围栏替代品;它叠加教育、公共土地合作和证明故事,指向首次部署之后更大面积、更多牛只、更多用例和更广运营采用。这是好消息。坏消息是,同一本公开客户簿留下了重大承保缺口。外部仍看不到收入集中度、头部账户依赖,或按队列实现的账户扩张。独立反向报道也让客户故事保持落地。多个外部来源继续指向高成本、电池或连接担忧、动物福利怀疑,以及有组织反对或监管拖累。这些信号没有任何一个能证明产品市场匹配弱,但合在一起说明,客户章节应落在“采用已验证,但耐久性和集中度风险未解”,而不是一个闭合的留存故事。[CU020, CU021, CU022, CU023, CU030, CU031]
| 扩张驱动因素 | 集中度风险 | 影响 | 尽调路径 |
|---|---|---|---|
| 首次部署跑通后,扩大面积或增加头数 | 不同队列里扩张常见还是少见,外部未知 | 若真实存在,可拉动强劲账户增长;若少见,则会高估 LTV | 要求提供队列级面积 / 头数扩张曲线 |
| 管理教育和顾问绑定 | 项目可能只会加深大客户粘性 | 可改善成熟牧场的留存和产品采用 | 要求提供资源项目参与者的采用率和留存结果 |
| 公地与保护合作 | 伙伴介导账户可能不均衡,并受政治条件影响 | 可加速拿下标杆客户,但会制造政策集中度风险 | 要求按伙伴主导 / 直销主导账户拆分管线和转化 |
| 高硬件 / 部署成本 | 会缩小潜在客户池或拖慢转化 | 如果只有资本最充足的牧场能采用,客户集中度风险会扩大 | 要求按分群提供销售漏斗转化和丢单原因 |
| 未公开头部账户 | 外部无法评估真实收入集中度 | 即便采用真实存在,客户质量承销仍不完整 | 要求提供头部客户收入占比和支出分布 |
公开账本证明存在多条扩张路径,但不能证明这些路径会复合成分散的高质量收入。
[CU021, CU022, CU023, CU030, CU031, CU034]公开漏斗不只是从演示到成交;它从放牧痛点出发,经过现场导入,走到正式部署和更大范围的运营铺开。
[CU019, CU020, CU022, CU031, CU034, CU036]6.5 图表
07风险
7.1 监管和动物福利正当性在 2026 年改善,但风险并未闭合
Halter 的公开记录在法律和政策动能上比一年前强得多,但关键判断并不是监管消失了,而是争论从抽象可行性转向有条件接受。Dairy Australia 的 2026 年更新显示,虚拟围栏现在已在澳大利亚全部六个奶业州合法;美国公共土地机构也公开支持包含虚拟围栏的创新试点。这是真进展。同时,UK Animal Welfare Committee 意见和 New Zealand Veterinary Association 立场声明,仍让动物福利问题保持开放,足以要求公司谨慎使用、可辩护的训练协议,以及更长期的独立证据。做承保时,这意味着 Halter 不再像被标题层面的合法性卡住,但仍暴露于政策反转、地方反对,以及未来任何会削弱动物福利正当性的事故流。[CR017, CR018, CR019, CR020, CR021, CR022]
| 规则 / 议题 | 司法辖区 | 状态 | 可能性 | 严重程度 | 缓释措施 | 剩余敞口 | 尽调路径 |
|---|---|---|---|---|---|---|---|
| 动物福利合法性和运营条件 | 英国 / 新西兰 / 更广品类 | AWC 意见和 NZVA 立场仍是当前参考点;证据基础仍有争议 | 中 | 高 | 已发布福利指南、培训协议、更低能量脉冲设计和外部研究库 | 一旦出现重大事故流或关键长期研究,合法性风险可能迅速重启 | 获取事故历史、独立长周期研究,以及上线以来任何监管或兽医投诉 |
| 奶牛州虚拟围栏合法性 | 澳大利亚 | Dairy Australia 称,六个奶牛州目前均已合法 | 中低 | 中 | 公司如今可引用核心市场 2026 年明确扩大合法性的事实 | 州层面合法不保证每种生产场景里的客户经济性或接受度 | 要求提供逐市场合规备忘录,以及按州或物种划分的使用限制 |
| 公地许可和放牧政策依赖 | 美国联邦土地 | USDA/DOI 鼓励创新;BLM 放牧规则仍在推进,且对评论敏感 | 中 | 高 | 机构沟通、伙伴资助试点、牧场主咨询、AUM 无净损失表述 | 公地准入仍受政策中介,容易受程序或政治变化影响 | 审查许可依赖、评论敞口,以及 BLM 规则制定变化时的应急计划 |
| 保护 / 准入合作 | 加州 / 新墨西哥州 / 美国西部 | Halter 已发布活跃合作公告,但执行仍依赖对手方 | 中 | 中高 | 基金会资金和本地伙伴关系提供近期证据点 | 如果机构、基金会或本地团体降低项目优先级,项目可能停摆 | 要求提供项目经济性、已签承诺、续约条款和按伙伴划分的集中度 |
| 公司与跨司法辖区治理面 | 新西兰 / 美国 | 新西兰实体摘录显示美国最终控股公司,但公开治理细节有限 | 中 | 中 | 正式实体状态有效,且可公开查询 | 董事会、委员会和公司间细节仍大多不公开 | 要求提供完整架构图、董事名单、保留事项和公司间转让条款 |
覆盖范围有限,只包括截至 2026-05-31 可见的公开法律、监管和政策材料;私下投诉、威胁性争议和内部合规备忘录无法从外部观察。
[CR017, CR018, CR021, CR022, CR023, CR024]Halter 主要风险集群的相对发生概率、影响、缓释成熟度和剩余严重度。
标签是基于所引公开证据的判断,而非内部损失模型。
[CR003, CR006, CR010, CR017, CR022, CR023]7.2 运营风险已从“能不能跑”转为“能否低成本、可重复地规模化运行”
运营风险故事已经不再是简单的技术就绪问题。多个来源如今支持虚拟围栏能在真实牧场环境中运行,尤其是在动物经过训练之后。更难的承保问题,是成本调整后的可重复性。North Dakota State University 的定价框架、Farm Progress 的成本怀疑,以及 New Zealand 需要折扣,都指向一个仍要求客户吞下可观运营和融资负担的品类。卫星连接减少了一个障碍,但它把塔台依赖换成了新的外部基础设施依赖,并没有抹掉现场维护或可靠性风险。Colorado State 的真实经验——项圈脱落、受潮问题和电池失望——很重要,因为它显示采用风险现在落在部署质量的混乱中段,而不只是演示能否打动人。公司可能领先,但卖的仍是现场系统,不是无摩擦软件。[CR004, CR005, CR006, CR007, CR008, CR009]
| 失效模式 | 可能性 | 严重程度 | 缓释成熟度 | 剩余敞口 | 未解决缺口 |
|---|---|---|---|---|---|
| 农场级经济性仍太重,难以支撑大品类采用 | 中高 | 高 | 中 — 已有降价和产品 ROI 叙事,但硬性的单位经济性证据仍稀少 | 早期采用者饱和后,即便中等采用摩擦也会拖慢增长 | 未公开回本分布、按价格队列拆分的流失,或折扣对毛利率的影响 |
| 硬件可靠性、维护和电池表现削弱田间信任 | 中 | 高 | 中低 — 研究和产品迭代可见,但公开事故数据缺失 | 牧场上的可见故障会伤害留存和口碑,而这个品类依赖信任 | 未公开 MTBF、更换节奏或按严重程度加权的事故历史 |
| 连接能力改善后,第三方基础设施风险仍在 | 中 | 中高 | 中 — 卫星在部分用例里移除了基站塔,但场外依赖上升 | 远程服务质量仍可能受外部连接经济性或中断约束 | 未公开 SLA、回退流程,或按地形划分的覆盖质量直方图 |
| 客户培训和变革管理质量因牧场环境而异 | 中 | 中高 | 中 — 推广材料和福利材料存在,但未披露标准化田间结果 | 入门培训弱的牧场即便技术原理可行,也可能表现不佳 | 未公开按账户类型划分的上线时间、培训成功率或牲畜逃逸率 |
| 竞争性价格压力压缩品类经济性 | 中 | 中 | 中低 —— Halter 仍在拿份额,但竞品架构依然可行 | 要继续拿份额,可能需要比当前叙事暗示的更低价格或更高支持强度 | 公开证据未披露实际转化率、折扣后的毛利率响应,或按对手拆分的赢单 / 输单 |
各行综合公司、推广机构、田野研究和媒体来源;描述的是公开运营风险,而不是经审计的工程 KPI。
[CR004, CR005, CR006, CR007, CR008, CR009]运营、监管和合作伙伴风险如何传导到增长质量、烧钱速度和估值支撑。
[CR005, CR010, CR015, CR022, CR023, CR030]7.3 增长故事依赖外部交易对手,不只是 Halter 的项圈和应用
Halter 2026 年叙事的一个显著特征,是它对外部参与者的依赖程度。公共土地增长依赖联邦机构、保护组织和基金会资金;卫星覆盖依赖 Starlink 类基础设施;顾问采用循环依赖咨询顾问和兽医;2026 年运营计划还依赖一轮大规模招聘同时在多个地理市场执行到位。这些依赖本身都不致命,但合在一起形成的风险画像,更像基础设施编排,而不是纯软件铺开。Companies Office 摘录也显示,公司是跨司法辖区结构:美国最终控股公司叠在 New Zealand 实体之上,而公开治理可见度仍然偏薄。对投资人来说,这意味着公司可以赢下品类领导地位,但如果伙伴经济性、人才吸收或治理成熟度落后于扩张节奏,仍可能令人失望。[CR003, CR025, CR026, CR027, CR028, CR029]
| 依赖项 | 对手方 | 作用 | 集中度 | 失效情景 | 严重度 | 缓释措施 | 剩余敞口 |
|---|---|---|---|---|---|---|---|
| 远程连接栈 | Starlink / 卫星基础设施 | 2026 年卫星发布后,让离网控制和遥测跑通 | 远程牧场用例中集中度高 | 服务、成本或接入变化削弱无塔部署价值 | 高 | 混合架构经验、产品迭代,以及按地形选择性部署 | 影响大,因为远程扩张现在依赖 Halter 不控制的基础设施 |
| 公共土地商业化 | USDA / BLM / 许可框架 | 为联邦土地牧场打开准入并提供验证 | 中,但战略意义高 | 规则调整、许可摩擦或政治变化拖慢公共土地切入点 | 高 | 机构沟通、牧场主咨询、合作伙伴出资试点 | 仍然重要,因为公共土地使用受政策调节,不是纯商业决策 |
| 保护资金和准入项目 | Foundation for America’s Public Lands 与地方保护组织 | 为试点出资,并为多方参与项目背书 | 中 | 项目转成稳定商业账户前,资金或利益方支持退坡 | 中高 | 案例研究管线多元,项目形式不止一种 | 公开证据尚未显示续约韧性或资助转付费转化 |
| 顾问与咨询渠道 | 乡村专业人士、兽医、顾问 | 影响农场采用和工作流深度 | 中 | 顾问参与无法规模化,或跨地区执行不一致 | 中 | 官方顾问工具和公司可控产品界面 | 公开资料未拆分渠道来源管线,也未披露受渠道影响账户的留存 |
| 后续融资可得性与叙事支撑 | 领投方 / 风险投资市场 | 设定继续扩张与估值支撑预期 | 完成估值 $2B 融资后,象征意义高 | 增长或经济性低于预期,被迫以惩罚性条款重置融资 | 高 | 大额现金注入、可见动能和品类龙头叙事 | 公开资料仍缺足够经济性数据,难判断融资故事有多稳 |
本登记表强调影响 GTM 执行的外部对手方和系统,而不只看已签商业供应商。
[CR003, CR022, CR023, CR025, CR026, CR027]| 职位 / 职能 | 依赖或缺口 | 可能性 | 严重度 | 缓释措施 | 尽调路径 |
|---|---|---|---|---|---|
| 创始人 / CEO 领导力 | 公开叙事、产品愿景和投资人信心仍高度绑定 Craig Piggott | 中 | 高 | 更广的资本基础和逐渐成熟的组织,可随时间降低单人依赖 | 要求提供继任计划、授权后的运营节奏,以及创始人之外的客户 / 投资人接触点 |
| 客户成功与现场运营 | 超过 200 人的招聘潮必须转化为跨区域一致的部署质量 | 中高 | 高 | 招聘计划明确包括客户岗位和产品改进 | 要求提供产能模型、实施人员配比,以及新现场员工达产周期 |
| 工程与产品执行 | 公司一边推出重大升级,一边扩张地域和动物健康范围 | 中 | 中高 | 新资金和聚焦产品的招聘 | 要求提供路线图治理、发布质量指标,以及可靠性与新模块之间的 backlog 拆分 |
| 治理 / 董事会监督 | 相比规模和估值,公开实体记录只露出很薄的治理信息 | 中 | 中 | 已有注册公司状态和公开申报 | 要求提供 NZ 与美国实体的董事会构成、委员会授权和保留事项 |
公开材料确认了规模和招聘野心,但管理带宽、领导层冗余和治理成熟度披露仍不够,无法关闭执行风险。
[CR003, CR030, CR031, CR033, CR034]如今位于 Halter 和可持续执行之间的关键外部系统和交易对手。
[CR022, CR027, CR028, CR029, CR031, CR033]7.4 剩余风险只有在公司持续把政策、产品和融资动能转化为更低风险的证明时,才具备可投资性
重要的组合结论是,只有几件事同时继续走对,Halter 的风险栈才算可控。公司确实有势能:资本更多、合法性范围更广、连接能力更好,生产者采用也看得见。但公开记录在最关乎下行保护的问题前仍然断开:真实事故率、公有土地依赖的经济性、合作伙伴集中度,以及 200 多人的招聘计划能否撑住客户交付质量。由此形成的风险画像是:缓释手段存在,但成熟度还不够,不足以把剩余敞口降到可以忽略。投资委员会因此不能用二元框架,而要用条件框架。如果卫星铺开确实降低服务摩擦,降价不意味着单位经济性走弱,动物福利或监管反对也保持可控,风险调整后的投资案例就会改善;如果其中任何一个变量反转,估值支撑会很快变弱。换句话说,Halter 已经跨过一个门槛:执行质量现在比叙事质量更重要。也正是在这里,今天要让机构投资核算信得过价格,尽调纪律必须先收紧。[CR001, CR002, CR006, CR010, CR015, CR017]
| 风险 | 可监控触发因素 | 阈值 / 事件 | 行动含义 |
|---|---|---|---|
| 动物福利正当性恶化 | 独立投诉、反向研究或监管介入 | 任何正式限制、可信的福利事件集群,或显示重大伤害的长期研究 | 暂停承销判断,直到事件数据、整改方案和监管路径清晰 |
| 降价掩盖疲弱经济性 | 折扣节奏和回本叙事 | 再次大幅重定价,却没有服务成本下降或转化改善的配套证据 | 将投资逻辑从增长质量转为增长承压,并收紧入场纪律 |
| 卫星依赖不及预期 | 远程地形覆盖可靠性和客户证言 | 2026 年发布后出现明显故障模式、SLA 薄弱,或远程牧场采用停滞 | 下调对美国扩张切入点的信心,并折减估值假设 |
| 公共土地策略停滞 | 试点转商业转化和政策连续性 | BLM 或合作伙伴项目未能续约,或规则变化给获许可方带来新摩擦 | 将公共土地验证视为非核心,把尽调重新聚焦纯商业账户 |
| 招聘潮跑得比运营系统快 | 现场运营生产率和客户质量信号 | 2026-2027 年扩张期间出现实施延迟、明显支持压力或离职率上升 | 在运营指标证明相反之前,假设增长更慢、烧钱更高 |
触发因素选择以尽调和 IC 跟进中的可观察性为准;它们不是精确内部公司 KPI。
[CR003, CR006, CR010, CR017, CR022, CR023]08估值
8.1 公司基本面确实成立,但当前估值已经计入大量成功预期
Halter 的估值章节从一个不舒服但有用的拆分开始。投资逻辑并非凭空想象:公司已有大额融资、可见的装机基数验证、优质投资人阵容、足够强的产品野心,以及足以说明它不只是新奇硬件生意的客户结果证据。反向逻辑同样真实:公开材料在高估值通常决胜的位置仍然很薄,包括经审计 ARR、毛利率、留存、股权结构条款,以及部署强度归一化之后的队列经济性。因此,当前估值更容易让人欣赏,却不容易让人按机构标准核算通过。公开来源足以解释为什么公司相对普通农机业务值得一个溢价叙事,但还不足以支持把纯 SaaS 式倍数照单全收。2026 轮融资因此像是一项高质量资产,只是价格已经假设了强执行会持续。[CV001, CV002, CV003, CV004, CV005, CV006]
| 建议 | 信心 | 风险评级 | 估值立场 | 决策含义 |
|---|---|---|---|---|
| 仅在价格纪律下观察 / 接触 | 中 | 高 | 相对公开 $2B 标记偏高 / 已打满 | 公司有吸引力,但没有经审计经济性、股权结构表透明度和结构性下行保护,不应按完整标记承销 |
此行仅反映基于公开证据的承销判断;私人资料室可能显著改变立场,方向可上可下。
[CV001, CV011, CV038, CV044, CV045]| 论点 | 公开资料显示 | 哪些证据会改变判断 |
|---|---|---|
| 投资逻辑:有真实验证的品类龙头 | 装机基础、大额融资、客户使用和产品宽度说明 Halter 不只是概念公司 | 经审计留存和毛利率数据会把这一判断从可信变成可承销 |
| 投资逻辑:高溢价倍数有机会兑现 | 投资人和媒体越来越把 Halter 定义为畜牧运营的基础设施 / 软件 | 证明 ARR 质量和毛利率更接近软件,而不是现场服务 |
| 反向逻辑:估值跑在披露前面 | 公开 ARR 和价格代理指标都粗糙,股权结构和优先权条款仍隐藏 | 完整财务包和清算瀑布,证明当前标记不会伤害普通股 |
| 反向逻辑:硬件加服务经济性限制上行 | 降价、现场成本和品类摩擦显示,经常性经济性质量可能低于纯 SaaS | 持续队列证据显示定价权、低流失和可控支持负担 |
各行有意把公司质量与入场价格质量分开;两者可以同时成立。
[CV005, CV015, CV016, CV018, CV019, CV037]规模证据、经济性不透明、公开可比公司边界和风险如何汇合成价格克制的建议。
[CV005, CV011, CV032, CV037, CV038, CV044]8.2 公开可比公司给足额追价设置了硬上限,不能随意越过
最干净的公开估值校验不是完美可比公司,而是一个有边界的可比区间。Deere 和 AGCO 展示了公开市场愿意为硬件占比高的农业敞口支付什么价格。Trimble 展示了工作流更重的工业技术如何拿到更好倍数。Merck 提供了大规模动物健康参照,但仍不是完美类比,因为 Vence 只是一个庞大企业里的一小块。即便有这些保留,结果方向上仍然醒目:选取的公开公司大约在 0.7x 到 4.6x 收入之间,而 Halter 自己的公开 ARR 估计值意味着当前估值约为 20x 到 29x ARR。这个差距不能证明本轮定价错误,但证明投资人买的是一个非常具体的未来——Halter 复利成长为定义品类的畜牧运营系统,而不是最后落成一个资本密集型小众工具:采用不错,利润率中等。[CV007, CV008, CV009, CV010, CV011, CV020]
| 可比对象 | 指标 | 倍数 / 估值 / 状态 | 参考价值 | 局限 |
|---|---|---|---|---|
| AGCO | US$8.13B 市值 / US$11.662B TTM 收入 | ~0.7x 市值 / 收入 | 有用的农业硬件偏重下限可比对象 | 太像 OEM,经常性软件属性弱于 Halter |
| Deere | US$146.44B 市值 / US$44.433B TTM 收入 | ~3.3x 市值 / 收入 | 说明高端工业 / 农业敞口的交易倍数可明显高于纯机械落后者 | 仍是比 Halter 宽得多的工业平台 |
| Trimble | US$13.14B 市值 / US$3.601B TTM 收入 | ~3.65x 市值 / 收入 | 这组里最像工作流驱动的工业科技可比对象 | 非畜牧原生,且已是成熟上市公司 |
| Merck | US$293.21B 市值 / US$64.235B TTM 收入 | ~4.56x 市值 / 收入 | 可作为动物健康规模参考,也提供战略收购方视角 | Vence 只是 Merck 很小一部分,该倍数作为代理信号噪声很大 |
覆盖范围有意保持部分:这组提供方向性的公开边界,不是 Halter 一一对应的完美同行篮子。
[CV020, CV021, CV022, CV023, CV024, CV025]将不同 ARR 倍数套用在公开中点 ARR 估计 US$85M 上得到的隐含股权价值。
仅用公开的低信心中点 ARR 估计展示敏感性;经审计 ARR 和毛利率明确后,真实估值可能大幅变化。
[CV010, CV011, CV039, CV040, CV041]8.3 乐观、基准、悲观情景区间很宽,因为缺失数据正落在估值精度最该来自的位置
这里情景区间宽,不是分析弱,而是经济性证据缺失后的诚实结果。乐观情景能成立,是因为 Halter 确实已有规模、国际扩张野心和产品宽度的证据,也因为当增长仍然陡峭时,私募市场有时会给品类领导者远高于上市工业可比公司的估值。悲观情景也能成立,因为公开记录不断提醒我们:这是一套硬件加服务的系统,卖给成本敏感的运营环境,同时还带着动物福利、执行和价格敏感性问题。基准情景因此落在当前估值附近,但并不会自动高于当前估值。关键结论是:公开证据足以让 Halter 留在可投资范围内,但若没有能显著提高留存、毛利率和融资栈质量信心的私有数据,还不足以从当前入场点核算出可观上行。[CV011, CV012, CV013, CV037, CV038, CV039]
| 情景 | 假设 | 估值 / 回报逻辑 | 关键风险 | 概率信号 |
|---|---|---|---|---|
| 乐观 | ARR 明显跑赢公开粗略估计;卫星扩大美国可触达市场;新市场转化良好;留存和毛利率证明高质量溢价 | 结果区间 US$2.6B-US$3.4B;若以 US$2.0B 入场,稀释前大约对应 1.3x-1.7x 总价值提升 | 执行落空、事件风险或毛利率失望会击穿溢价情景 | 有可能,但需要多项目前未公开的私人证据支撑 |
| 基准 | 增长仍强,但经济性更像高端工业 / 垂直软件混合体,而不是纯 SaaS 异常值 | 区间 US$1.6B-US$2.2B;除非私有指标明显好于公开代理指标,否则相对当前标记上行有限 | 入场纪律很关键,因为平庸经济性可能抹掉大部分上行 | 当前公开证据最能支持 |
| 悲观 | 价格敏感、品类摩擦或服务强度暴露出业务偏硬件,经常性质量更低、扩张更弱 | 区间 US$0.9B-US$1.3B;若以 US$2.0B 入场,意味着显著下行,且大概率痛苦地重置融资 | 稀释、优先权包袱或公共土地 / 动物福利受挫,可能加速这一情景 | 可信,因为公开记录恰好在通常支撑高溢价轮次的指标上仍然很薄 |
区间是作者基于公开可比公司、价格代理指标和当前私有标记设定的情景带;不是 DCF 输出。
[CV011, CV039, CV040, CV041]基于公开证据的估值区间,对比当前 US$2.0B 估值标记。
区间来自公开可比公司边界、已披露规模和公开风险因素下的情景研判。它们不是 DCF 输出,也未计入优先股堆叠影响。
[CV039, CV040, CV041, CV044]8.4 建议:保持接触,但要求价格纪律和私有尽调证据
最终估值立场不是否定公司,而是拒绝把公司质量和入场质量的确定性混为一谈。Halter 已经有足够公开证据,值得认真尽调:真实客户、真实融资、竞争上确实有分量,以及足够支撑持续关注的运营野心。但公开材料仍没有展示股权结构表、优先股堆叠、经审计 ARR、毛利率或分细分的回本数据,投资人还不能放心地把 $2B 估值视为已经核算通过。决策问题因此从「这家公司有意思吗?」变成「什么条款下才有吸引力?」合理姿态是观察,或在严格价格纪律下接触,而不是按完整公开估值领投或追价。退出路径上,下一轮私募融资或战略方兴趣比近期 IPO 更可信。最快改变判断的不是另一篇叙事文章,而是硬的私有证据:公司增长具备软件式质量,而不只是软件式叙事,尤其是在留存、毛利率和稀释方面。[CV014, CV042, CV043, CV044, CV045, CV046]
| 触发因素 | 阈值 | 对投资逻辑的传导 | 行动含义 |
|---|---|---|---|
| 经审计 ARR 或毛利率明显低于溢价预期 | ARR 低于公开粗略区间,或毛利率结构性远低于软件式预期 | 削弱「当前标记买到高质量经常性经济性」这一论点 | 不按完整公开标记出价或支持 |
| 股权结构表条款对投资人不友好 | 清算优先权、参与分配或稀释堆叠显著限制普通股结果 | 标称估值不再是回报分析的合适框架 | 重新定价,或要求结构性保护 |
| 当前折扣后价格敏感性继续恶化 | 进一步大幅重定价,或打折后转化仍弱 | 显示采用受成本约束,高溢价倍数被高估 | 转向悲观情景承销 |
| 运营 / 动物福利挫折冲击扩张质量 | 事件集群、公共土地摩擦,或招聘潮带来的服务压力 | 将增长叙事转化为估值下调风险 | 暂停或收紧敞口,直到韧性被证明 |
这些是面向 IC 的否决触发因素,不是管理层 KPI;它们聚焦最快打破估值支撑的事项。
[CV013, CV038, CV039, CV044]| 主题 | 缺失证据 | 重要性 | 负责人或尽调路径 |
|---|---|---|---|
| 经审计 ARR / 收入质量 | 经审计 ARR、收入结构、毛利率、流失率和按队列留存 | 这是验证或否定高溢价倍数情景最快的路径 | 财务团队 / 资料室请求 |
| 股权结构表和清算堆叠 | 优先权条款、稀释、期权池和瀑布示例 | 即使标称估值看起来稳定,回报也可能不及预期 | 公司法律顾问和 CFO |
| 分部经济性 | 按 NZ 奶牛、美国肉牛和新地域拆分 CAC、回本周期、支持成本、扩张和烧钱 | 显示规模是在复利增长,还是变得更贵 | 收入运营和 FP&A |
| 退出准备度 | 董事会成熟度、审计准备度、政策历史和分部披露包 | 明确下一步更可能是 IPO、战略出售,还是新一轮私募 | 管理团队和投行 |
这些要求有意收窄到少数最能改变估值舒适度的缺失项。
[CV014, CV038, CV042, CV043, CV045]IC 口径评分:从证据、经济性、风险和估值匹配度评估 Halter。
评分是作者仅基于公开证据按 1-10 分做出的判断,分数越高越好。整体读法:资产有潜力,入场价偏高。
[CV005, CV018, CV032, CV038, CV042, CV044]免责声明
本报告是基于公开证据的尽调快照,不构成投资建议。重要的财务、法律、技术和合同事实仍未公开;做出任何投资决定前,应直接向管理层和一手文件核验。
证据索引
| 编号 | 陈述 | 可信度 | 来源 |
|---|---|---|---|
| CO001 | Halter was founded in 2016 in New Zealand by Craig Piggott, with investor and alumni sources also identifying Max Olson as a co-founder. | 高 | SO015, SO016, SO017 |
| CO002 | Max Olson is publicly described as Halter’s co-founder and first CTO, but he now appears outside the current operating roster at Vessev. | 中 | SO015, SO017 |
| CO003 | Halter is headquartered in Auckland, New Zealand, with Australian operations in Melbourne and a U.S. office in Colorado. | 高 | SO002, SO003, SO008 |
| CO004 | Halter describes itself as the leading operating system to run a dairy or beef farm. | 高 | SO001, SO004 |
| CO005 | Halter’s product system combines a solar-powered smart collar, farm connectivity, and an app that lets farmers fence, move, and monitor cattle remotely. | 高 | SO001, SO004 |
| CO006 | Halter’s collars use audio cues and gentle vibrations to contain and herd cattle within virtual boundaries. | 高 | SO002, SO003 |
| CO007 | In April 2026 Halter launched direct-to-satellite connectivity for beef collars, removing the need for cell towers or on-ranch infrastructure. | 中 | SO008 |
| CO008 | Halter serves more than 2,000 farmers and ranchers across New Zealand, Australia, and the United States. | 高 | SO002, SO003, SO011 |
| CO009 | Halter had sold one million solar-powered collars by the March 2026 Series E cycle. | 高 | SO002, SO003, SO011 |
| CO010 | Official Series E materials say American ranchers had built 60,000 miles of virtual fencing since Halter launched in the U.S. in 2024. | 高 | SO002, SO003 |
| CO011 | Halter’s 2026 expansion plan names the United Kingdom, Ireland, Canada, and parts of South America as next markets beyond NZ, Australia, and the U.S. | 高 | SO002, SO009, SO011 |
| CO012 | Halter employed over 400 people across New Zealand, Australia, and the U.S. in its March 2026 official funding announcement. | 中 | SO002 |
| CO013 | Multiple March 2026 articles say Halter planned to hire more than 220 people, focused on product, engineering, and customer roles at Auckland HQ. | 中 | SO010, SO011 |
| CO014 | Craig Piggott remains Halter’s founder and CEO in 2026 and is still the company’s primary external spokesperson. | 高 | SO002, SO007, SO009 |
| CO015 | Halter’s careers page highlights internally promoted leaders across product engineering, marketing, strategy, and operations, indicating a growing management bench below the founder. | 中 | SO015 |
| CO016 | Public materials show Max Olson as a founder-alumnus rather than a current listed Halter executive. | 中 | SO015, SO017 |
| CO017 | The reviewed public source set does not disclose Halter’s board composition, committee structure, or investor control rights in enough detail to map governance formally. | 低 | |
| CO018 | Halter’s June 2025 Series D announcement described the round as $165 million and valued the company at NZ$1.65 billion or about $1 billion USD. | 中 | SO004 |
| CO019 | AgFunderNews reported the same 2025 Series D as a $100 million raise led by BOND with support from NewView, Bessemer, DCVC, Blackbird, Icehouse, and Promus. | 中 | SO005 |
| CO020 | Halter’s March 2026 Series E raised $220 million at a $2 billion valuation. | 高 | SO002, SO003 |
| CO021 | Founders Fund led the 2026 Series E, with Blackbird, DCVC, Bond, Bessemer, NewView, Ubiquity, Promus, and Icehouse also participating. | 高 | SO002, SO003 |
| CO022 | Public March 2026 coverage implies Halter’s headline U.S.-dollar valuation roughly doubled from about $1 billion in 2025 to $2 billion in 2026. | 高 | SO004, SO009, SO010 |
| CO023 | Series E materials say Founders Fund first provided early capital to Halter in its Series A round in 2017. | 高 | SO002, SO003 |
| CO024 | TechCrunch reported in April 2026 that Halter had raised roughly $400 million in total, but exact lifetime funding still depends on paid databases and company summaries for the earliest rounds. | 中 | SO007 |
| CO025 | In August 2025 Halter announced a partnership with the Bureau of Land Management and the Foundation for America’s Public Lands that included $2.7 million of support for ranchers using Halter on BLM-managed land. | 高 | SO013, SO014 |
| CO026 | TIME named Halter’s app and collar in its 2025 Best Inventions special mentions. | 中 | SO012 |
| CO027 | An AgFirst and Transform Agri study of 10 high-performing New Zealand dairy farms found average gains of 13.2% in profit before tax, 8.9% in pasture harvested, and 9.5% in milk solids per hectare. | 高 | SO023, SO024, SO025 |
| CO028 | Halter’s ROI materials cite named farms such as Willowcliff, Harakeke, and Grassmere achieving double-digit EBIT or labor-efficiency gains after adopting the system. | 中 | SO023 |
| CO029 | Halter’s welfare charter said that approximately 700,000 animals and hundreds of farms were using the system as at January 2026. | 中 | SO018 |
| CO030 | Halter’s welfare charter says the maximum pulse strength is 0.45 joules and that roughly 80–98% of cows at pasture receive pulses below 0.2 joules depending on season. | 中 | SO018 |
| CO031 | Halter’s cited independent Tasmanian research says cows learned to respond to sound cues within a day and showed no evidence of increased chronic stress versus conventional management. | 中 | SO019 |
| CO032 | The New Zealand Veterinary Association says current virtual-fencing literature is mostly short-term and often industry-funded, leaving limited independent long-term evidence. | 中 | SO020 |
| CO033 | NZVA supports responsible virtual fencing adoption only with clear safeguards, operator training, monitoring systems, and regulatory oversight. | 中 | SO020 |
| CO034 | Dairy Australia said in January 2026 that virtual fencing and herding had become legal in all six Australian dairying states after late-2025 and early-2026 legislative changes. | 中 | SO021 |
| CO035 | Halter looks founder-led with improving managerial depth, but public materials do not yet make governance sufficiently transparent for outside mapping of checks and succession. | 中 | SO015 |
| CO036 | TechCrunch described Halter as an outlier in an agtech sector that has recently slumped as startups struggle with farmer adoption and high operational costs. | 中 | SO007 |
| CO037 | CMK’s 2026 farm-finance commentary says advanced Halter pricing sits at about NZ$9.90 per cow per month, with additional infrastructure such as drafting gates often required to maximize returns. | 中 | SO022 |
| CO038 | The same CMK analysis argues wearables amplify good farm management rather than fixing weak underlying systems, which tempers simplistic ROI assumptions. | 中 | SO022 |
| CO039 | Halter’s positioning increasingly emphasizes software, sensors, and AI inside livestock operations rather than a narrow agtech label. | 高 | SO003, SO006 |
| CO040 | TechCrunch reported Halter was operating in 22 U.S. states by April 2026. | 中 | SO007 |
| CO041 | SmartCompany said American farmers and ranchers had built close to 100,000 kilometres of virtual fencing since Halter’s 2024 U.S. launch. | 中 | SO009 |
| CO042 | Icehouse Ventures said Halter had already reached more than 500,000 cattle across 1,000-plus farms and 200-plus employees by its 2025 unicorn milestone. | 中 | SO016 |
| CO043 | The BLM partnership named Cotoni-Coast Dairies National Monument in California as the first ranching site to use the funding support, giving Halter a tangible public-lands showcase. | 中 | SO013 |
| CM001 | Halter should be analyzed against the market for pasture-based cattle operating systems rather than a generic agritech category. | 高 | SM015, SM016, SM017 |
| CM002 | Included market spend for Halter covers collars, software, connectivity, and farm workflows for fencing, shifting, pasture, and animal monitoring. | 高 | SM015, SM016, SM017 |
| CM003 | Virtual fencing does not replace all permanent fence, corrals, or high-pressure handling infrastructure. | 中 | SM010, SM011 |
| CM004 | LIC says New Zealand farms just under 5 million dairy cows in about 10,000 dairy herds. | 中 | SM002 |
| CM005 | Stats NZ reported 5.75 million dairy cattle and 3.833 million beef cattle in New Zealand at 30 June 2025. | 中 | SM001 |
| CM006 | USDA NASS reported 86.2 million cattle and calves in the United States on 1 January 2026, including 27.6 million beef cows and 9.57 million milk cows. | 高 | SM004, SM005 |
| CM007 | MLA projected the Australian cattle herd at 31.052 million head for 2026. | 中 | SM006 |
| CM008 | Halter’s one million installed collars sit against a much larger cattle base across its live markets. | 高 | SM005, SM006, SM017 |
| CM009 | A core current-geography market lens using New Zealand cattle, the Australian herd, and U.S. beef cows yields about 68.2 million head. | 中 | SM001, SM005, SM006 |
| CM010 | A broader current-geography ceiling using all U.S. cattle and calves yields about 126.8 million head. | 中 | SM001, SM005, SM006 |
| CM011 | One million collars implies roughly 1.5% penetration of the core cattle lens and less than 1% of the broader ceiling. | 中 | SM005, SM006, SM017 |
| CM012 | Public-land stewardship, wildfire-fuel management, and riparian protection widen the economic buyer conversation beyond pure farm convenience. | 高 | SM011, SM012 |
| CM013 | The economic buyer for virtual fencing is usually the farm or ranch owner or the operating entity that controls the grazing budget. | 中 | SM015, SM016, SM010 |
| CM014 | Adoption often involves additional stakeholders such as agencies or stewardship partners when public-land or environmental outcomes matter. | 中 | SM012, SM013, SM014 |
| CM015 | Halter’s dairy workflow centers on pasture allocation, heat detection, health monitoring, and workload reduction. | 中 | SM016 |
| CM016 | Halter’s beef workflow centers on remote coverage, exclusion zones, rotational grazing, and more effective use of rough or steep ground. | 中 | SM015 |
| CM017 | The status quo substitute for Halter remains physical fencing plus labor-intensive cattle movement using people, vehicles, and dogs. | 高 | SM010, SM011, SM012 |
| CM018 | WireMesh and Farm Progress both frame the current U.S. virtual-fencing category around four main vendors: Halter, Vence, Gallagher eShepherd, and Nofence. | 中 | SM010, SM011 |
| CM019 | Datamars, Cowlar, and Moocall are adjacent livestock-monitoring products rather than full virtual-fencing systems. | 中 | SM021, SM022, SM023 |
| CM020 | Labor savings from fence setup, checks, and herd movement are one of the clearest value drivers in current virtual-fencing coverage. | 高 | SM010, SM011, SM012 |
| CM021 | Pasture utilization, rotational grazing, and environmental exclusion zones are core productivity drivers for virtual fencing adoption. | 高 | SM011, SM012 |
| CM022 | Dairy Australia said virtual fencing and herding had become legal in all six Australian dairying states by January 2026. | 中 | SM013 |
| CM023 | Labor scarcity and aging agricultural workforces remain a structural tailwind for automation in cattle systems. | 高 | SM008, SM025 |
| CM024 | The University of Arizona estimated first-year virtual-fencing costs at roughly $175 to $400 per cow depending on vendor and herd size, with year-two-and-beyond costs around $80 to $130 per cow. | 中 | SM009 |
| CM025 | Farm Progress quoted Halter’s U.S. pricing at $72 per collar per year plus a one-time $4,500 per tower with a required three-year commitment. | 中 | SM010 |
| CM026 | CMK’s New Zealand commentary puts advanced Halter pricing at about NZ$9.90 per cow per month. | 中 | SM024 |
| CM027 | Connectivity, battery life, collar handling, and training time remain material adoption constraints across virtual-fencing systems. | 中 | SM010, SM011 |
| CM028 | Even enthusiastic trade explainers say virtual fencing works best as a management layer and still relies on perimeter fencing and handling infrastructure for safety. | 中 | SM010, SM011 |
| CM029 | NZVA says longer-term independent evidence on virtual-fencing welfare remains limited and that strong safeguards and oversight are still necessary. | 中 | SM014 |
| CM030 | UC Davis CLEAR frames virtual fencing as a tool that can improve soil health, protect water quality, reduce erosion, and support wildfire resilience. | 中 | SM012 |
| CM031 | Vendor architectures differ materially: Gallagher offers outright hardware purchase with cellular or LoRa options, Vence uses base stations, Nofence uses collars plus app access without base stations, and Halter has historically used towers while adding satellite. | 高 | SM018, SM019, SM020, SM010 |
| CM032 | Halter’s direct-to-satellite launch specifically reduced a prior adoption barrier for remote beef operations that lacked tower-friendly connectivity. | 高 | SM015, SM017 |
| CM033 | Current U.S. market coverage still describes virtual fencing as moving from pilots toward early adoption rather than as a mature default system. | 中 | SM010, SM011 |
| CM034 | New Zealand’s dairy sector is a natural home market for Halter because it is large, pasture-based, export oriented, and organized around grazing efficiency. | 高 | SM002, SM003 |
| CM035 | Australian cattle adoption economics sit inside a broader sector shaped by strong export orientation, labor constraints, and biosecurity regimes. | 高 | SM006, SM008, SM013 |
| CM036 | Even conservative headcount math leaves Halter with very large runway relative to its current installed base. | 高 | SM005, SM006, SM017 |
| CM037 | No public source in the reviewed set isolates a clean spend-based SAM or SOM for Halter’s exact bundle, so any narrow market estimate remains analytical rather than directly sourced. | 低 | |
| CM038 | Current evidence supports real demand for virtual fencing, but payback still varies materially by herd size, terrain, management intensity, and the mix of labor and fencing costs. | 高 | SM009, SM010, SM011 |
| CP001 | Buyers evaluating Halter can solve the same grazing-control job through direct virtual-fencing platforms, adjacent monitoring tools, or the status quo of physical fence plus manual labor. | 高 | SP011, SP013, SP015 |
| CP002 | South Dakota State University says producers can currently purchase virtual-fence systems from Vence, Nofence, eShepherd, and Corral Technologies. | 中 | SP011 |
| CP003 | WireMesh says the 2026 category conversation is dominated by Vence, Gallagher eShepherd, Nofence, and Halter. | 中 | SP016 |
| CP004 | Halter’s beef product markets virtual fencing and shifting for beef with direct satellite connectivity and no towers required. | 高 | SP001, SP026 |
| CP005 | Halter’s dairy product bundles virtual fencing, shifting, pasture management, and heat and health monitoring. | 中 | SP002 |
| CP006 | Gallagher eShepherd uses solar-powered GPS neckbands and an app to fence, track, and move cattle. | 中 | SP004 |
| CP007 | Gallagher eShepherd offers either cellular connectivity or LoRa base stations for remote terrain with limited coverage. | 中 | SP004, SP014 |
| CP008 | Merck positions Vence as a cattle virtual-fencing system that controls movement, manages grazing, and monitors animal location and movement. | 中 | SP005 |
| CP009 | Merck says one Vence base station covers roughly 5,000 to 10,000 acres depending on terrain. | 中 | SP005 |
| CP010 | Nofence targets both beef operations and dry-cow or replacement-heifer dairy workflows from the same cattle platform. | 中 | SP006 |
| CP011 | Nofence uses solar-powered GPS collars and says no base stations are required. | 中 | SP006, SP007 |
| CP012 | Datamars Active Tag is positioned as real-time livestock monitoring rather than a full virtual-fencing system. | 中 | SP008 |
| CP013 | Cowlar markets smart neck collars for temperature, activity, and behavior monitoring to improve dairy operations and reproduction outcomes. | 中 | SP009 |
| CP014 | Moocall markets a tail-mounted sensor that alerts farmers ahead of calving rather than managing grazing boundaries. | 中 | SP010 |
| CP015 | Corral says its virtual-fencing system is built for cow-calf operations and that early partners receive a discount rate with dedicated support. | 中 | SP019 |
| CP016 | Drover markets solar ear tags for virtual fencing and says buyers do not need expensive base stations or infrastructure. | 中 | SP020 |
| CP017 | Collie markets a combined virtual-fencing and virtual-herding product and claims 230 euros saved per cow per year plus around three hours saved per day. | 中 | SP021 |
| CP018 | Ranchflow markets AI-powered livestock health monitoring combined with digital fencing. | 中 | SP022 |
| CP019 | Across the reviewed vendor set, virtual fencing generally combines app-drawn boundaries, GPS positioning, and audio or stimulus cues at the collar. | 高 | SP011, SP013, SP018 |
| CP020 | DTN describes Gallagher eShepherd as a two-piece system with neckbands plus remote base stations used when coverage is limited. | 中 | SP014 |
| CP021 | DTN describes Halter’s earlier ranch setup as collars plus a central communications tower, showing that Halter’s current satellite promise is an architectural evolution rather than the original default. | 中 | SP014, SP026 |
| CP022 | The Rangelands Gateway comparison says virtual-fence components from different manufacturers are generally not interoperable or interchangeable. | 中 | SP012 |
| CP023 | The Rangelands Gateway comparison lists a public eShepherd price signal of 4,500 US dollars per base station and 66 US dollars per collar in yearly cost. | 中 | SP012 |
| CP024 | University of Arizona cost work says Halter and Vence are roughly equal in year-one and year-two onward cost comparisons. | 中 | SP017 |
| CP025 | CMK places advanced Halter pricing around NZ$9.90 per cow per month. | 中 | SP023 |
| CP026 | Farm Progress quotes a rancher hearing 70 to 80 US dollars per head per year for virtual fencing and judging that level uneconomic for broad use. | 中 | SP015 |
| CP027 | Powerflex says category-level collar prices often run 300 to 500 US dollars per animal plus base-station hardware and 30 to 100 US dollars of annual subscription fees. | 中 | SP013 |
| CP028 | Nofence says each cattle collar includes the first year of app access, support, and software updates and then moves to a monthly or annual subscription. | 中 | SP006 |
| CP029 | Corral offers only an early-partner discount signal publicly rather than a durable list-price schedule. | 中 | SP019 |
| CP030 | Drover’s pitch is explicitly infrastructure-light because it uses ear tags and says no base stations are needed. | 中 | SP020 |
| CP031 | Halter’s direct-to-satellite beef positioning reduces infrastructure friction in remote terrain relative to base-station-led systems. | 中 | SP001, SP026, SP014 |
| CP032 | Gallagher and Vence remain structurally strong where buyers want purpose-built extensive-grazing systems and can support more infrastructure. | 中 | SP004, SP005, SP014 |
| CP033 | Nofence’s no-base-station design lowers pilot friction but still keeps buyers committed to collar subscriptions and training. | 中 | SP006, SP007, SP018 |
| CP034 | Adjacent monitoring products from Datamars, Cowlar, Moocall, and Ranchflow compete for animal-data budget without matching full grazing-control breadth. | 中 | SP008, SP009, SP010, SP022 |
| CP035 | The status-quo substitute for many ranches is still physical fence, portable wire, and manual shifting rather than software-defined grazing. | 中 | SP013, SP015 |
| CP036 | Because vendor components are not interoperable, a buyer that multi-homes across systems takes on integration and training burden rather than gaining clean swapability. | 中 | SP012, SP018 |
| CP037 | Welfare and regulatory scrutiny remain category-wide trust friction for virtual fencing even as adoption expands. | 高 | SP018, SP024, SP025 |
| CP038 | Dairy Australia says Halter is currently the most broadly adopted virtual-fencing solution in Australian dairy herds. | 中 | SP025 |
| CP039 | Dairy Australia says legalization in New South Wales, South Australia, and Victoria should bring broader adoption and new market entrants. | 中 | SP025 |
| CP040 | Halter’s March 2026 financing disclosure says the company has sold one million collars to more than 2,000 farmers and ranchers. | 中 | SP003 |
| CP041 | Powerflex and Farm Progress both present cost as a central adoption brake even when virtual fencing promises labor and flexibility benefits. | 高 | SP013, SP015, SP017 |
| CP042 | Corral, Drover, Collie, and Ranchflow show that entrant pressure can come from cow-calf specialization, ear tags, virtual herding, or AI monitoring rather than only collar-for-collar copies of Halter. | 中 | SP019, SP020, SP021, SP022 |
| CP043 | The reviewed public source set does not provide enough verified deployment counts or funding detail to rank early entrants like Corral, Drover, Collie, or Ranchflow against Halter on scale. | 低 | SP019, SP020, SP021, SP022 |
| CP044 | The reviewed public source set does not provide realized net pricing, discount schedules, or renewal behavior for most virtual-fencing vendors. | 低 | SP012, SP019, SP020 |
| CP045 | Competitive moat in this category looks more like product-system breadth, deployment flexibility, and trust than an uncontested hardware monopoly. | 中 | SP011, SP012, SP024, SP025 |
| CI001 | Halter sells a hardware-and-software system built around smart collars, virtual fencing, and herd or pasture management workflows. | 高 | SI001, SI007, SI008 |
| CI002 | Halter’s dairy offer includes pasture management, heat monitoring, health monitoring, and shifting workflows. | 中 | SI008 |
| CI003 | Halter says beef required its own product and engineering rather than a repurposed dairy product. | 中 | SI015, SI007 |
| CI004 | Halter’s rural-professionals page shows the product surfaces live KPI views such as feed wedge, growth rates, paddock APC, and stock numbers between farm visits. | 中 | SI006 |
| CI005 | Halter says the product is available to dairy and beef farmers with unique packages for each. | 中 | SI006, SI007, SI008 |
| CI006 | AgFunder says Halter’s business model is subscription-based, with pricing around $6-$10 per cow per month plus a one-time infrastructure fee for towers. | 中 | SI009 |
| CI007 | iStart says Halter positions its system as full herd management at US$5 to US$8 per animal per month. | 中 | SI011 |
| CI008 | CMK says advanced Halter pricing recently reduced to NZ$9.90 per cow per month. | 中 | SI016 |
| CI009 | DTN says Halter’s satellite-connected system is available at nearly the same collar and subscription prices as before while removing the need for a tower. | 中 | SI024 |
| CI010 | Halter’s March 2026 financing disclosure says the company has sold one million collars to more than 2,000 farmers and ranchers. | 高 | SI001, SI002 |
| CI011 | Halter’s March 2026 financing disclosure says the company has more than 400 employees. | 中 | SI001 |
| CI012 | Business Wire, iStart, and Farmers Weekly all describe Halter’s largest-ever hiring push at more than 200 or 220-plus roles focused on product, engineering, and customer operations. | 高 | SI002, SI011, SI019 |
| CI013 | Official and independent round coverage says Series E funds are being used for U.S., New Zealand, and Australia expansion, UK and Ireland entry, and further product development. | 高 | SI001, SI010, SI018, SI019 |
| CI014 | Farmers Weekly says Halter will keep investing in product development including animal health monitoring and pasture management. | 中 | SI019 |
| CI015 | Halter’s ROI materials say an AgFirst and Transform Agri study found roughly 9% more pasture eaten, 9.5% more milk solids per hectare, and about a 13% increase in profit before tax across ten farms. | 高 | SI004, SI005 |
| CI016 | Halter’s rural-professionals page includes customer anecdotes of about three hours of labor saved per day plus production benefits. | 中 | SI006 |
| CI017 | TechCrunch and Promus both frame Halter as software, sensors, and AI embedded directly into livestock operations rather than a simple hardware SKU. | 中 | SI010, SI022 |
| CI018 | Using Halter’s disclosed one million collars and 2,000-plus customers implies roughly 500 collars per customer on average. | 中 | SI001, SI002 |
| CI019 | Using Halter’s disclosed 2,000-plus customers and 400-plus employees implies roughly five customers per employee. | 中 | SI001 |
| CI020 | Hiring concentrated in product, engineering, and customer operations implies a service-heavy direct go-to-market and onboarding model rather than a self-serve software motion. | 中 | SI002, SI011, SI020 |
| CI021 | Satellite connectivity materially improves the customer economics of deploying Halter in remote terrain because it removes tower requirements and expands reachable beef acreage. | 中 | SI014, SI024, SI025, SI026 |
| CI022 | Halter’s beef-specific product work implies ongoing vertical-specific R&D and engineering cost rather than simple software reuse. | 中 | SI015, SI007 |
| CI023 | Merck’s Q1 2026 10-Q shows that a scaled animal-health platform carries explicit cost of sales, selling and administrative expense, and research and development expense layers. | 高 | SI012, SI013 |
| CI024 | The availability of Merck 10-K and 10-Q filings highlights how much more transparent public comparables are than Halter on audited revenue, cost, and balance-sheet data. | 中 | SI012, SI013, SI001, SI002 |
| CI025 | AgFunder estimates that Halter’s disclosed subscription rates and installed collar base imply ARR in the rough range of US$70 million to US$100 million. | 低 | SI009 |
| CI026 | Public sources disclose pricing proxies and customer outcome signals more readily than audited ARR, revenue mix, or margin. | 中 | SI009, SI011, SI016, SI001 |
| CI027 | Reviewed public sources do not disclose Halter’s split between hardware revenue, recurring subscription revenue, and services revenue. | 中 | SI001, SI002, SI009, SI011 |
| CI028 | Reviewed public sources do not disclose Halter’s gross margin or hardware margin. | 中 | SI001, SI002, SI009, SI012, SI013 |
| CI029 | Reviewed public sources do not disclose Halter’s cash on hand, monthly burn, or runway. | 中 | SI001, SI002, SI010, SI011 |
| CI030 | Reviewed public sources do not disclose debt, credit facilities, or project-finance obligations for Halter. | 中 | SI001, SI002, SI010, SI011 |
| CI031 | Reviewed public sources do not disclose CAC, payback period, net revenue retention, or churn. | 中 | SI001, SI002, SI010, SI011 |
| CI032 | Farm Progress shows that virtual fencing can still look uneconomic to ranchers at 70 to 80 US dollars per head per year. | 中 | SI017 |
| CI033 | The New Zealand Veterinary Association says welfare and safeguard requirements remain material for virtual fencing, which can raise service burden or slow adoption. | 中 | SI023 |
| CI034 | Halter should be analyzed as a hardware-enabled recurring-revenue system rather than as pure SaaS. | 中 | SI009, SI011, SI017, SI022 |
| CI035 | Capital adequacy improved materially with the $220 million Series E, but next-round timing still depends on opaque burn, margin, and international rollout costs. | 中 | SI001, SI010, SI011, SI019 |
| CI036 | The strongest public traction metrics are customers, collars sold, headcount, and hiring plans rather than audited revenue or margin disclosures. | 中 | SI001, SI002, SI011 |
| CI037 | The ten-farm profit study supports revenue quality by suggesting Halter is tied to measurable operating outcomes rather than novelty alone. | 高 | SI004, SI005, SI006 |
| CI038 | Satellite launch coverage says Halter is widening the product surface customers pay for with reproduction, animal behavior, and precision pasture tools alongside connectivity. | 中 | SI014, SI024, SI025, SI026 |
| CI039 | The hardest financial diligence blockers are revenue mix, gross margin, burn or runway, and retention rather than whether demand exists. | 中 | SI001, SI004, SI009, SI011, SI017 |
| CI040 | The supportable financial verdict is that Halter has strong value proof and fresh capital, but public disclosure remains too thin for full unit-economics underwriting. | 中 | SI001, SI004, SI009, SI013, SI017 |
| CI041 | Halter’s careers page shows active hiring across product, engineering, operations, and commercial functions, reinforcing post-Series-E operating-expense expansion. | 中 | SI020 |
| CI042 | Rural News restates the Series E round as NZ$377 million and frames it as funding for global expansion. | 中 | SI021 |
| CE001 | Halter publicly presents one connected product system spanning collars, connectivity infrastructure, apps, and intelligence rather than a standalone collar accessory. | 中 | SE001 |
| CE002 | Halter says its dairy product supports almost-instant 90 second response to app-driven shifts plus minute-level heat and health monitoring. | 中 | SE001 |
| CE003 | The company describes the collar as solar-powered, weather-resistant, dust-sealed, ergonomic, and designed for year-round reliability. | 中 | SE001 |
| CE004 | Halter describes the tower as a self-powered, long-range connectivity layer that does not rely on mobile reception. | 中 | SE001 |
| CE005 | Halter says the beef system connects collars by satellite with no towers or cell coverage required, enabling use on public land, leased ground, and seasonal country. | 高 | SE001, SE015 |
| CE006 | Halter states it is the only satellite-enabled virtual fencing system in the world. | 高 | SE001, SE014 |
| CE007 | The public technology page says over 100 engineers and designers build Halter across hardware, software, and AI. | 高 | SE001, SE010 |
| CE008 | Public architecture language centers on collars, connectivity, app surfaces, cloud processing, and machine-learning models that turn raw data into farmer-facing actions. | 中 | SE001 |
| CE009 | Halter says each collar sends over 6,000 data points every minute to its cloud-based data platform. | 中 | SE001 |
| CE010 | Halter says it has collected over 7 billion hours of animal behaviour data. | 中 | SE001 |
| CE011 | Halter presents Cowgorithm as the algorithm that automatically trains cattle to respond to directional sound cues and unlocks virtual fencing and remote shifting. | 高 | SE001, SE006 |
| CE012 | The training guidance page says primary cues are designed to be predictable and controllable so cattle learn how to avoid the secondary cue. | 中 | SE006 |
| CE013 | Halter’s software-engineer profile says paddock-growth modeling work sits between infrastructure and machine-learning experimentation. | 中 | SE010 |
| CE014 | Halter’s patent-marking page lists a PCT filing for a machine-learning model for determining foliage cover and training method therefor. | 高 | SE002, SE017 |
| CE015 | The same patent-marking page lists 2024 filings covering livestock guidance devices and wireless message relaying, implying active IP work beyond a single collar form factor. | 高 | SE002, SE017 |
| CE016 | The rural-professionals software policy says Halter’s service stack includes georeferenced physical-world imagery data and Hexagon software terms. | 中 | SE011 |
| CE017 | Halter’s new-features page markets more granular pasture data as an active product update. | 中 | SE003 |
| CE018 | The same roadmap page markets heat detection and breeding management including daily heat lists, cycling activity, and estimated pregnancy status. | 高 | SE003, SE027 |
| CE019 | Halter’s roadmap page labels map-based task assignment with photos and comments as coming soon rather than already deployed everywhere. | 中 | SE003 |
| CE020 | The reduce-workload page frames remote shifting and digital break setting as explicit workload-reduction features. | 中 | SE025 |
| CE021 | The harvest-more-pasture page positions Halter as a pasture-allocation and utilization tool rather than only a containment system. | 中 | SE026 |
| CE022 | The improve-mating-results page presents heat detection and breeding management as a first-class workflow module. | 中 | SE027 |
| CE023 | Halter publishes a dedicated Veterinary Advisory Board page as part of its animal-welfare control framework. | 高 | SE004, SE007 |
| CE024 | The animal-health-benefits page says customers have reported reduced non-environmental lameness after adopting Halter guidance. | 低 | SE005 |
| CE025 | Leadership-and-governance materials make animal welfare a governance topic rather than only a product FAQ. | 高 | SE007, SE024 |
| CE026 | Halter publishes legislative-compliance guidance that explicitly ties the system to animal-welfare and legal obligations across operating markets. | 高 | SE008, SE024 |
| CE027 | Halter’s Australia regulation page says New South Wales, Victoria, South Australia, and the Northern Territory implemented cattle virtual-fencing regulations in the second half of 2025, leaving all Australian jurisdictions permissive by 2026. | 高 | SE009, SE018 |
| CE028 | Dairy Australia said virtual fencing and herding became available to all dairy farmers in January 2026. | 中 | SE018 |
| CE029 | DTN reported that Halter’s April 2026 beef update shifted collars from cell towers or on-ranch infrastructure onto Starlink connectivity. | 高 | SE015, SE014 |
| CE030 | DTN identified High Lonesome Ranch in western Colorado, spanning over 225,000 acres, as an early satellite-enabled Halter deployment. | 中 | SE015 |
| CE031 | Farm Progress said U.S. ranchers still weigh return on investment carefully because virtual fencing remains expensive despite rising adoption. | 高 | SE016, SE021 |
| CE032 | VetSalus said the RSPCA has banned virtual collars on farms operating under its revised dairy-cattle welfare standards in the UK. | 中 | SE019 |
| CE033 | Google Play listed Halter with 5K+ downloads and a 2026-05-28 update at access time. | 中 | SE012 |
| CE034 | Apple’s App Store showed Halter version 1.296.0 with a 4.6 out of 5 rating from 59 ratings at access time. | 中 | SE013 |
| CE035 | BusinessWire said Halter’s U.S. customers had created more than 11,000 miles of virtual fencing by November 2025. | 中 | SE020 |
| CE036 | WireMeshNews said U.S. virtual fencing moved from pilot to early adoption in 2025 but battery life, connectivity, and hardware cost remained the main friction points. | 中 | SE021 |
| CE037 | No public uptime dashboard, SLA, or field-failure-rate disclosure was surfaced during this review. | 中 | SE001, SE003, SE011 |
| CE038 | Public materials surfaced app surfaces and software policies but no public API documentation or third-party integration reference for implementers. | 中 | SE001, SE011, SE010 |
| CE039 | No public security-certification register, external audit report, or trust portal was surfaced during review. | 低 | SE001, SE024, SE011 |
| CU001 | Halter’s visible customer proof splits into two main clusters: New Zealand dairy farmers and U.S. beef ranchers. | 中 | SU001, SU002 |
| CU002 | The U.S. rancher stories page samples named accounts in Colorado, Nebraska, Montana, California, and Oklahoma. | 中 | SU001 |
| CU003 | The New Zealand farmer stories page samples Canterbury, Tasman, Waikato, and other dairy regions with herd sizes from hundreds to more than 1,500 cows. | 中 | SU002 |
| CU004 | Halter’s 2024 U.S. launch release said the company had 200,000 cattle under management across the U.S., New Zealand, and Australia, with U.S. customers in California, Oregon, Colorado, Texas, and Louisiana. | 中 | SU008 |
| CU005 | BusinessWire said Halter’s U.S. customers had created more than 11,000 miles of virtual fencing by November 2025. | 中 | SU016 |
| CU006 | The same BusinessWire release estimated that 11,000 miles of virtual fencing represented roughly $220 million of avoided conventional-fencing cost at $20,000 per mile. | 中 | SU016 |
| CU007 | Animal AgTech quoted Halter as partnered with 250 ranchers across 24 states and live on more than 300 ranches nationally. | 中 | SU017 |
| CU008 | DTN identified High Lonesome Ranch on more than 225,000 acres as an early satellite-enabled Halter deployment. | 中 | SU023 |
| CU009 | A Grady Grissom rancher story claims Halter lifted carrying capacity by 20 percent. | 低 | SU001 |
| CU010 | A Ron Jespersen rancher story says Halter avoided a $20,000 fence build. | 中 | SU001 |
| CU011 | A Halter YouTube testimonial frames one California rancher around $210,000 of fencing-equivalent value and 7 miles of fence. | 中 | SU012 |
| CU012 | A Lauren Sizemore story tile says Halter saves hundreds of hours without polywire. | 中 | SU001 |
| CU013 | A Bart and Wendy Morris story says Halter supports two to three times county-average stocking rate. | 中 | SU001 |
| CU014 | Emerald View Farm appears on Halter’s farmer stories page as a 1,545-cow Canterbury dairy account. | 中 | SU002 |
| CU015 | Lizzy and Cameron Te Brake appear on Halter’s farmer stories page as a 440-cow Waikato dairy account. | 中 | SU002 |
| CU016 | Halter’s New Zealand proof set leans on “more milk, less effort” and “doubling stock units” narratives, implying dairy ROI is communicated mainly through labor and pasture outcomes. | 中 | SU002, SU007 |
| CU017 | Halter’s dairy ROI article says an independent AgFirst and Transform Agri study covered 10 high-performing New Zealand dairy farms using Halter. | 中 | SU007 |
| CU018 | That same article says the observed returns came through pasture utilization, labor efficiency, reproduction, and cost control rather than one single lever. | 中 | SU007 |
| CU019 | Halter’s winter-grazing article says ranchers use the system for winter grazing and cattle management, not only for peak-season rotations. | 中 | SU005 |
| CU020 | Halter’s “Two Smart Ways” article says ranchers use the system for less-obvious jobs that previously required more fence, labor, and stress on stock. | 中 | SU006 |
| CU021 | Halter’s Ranch Management Resource Initiative is available to ranchers with 250 or more collared head and covers up to $2,500 of outside training cost. | 中 | SU011 |
| CU022 | Halter’s competition-winners page shows a free-year onboarding incentive across two dairy farms and two beef farms. | 中 | SU009 |
| CU023 | Halter’s Rio Grande partnership shows expansion can run through conservation and public-land partners rather than only direct one-farm selling. | 中 | SU010 |
| CU024 | WyoFile coverage quoted a Wyoming rancher calling virtual fencing a win-win because it reduces wire conflict with wildlife while improving land-use flexibility. | 中 | SU004 |
| CU025 | Apple and Google Play show Halter’s app is live across iOS and Android, with 5K+ Google Play downloads and a 4.6 out of 5 iOS rating from 59 ratings at access time. | 高 | SU013, SU014 |
| CU026 | AppBrain estimated about 6.2 thousand Android downloads and 460 downloads in the prior 30 days, but no user ratings yet. | 中 | SU015 |
| CU027 | The visible proof set is production-oriented rather than pilot-oriented; the named stories describe current farms and ranches already using Halter in live operations. | 中 | SU001, SU002 |
| CU028 | Most public customer proof is still Halter-authored or Halter-amplified rather than contract-level or customer-authored procurement evidence. | 中 | SU001, SU002, SU003, SU004, SU008 |
| CU029 | No public NRR, GRR, logo-churn, or cohort-retention disclosure was surfaced in this review. | 中 | SU001, SU002, SU013, SU014 |
| CU030 | No public top-customer spend, concentration, or revenue-per-account breakdown was surfaced in this review. | 中 | SU001, SU002, SU016, SU017 |
| CU031 | Farm Progress said high cost still forces ranchers to weigh return on investment carefully before fully committing to virtual fencing. | 高 | SU019, SU022 |
| CU032 | VetSalus said the 2026 Halter funding story was not all positive and highlighted RSPCA resistance to virtual collars. | 中 | SU020 |
| CU033 | Stuff published a piece titled “Petition opposes virtual fencing,” showing organized opposition exists in New Zealand. | 中 | SU021 |
| CU034 | WireMeshNews said early U.S. adoption is real but battery life, connectivity, and hardware cost remain the main friction points across the category. | 中 | SU022 |
| CU035 | Halter’s NSW farmers page shows customer demand for virtual fencing can outpace the speed of enabling regulation. | 中 | SU025 |
| CU036 | Halter’s VF101 guide explains onboarding around communication towers, smart collars, and managed cattle movement, implying deployment is operational rather than self-serve software. | 中 | SU024 |
| CU037 | BusinessWire’s 11,000-mile figure plus Animal AgTech’s 250-rancher, 24-state figure together support real U.S. customer breadth beyond a handful of showcase logos. | 高 | SU016, SU017 |
| CU038 | Public satisfaction proxies are app ratings and anecdotal stories, not audited buyer-satisfaction or renewal data. | 中 | SU013, SU014, SU015, SU001, SU002 |
| CU039 | The economic buyer in visible proof appears to be the farm or ranch owner-operator because outcome language centers on fencing cost, carrying capacity, labor, pasture, and productivity. | 中 | SU001, SU002, SU007, SU016 |
| CU040 | Halter’s public customer journey appears land-and-expand shaped: initial proof or onboarding, live operational value, then broader acreage, headcount, or management-education attachment. | 中 | SU005, SU011, SU024 |
| CR001 | Halter’s March 2026 Series E announcement said the company raised $220 million at a $2 billion valuation. | 高 | SR001, SR002 |
| CR002 | The same financing materials said Halter served more than 2,000 farmers and ranchers and had sold one million collars across New Zealand, Australia, and the U.S. | 高 | SR001, SR002 |
| CR003 | Halter publicly tied the Series E proceeds to global expansion and a 200-plus hiring plan centered on product, engineering, and customer roles. | 高 | SR001, SR002, SR031 |
| CR004 | AgFunder described Halter’s business model as subscription-led, with pricing around $6 to $10 per cow per month plus a one-time infrastructure fee. | 中 | SR003 |
| CR005 | North Dakota State University estimated a Halter deployment at about $11.08 per head per month in a 100-head, five-year scenario that included a $4,500 tower, $50 monthly internet fee, $1,000 delivery fee, and $6 monthly collar fee. | 中 | SR020 |
| CR006 | Otago Daily Times reported that Halter cut its dairy package price by 37% to NZ$9.90 per cow per month. | 中 | SR028 |
| CR007 | Farm Progress characterized virtual fencing adoption in the U.S. as still exposed to high costs and ROI scrutiny. | 中 | SR027 |
| CR008 | The University of Arizona vendor comparison said virtual fence components from different manufacturers are generally not interoperable or interchangeable. | 中 | SR023 |
| CR009 | The same vendor comparison showed Halter, Vence, Gallagher, and Nofence each require a distinct hardware and infrastructure stack rather than a commodity standard. | 高 | SR020, SR023, SR024, SR025 |
| CR010 | Halter’s satellite launch said direct-to-satellite collars remove the need for cell towers or on-ranch infrastructure. | 高 | SR005, SR006 |
| CR011 | The satellite launch also claimed Halter’s internal modeling expands coverage of the U.S. beef cattle market by 2.5 times. | 中 | SR005, SR006 |
| CR012 | High Plains Journal said connectivity had been the final barrier to bringing virtual fencing across remote and expansive ranches. | 中 | SR006 |
| CR013 | North Dakota State University described virtual fencing as a flexible grazing tool but framed product and pricing differences as system-dependent rather than universally solved. | 中 | SR020 |
| CR014 | Colorado State University reported that one rancher’s first year with virtual fencing included collars falling off, moisture-related malfunctions, and faster-than-expected battery drain. | 中 | SR022 |
| CR015 | PERC argued virtual fencing must make economic sense for producers and become widely adopted before conservation benefits can scale. | 中 | SR021 |
| CR016 | PERC also said physical fences will remain necessary in many contexts such as boundary law, roads, and wildlife exclusion. | 中 | SR021 |
| CR017 | The UK Animal Welfare Committee opinion said virtual fencing can be used without detriment only if proper-use conditions are met and further research continues. | 高 | SR013, SR014 |
| CR018 | The New Zealand Veterinary Association said the current literature base is still mostly short-term and often industry-funded, with limited independent long-term research. | 中 | SR014 |
| CR019 | Halter’s welfare overview says a collar pulse has a maximum strength of 0.45 joules, below a typical mains-powered electric fence. | 低 | SR011 |
| CR020 | Halter’s animal-welfare research page anchors the company’s welfare case in external studies rather than only internal assertion. | 中 | SR012 |
| CR021 | Dairy Australia said legislative changes in New South Wales, South Australia, and Victoria made virtual fencing legal in all six dairying states. | 中 | SR015 |
| CR022 | USDA and DOI said their March 2026 MOU encourages adoption of innovative technologies such as virtual fencing on public lands. | 高 | SR016, SR017 |
| CR023 | The SBA Advocacy summary said BLM’s proposed grazing-rule changes may significantly affect small ranching operations that hold public-land permits or leases. | 中 | SR017 |
| CR024 | The SBA summary also said comments on the proposed BLM rule were due by July 13, 2026, confirming the framework is still live rather than settled. | 中 | SR017 |
| CR025 | University of Idaho’s 2026 study put 550 mother cows with calves on virtual fencing in a federal grazing-allotment case study using a portable cellular base station. | 中 | SR019 |
| CR026 | The same University of Idaho study said the Foundation for America’s Public Lands funded the work with a $235,000 grant. | 中 | SR019 |
| CR027 | Halter’s California BLM partnership announcement said the program included $2.7 million in funding to support ranchers on BLM-managed land. | 中 | SR007 |
| CR028 | Halter’s Rio Grande monument announcement shows the company’s public-land story depends on cooperation with ranchers and conservation groups, not only product capability. | 高 | SR008, SR019 |
| CR029 | Halter’s rural-professionals page positions consultants and vets as external actors in farm execution, indicating some workflow depth is reinforced by advisor relationships. | 中 | SR010 |
| CR030 | The careers page and iStart coverage show Halter is trying to add more than 200 roles across New Zealand, Australia, and the U.S. on a compressed timeline. | 高 | SR009, SR031 |
| CR031 | TechCrunch framed Founders Fund’s backing as a large strategic bet on Halter’s cow-collar platform rather than a routine agtech follow-on. | 中 | SR004 |
| CR032 | Agtech Industry Examiner argued the company is now being judged as operational infrastructure, which raises the burden of proof on unit economics and repeatable adoption. | 中 | SR029, SR003 |
| CR033 | The New Zealand Companies Office extract lists Halter USA Inc. as the ultimate holding company for Halter Limited. | 中 | SR030 |
| CR034 | The same company extract shows public director visibility concentrated in two named directors on the NZ entity. | 中 | SR030 |
| CR035 | Gallagher’s eShepherd page highlights both cellular and base-station options, implying Halter does not own the only architecture available to ranchers. | 中 | SR024 |
| CR036 | Merck Animal Health’s Vence page shows a large incumbent animal-health owner is one of the main competitive alternatives in virtual fencing. | 中 | SR025 |
| CR037 | Farm Progress and PERC both imply that adoption outside the early-adopter cohort will require better economics and trust, not just more awareness. | 中 | SR021, SR027 |
| CR038 | Otago Daily Times and AgFunder together indicate Halter is already adapting price points to a farm economy that remains cost sensitive. | 中 | SR003, SR028 |
| CR039 | The University of Arizona comparison and Colorado State field report together imply switching costs and hardware reliability both matter because the category is not yet standardized. | 中 | SR022, SR023 |
| CR040 | The AWC opinion and the NZVA statement together show welfare legitimacy remains a live diligence item rather than a closed debate. | 高 | SR013, SR014 |
| CV001 | Halter’s March 2026 Series E materials said the company raised $220 million at a $2 billion valuation. | 高 | SV001, SV002 |
| CV002 | SmartCompany framed the same round as NZ$315 million at a NZ$2.9 billion valuation. | 中 | SV005 |
| CV003 | Halter’s June 2025 Series D announcement said the company was valued at US$1 billion after raising $165 million. | 中 | SV008 |
| CV004 | The public mark roughly doubled from US$1 billion in 2025 to US$2 billion in 2026. | 高 | SV001, SV008 |
| CV005 | Official 2026 round coverage said Halter served more than 2,000 ranchers and farmers and had sold one million collars. | 高 | SV001, SV015 |
| CV006 | Tech Funding News described Halter’s 2026 round as one of agtech’s largest financings. | 中 | SV015 |
| CV007 | AgFunder said Halter’s pricing starts around US$6 to US$10 per cow per month plus an infrastructure fee. | 中 | SV003 |
| CV008 | Otago Daily Times reported that Halter cut its dairy package price to NZ$9.90 per cow per month. | 中 | SV012 |
| CV009 | North Dakota State University estimated a Halter deployment at about US$11.08 per head per month in a modeled 100-head setup. | 中 | SV013 |
| CV010 | AgFunder estimated Halter’s disclosed pricing and collar base imply roughly US$70 million to US$100 million of ARR. | 低 | SV003 |
| CV011 | A US$2 billion mark against a public ARR estimate of US$70 million to US$100 million implies roughly 20x to 29x ARR. | 低 | SV001, SV003 |
| CV012 | Farm Progress reported that virtual fencing can still look uneconomic to ranchers at roughly US$70 to US$80 per head per year. | 中 | SV011 |
| CV013 | Price cuts and extension-cost estimates together show Halter’s monetization still sits inside a cost-sensitive farm economy rather than a frictionless software budget. | 中 | SV011, SV012, SV013 |
| CV014 | The Companies Office extract lists Halter USA Inc. as the ultimate holding company of Halter Limited. | 中 | SV010 |
| CV015 | TechCrunch and AgFunder both frame Halter as more than a conventional agtech point solution, which is the public logic behind a premium narrative. | 高 | SV003, SV004 |
| CV016 | Agtech Industry Examiner argued the $2 billion mark must be justified in a tougher funding market by proving software-like infrastructure economics. | 中 | SV014 |
| CV017 | Farmers Weekly and Rural News emphasized that the raise funds global expansion, not just New Zealand dairy penetration. | 中 | SV006, SV007 |
| CV018 | Halter’s ROI study announcement said an independent study across 10 farms found a 13.2% average increase in profit before tax. | 中 | SV009 |
| CV019 | That ROI study is directionally supportive but too small and too curated to close valuation comfort across global markets by itself. | 中 | SV009 |
| CV020 | CompaniesMarketCap listed Deere’s market capitalization at US$146.44 billion as of May 2026. | 中 | SV018 |
| CV021 | Macrotrends listed Deere’s trailing-twelve-month revenue at US$44.433 billion for the period ending July 31, 2025. | 中 | SV019 |
| CV022 | Using those public figures, Deere trades at roughly 3.3x market cap to trailing revenue. | 中 | SV018, SV019, SV025 |
| CV023 | CompaniesMarketCap listed AGCO’s market capitalization at US$8.13 billion as of May 2026. | 中 | SV020 |
| CV024 | Macrotrends listed AGCO’s trailing-twelve-month revenue at US$11.662 billion for the period ending December 31, 2024. | 中 | SV021 |
| CV025 | Using those public figures, AGCO trades at roughly 0.7x market cap to trailing revenue. | 中 | SV020, SV021, SV026 |
| CV026 | CompaniesMarketCap listed Trimble’s market capitalization at US$13.14 billion as of May 2026. | 中 | SV022 |
| CV027 | Macrotrends listed Trimble’s trailing-twelve-month revenue at US$3.601 billion for the period ending September 30, 2025. | 中 | SV023 |
| CV028 | Using those public figures, Trimble trades at roughly 3.65x market cap to trailing revenue. | 中 | SV022, SV023, SV027 |
| CV029 | CompaniesMarketCap listed Merck’s market capitalization at US$293.21 billion as of May 2026. | 中 | SV016 |
| CV030 | Macrotrends listed Merck’s trailing-twelve-month revenue at US$64.235 billion for the period ending September 30, 2025. | 中 | SV017 |
| CV031 | Using those public figures, Merck trades at roughly 4.56x market cap to trailing revenue. | 中 | SV016, SV017, SV024 |
| CV032 | Across Deere, AGCO, Trimble, and Merck, the public market-cap-to-revenue band runs from roughly 0.7x to 4.6x. | 中 | SV016, SV017, SV018, SV019, SV020, SV021, SV022, SV023 |
| CV033 | AGCO is the punitive end of the comp set because it is a hardware-heavy agricultural OEM without a software-style rerating. | 中 | SV020, SV021, SV026 |
| CV034 | Deere sits above AGCO because precision-enabled agricultural exposure commands a better public multiple than plain farm equipment. | 中 | SV018, SV019, SV025 |
| CV035 | Trimble is the most supportive upper-mid public comp because markets already pay it a mid-3x revenue multiple for workflow-heavy industrial technology exposure. | 中 | SV022, SV023, SV027 |
| CV036 | Merck is useful as an animal-health scale reference but a weak direct match because Vence sits inside a much broader pharmaceutical platform. | 高 | SV016, SV017, SV028, SV030 |
| CV037 | Halter deserves a premium to AGCO-like machinery multiples because the public file shows recurring animal-based pricing, software workflows, and fast international growth rather than one-off equipment sales. | 高 | SV001, SV003, SV015, SV020, SV021 |
| CV038 | Halter deserves a discount to pure SaaS framing because gross margin, churn, CAC, support burden, and cap-table terms remain opaque while hardware and field operations are clearly material. | 高 | SV003, SV010, SV011, SV013, SV014 |
| CV039 | A bear case around US$0.9 billion to US$1.3 billion becomes plausible if price sensitivity, adoption friction, or incident risk force the business toward lower industrial-style multiples. | 中 | SV011, SV012, SV013, SV014, SV020, SV021 |
| CV040 | A base case around US$1.6 billion to US$2.2 billion requires Halter to hold premium growth and prove enough retention and operating leverage to avoid a sharp de-rating from the current mark. | 中 | SV001, SV003, SV015, SV018, SV019, SV022, SV023 |
| CV041 | A bull case around US$2.6 billion to US$3.4 billion requires ARR to outrun public rough estimates, satellite-enabled expansion to broaden U.S. coverage, and new markets such as Ireland and the U.K. to convert efficiently. | 中 | SV001, SV004, SV015 |
| CV042 | Public evidence supports an exit path through another private round or strategic interest before a near-term IPO because the disclosure surface is not yet public-market grade. | 中 | SV004, SV010, SV030 |
| CV043 | Merck’s filing and the competitor product pages show credible strategic acquirers or category consolidators exist, but they do not prove a transaction will occur at a premium price. | 高 | SV028, SV029, SV030 |
| CV044 | The public file supports a monitor or invest-only-with-price-discipline recommendation rather than chasing the full mark without private metrics. | 中 | SV001, SV003, SV014, SV015 |
| CV045 | Confidence should be medium because market, customer, and product proof are strong while audited economics, preference stack, and cohort durability remain private. | 中 | SV001, SV003, SV009, SV010, SV014 |
| CV046 | Credible competitor alternatives from Gallagher and Merck cap any monopoly-style multiple assumption for Halter. | 中 | SV028, SV029 |
| 编号 | 出版方 | 标题 | 引文 |
|---|---|---|---|
| SO001 | Halter | Halter® | Virtual Fencing and Pasture Management | |
| SO002 | Halter | Halter raises $220M in Series E to accelerate global expansion of virtual fencing | Halter serves more than 2000 ranchers and farmers across New Zealand, Australia, and the U.S., with a million of its solar-powered collars now sold. |
| SO003 | Business Wire | Halter Raises $220M in Series E to Accelerate Global Expansion of Virtual Fencing | Halter plans to hire more than 200 people - its largest-ever hiring effort - with a focus on product, engineering, and customer roles at its Auckland HQ. |
| SO004 | Halter | Halter raises $165M in funding to help farmers boost productivity | Halter has raised $165M in a Series D fundraising round, valuing Halter at $1.65billion (USD $1 billion). |
| SO005 | AgFunderNews | Halter to beef up US expansion following $100m raise led by BOND | The new funding will fuel Halter’s expansion across the US, where the company already supports around 150 ranchers across 18 states. |
| SO006 | AgFunderNews | Halter says it's not an agtech company on the heels of $220m raise | Halter shouldn’t be limited by the typical boundaries associated with agtech investment or identity. |
| SO007 | TechCrunch | Peter Thiel's big bet on solar-powered cow collars | Halter’s collar is on more than a million cattle across more than 2,000 farms in New Zealand, Australia, and the United States, where the company operates in 22 states. |
| SO008 | Halter | Halter launches world-first direct-to-satellite virtual fencing | Halter’s internal modelling estimates direct-to-satellite capability expands coverage of the U.S beef cattle market by 2.5x. |
| SO009 | SmartCompany | NZ agtech startup Halter raises $315 million at $2.9 billion valuation | Halter was founded in 2016 by CEO Craig Piggott, and its virtual farm fences are now used across more than 2,000 cattle farms in New Zealand, Australia and the US. |
| SO010 | Farmers Weekly | Halter eyes fresh pastures after huge capital raise | Halter is hiring for 220-plus roles across New Zealand, Australia, and the US in coming weeks. |
| SO011 | Rural News Group | Halter Raises NZ$377M to Expand Virtual Fencing Globally | Halter now serves more than 2000 farmers and ranchers across New Zealand, Australia, and the United States, with one million of its solar-powered collars now sold. |
| SO012 | TIME | Virtual cattle fences | New Zealand-based Halter makes wireless smart collars that allow for the virtual guiding and monitoring of cattle herds. |
| SO013 | Halter | Halter partners with the Foundation for America’s Public Lands and the Bureau of Land Management to expand joint access in California | The partnership includes $2.7 million in funding to support ranchers using Halter on BLM-managed land. |
| SO014 | AgFunderNews | Halter will equip more US ranchers with virtual fencing tech via new partnership with federal land manager | |
| SO015 | Halter | Careers | Halter® | Halter is not your normal 9 to 5, and we understand this is not for everyone. |
| SO016 | Icehouse Ventures | Icehouse Ventures | Halter | Founded in 2016 by Craig Piggott and Max Olson, Halter bridges deep tech into real-world farming. |
| SO017 | Vessev | Vessev | Our story, who we are, and what drives us | |
| SO018 | Halter | Animal Welfare Charter: System Overview | The maximum strength of a single pulse is 0.45 joules, which is significantly less energy than the shock received from a typical mains-powered electric fence. |
| SO019 | Halter | Animal Welfare Charter: Virtual Fencing Research | |
| SO020 | New Zealand Veterinary Association | Position Statement: Virtual Fencing and Virtual Herding for Cattle | The NZVA also acknowledges that current literature is mostly short-term and often industry-funded, with limited independent, long-term research. |
| SO021 | Dairy Australia | Virtual fencing and herding option now available to all dairy farmers | With recent legislative changes in New South Wales, South Australia and Victoria, these innovations are now legal in all six dairying states. |
| SO022 | CMK | Smart Farming Investment: Making Sense of Cow Wearables for your Bottom Line | Advanced systems with virtual fencing (Halter): Recently reduced to $9.90 per cow per month ($118.80 annually). |
| SO023 | Halter | Dairy Farm ROI Study & Productivity Insights | Halter® | An independent study by AgFirst and Transform Agri analysed ten high-performing dairy farms. |
| SO024 | Halter | Independent Research Confirms Profit Gains from Halter Technology on NZ Dairy Farms | Farms recorded an average 13.2% increase in profit before tax (EBIT). |
| SO025 | Farmers Weekly | Research backs profit gains from Halter | |
| SM001 | Stats NZ | Agriculture | Stats NZ | |
| SM002 | LIC | New Zealand Dairy Industry | LIC | |
| SM003 | DairyNZ and LIC | New Zealand Dairy Statistics 2024-25 | |
| SM004 | USDA NASS | United States cattle inventory down slightly | |
| SM005 | USDA NASS | Cattle 01/30/2026 | |
| SM006 | Meat & Livestock Australia | Industry projections 2026 – Australian cattle | |
| SM007 | Australian Bureau of Statistics | Livestock products, March 2026 | |
| SM008 | ABARES | Snapshot of Australian Agriculture 2026 | |
| SM009 | University of Arizona Cooperative Extension | Foundations of Virtual Fencing: Economics of Virtual Fence (VF) Systems | |
| SM010 | Farm Progress | Virtual fencing gains ground in U.S. despite high costs | There is a standard cost of $72 per year per collar with a required three-year commitment. For the required towers, there is a one-time $4,500 price per tower. |
| SM011 | WireMeshNews | Virtual Fencing Gains Ground in the U.S. — Costs, Vendors, Trials & 2026 Outlook | |
| SM012 | CLEAR Center at UC Davis | Virtual Fencing: Drawing the Lines of Sustainability | |
| SM013 | Dairy Australia | Virtual fencing and herding option now available to all dairy farmers | |
| SM014 | New Zealand Veterinary Association | Position Statement: Virtual Fencing and Virtual Herding for Cattle | |
| SM015 | Halter | Halter® Beef | |
| SM016 | Halter | Halter® Dairy | |
| SM017 | Halter | Halter raises $220M in Series E to accelerate global expansion of virtual fencing | |
| SM018 | Gallagher | eShepherd Virtual Fencing | Gallagher United States | |
| SM019 | Merck Animal Health | Vence - Merck Animal Health USA | |
| SM020 | Nofence | Virtual fencing for cattle with GPS collars | Nofence | |
| SM021 | Datamars | Datamars Livestock | Active Tag | |
| SM022 | Cowlar | Smart Neck Collars for Dairy Cows | |
| SM023 | Moocall | Livestock Monitoring Systems | Moocall | |
| SM024 | CMK | Smart Farming Investment: Making Sense of Cow Wearables for your Bottom Line | |
| SM025 | AgFunderNews | Halter to beef up US expansion following $100m raise led by BOND | |
| SP001 | Halter | Halter for Beef Farms | Halter® | Dependable remote coverage, no towers required. Halter’s virtual fencing collars now connect directly to satellites, unlocking limitless fencing in any terrain. |
| SP002 | Halter | Dairy Farm Automation & Herd Management | Halter® | |
| SP003 | Halter | Halter raises $220M in Series E to accelerate global expansion of virtual fencing | Halter serves more than 2000 ranchers and farmers across New Zealand, Australia, and the U.S., with a million of its solar-powered collars now sold. |
| SP004 | Gallagher | eShepherd Virtual Fencing | Gallagher United States | Choose from two proven options — cellular for quick setup and lower cost, or LoRa base stations for remote terrain with limited coverage. |
| SP005 | Merck Animal Health | Vence | One base station covers 5,000-10,000 acres, depending on terrain. |
| SP006 | Nofence | Virtual fencing for cattle with GPS collars | Nofence | Each collar comes with a 5-year warranty and includes the first year of app access, support, and software updates. After that, choose a monthly or annual subscription. |
| SP007 | Nofence | Nofence products for cattle, sheep and goats | Nofence | The system runs on GPS satellites and cellular networks, with HerdNet providing collar-to-collar communication in areas with limited signal. No base stations, no buried wire, no complicated setup. |
| SP008 | Datamars | Datamars Livestock | Active Tag | |
| SP009 | Cowlar | Smart Neck Collars for Dairy Cows | |
| SP010 | Moocall | Livestock Monitoring Systems | Moocall | |
| SP011 | South Dakota State University Extension | Virtual Fencing: Emerging Companies, Functionality and Benefits | Virtual fence companies that are currently available for producers to purchase include Vence, NoFence, eShepherd, and Corral Technologies. |
| SP012 | Rangelands Gateway / University of Arizona | Virtual Fence Vendors Basic Comparison (March 2025) | Virtual fence components from different manufacturers are generally not interoperable or interchangeable. |
| SP013 | Powerflex Fence | Virtual Fence for Livestock: A Quick Overview of Pros, Cons & Where It Actually Works | Collars run $300–$500 per animal at retail, plus base station hardware, plus annual subscription fees ($30–$100 per collar per year depending on system). |
| SP014 | DTN / Progressive Farmer | Virtual Fencing: A Rancher's New Best Friend | |
| SP015 | Farm Progress | Virtual fencing gains ground in U.S. despite high costs | In hearing the numbers, 70-80 bucks a head, if you want to turn out 80-100 cattle on a piece of land, that’s like $8,000 a year. |
| SP016 | WireMeshNews | Virtual Fencing for Cattle — Costs, Vendors, Trials & 2026 Outlook | Four vendors dominate the conversation—Vence (Merck Animal Health), Gallagher’s eShepherd, Nofence, and Halter. |
| SP017 | University of Arizona Cooperative Extension | Foundations of Virtual Fencing: Economics of Virtual Fence (VF) Systems | Halter & Vence costs roughly equal. |
| SP018 | CLEAR Center at UC Davis | Virtual Fencing: Drawing the Lines of Sustainability | Once the virtual boundary is uploaded to the collars—either through cellular networks or base stations that connect to those networks—the fence exists wherever the producer needs it to be. |
| SP019 | Corral Technologies | Corral Technologies - Ranching Reimagined - Lincoln Nebraska | Our early partners are receiving a discount rate with dedicated customer support. |
| SP020 | Drover | Drover - Virtual Fencing Ear Tags | No expensive base stations or infrastructure needed. Our tags use muscle contractions, not electric shock. |
| SP021 | Collie | Collie | Manage your herd with just a collar and an app | Collie is the only system in Europe that combines Virtual Fencing with Virtual Herding. |
| SP022 | Ranchflow | AI-Powered Livestock Monitoring & Digital Fencing | Smart collars with AI-powered health monitoring and virtual fencing technology. |
| SP023 | CMK | Smart Farming Investment: Making Sense of Cow Wearables for your Bottom Line | Advanced systems with virtual fencing (Halter): Recently reduced to $9.90 per cow per month ($118.80 annually). |
| SP024 | New Zealand Veterinary Association | Position Statement: Virtual Fencing and Virtual Herding for Cattle | |
| SP025 | Dairy Australia | Virtual fencing and herding option now available to all dairy farmers | With legalisation now in NSW, South Australia and Victoria, we can expect broader adoption and new market entrants. |
| SP026 | Halter | Halter launches world first virtual fencing via satellite | Halter’s virtual fencing collars now connect directly to satellites, unlocking limitless fencing in any terrain. |
| SI001 | Halter | Halter raises $220M in Series E to accelerate global expansion of virtual fencing | Halter serves more than 2000 ranchers and farmers across New Zealand, Australia, and the U.S., with a million of its solar-powered collars now sold. |
| SI002 | Business Wire | Halter Raises $220M in Series E to Accelerate Global Expansion of Virtual Fencing | Halter plans to hire more than 200 people - its largest-ever hiring effort - with a focus on product, engineering, and customer roles at its Auckland HQ. |
| SI003 | Halter | Halter raises $165M in funding to help farmers boost productivity | |
| SI004 | Halter | Dairy Farm ROI Study & Productivity Insights | Halter® | These farms using Halter achieved on average: 9% increase in pasture eaten, 9.5% more milk solids per hectare, 13% lift in profit before tax. |
| SI005 | Halter | Independent Research Confirms Profit Gains from Halter Technology on NZ Dairy Farms | On average these gains resulted in an increase in farm profit before tax of more than 13%. |
| SI006 | Halter | Farm Advisory Tools for Farm Consultants and Vets | Halter® | Through the Halter app, you can view real-time data on daily farm and paddock APC, live feed wedge, daily growth rates, and live stock numbers. |
| SI007 | Halter | Halter for Beef Farms | Halter® | |
| SI008 | Halter | Dairy Farm Automation & Herd Management | Halter® | |
| SI009 | AgFunderNews | Halter says it's not an agtech company on the heels of $220m Series E | Halter’s business model is subscription-based, with pricing starting at around $6-10 per cow per month depending on herd size and market, plus a one-time infrastructure fee for on-farm towers. |
| SI010 | Promus Ventures | Halter Raises $220M Series E at $2B Valuation - Physical AI is Eating the Farm and Ranch | Promus Ventures | |
| SI011 | iStart | Halter ramps up hiring after Thiel-backed US$220m raise | The company has positioned the systems as full herd management, from a smartphone, at US$5 to US$8 per animal per month. |
| SI012 | U.S. Securities and Exchange Commission | Merck & Co., Inc. 2025 Form 10-K | |
| SI013 | U.S. Securities and Exchange Commission | Merck & Co., Inc. Q1 2026 Form 10-Q | Three months ended March 31, 2026: Sales 16,286; Cost of sales 4,195; Selling, general and administrative 2,700; Research and development 12,592. |
| SI014 | High Plains Journal | Halter uses satellite technology for virtual fencing - High Plains Journal | Combined with a suite of new tools for reproduction, animal behavior and precision pasture management, the release expands what is possible for cattle ranch management. |
| SI015 | Halter | Built for Beef: How Halter created virtual fencing purpose-built for beef farmers | Beef required its own product, its own engineering, and its own way of thinking. |
| SI016 | CMK | Smart Farming Investment: Making Sense of Cow Wearables for your Bottom Line | Advanced systems with virtual fencing (Halter): Recently reduced to $9.90 per cow per month ($118.80 annually). |
| SI017 | Farm Progress | Virtual fencing gains ground in U.S. despite high costs | In hearing the numbers, 70-80 bucks a head, if you want to turn out 80-100 cattle on a piece of land, that’s like $8,000 a year. |
| SI018 | SmartCompany | Holy cow! NZ agtech startup Halter raises $314.4 million at $2.9 billion valuation | |
| SI019 | Farmers Weekly | Halter eyes fresh pastures after huge capital raise | Investment will continue across product development, including animal health monitoring and pasture management, shaped by how customers are using the system in the field. |
| SI020 | Halter | Careers | Halter® | |
| SI021 | Rural News | Halter Raises NZ$377M to Expand Virtual Fencing Globally | |
| SI022 | TechCrunch | Peter Thiel's big bet on solar-powered cow collars | TechCrunch | |
| SI023 | New Zealand Veterinary Association | Position Statement: Virtual Fencing and Virtual Herding for Cattle | |
| SI024 | DTN / Progressive Farmer | Halter's Satellite-Connected Cattle Collars Could Expand US Beef Market Coverage | This makes it available to them at nearly the same prices as before for the collar and subscription. And there's no need now for a tower. |
| SI025 | AGDaily | Virtual fencing goes fully off-grid with Halter’s satellite launch | AGDaily | |
| SI026 | Beef Magazine | Halter launches first virtual fencing via satellite | |
| SE001 | Halter | Industry Leading Technology - Now Direct to Satellite | Halter® | Each collar collects and sends over 6,000 data points every minute to Halter’s cloud-based data platform. |
| SE002 | Halter | Patents | Halter® | |
| SE003 | Halter | What's New in Halter - Latest Features | Halter® | |
| SE004 | Halter | Halter's Animal Welfare Charter: Veterinary Advisory Board | |
| SE005 | Halter | Halter's Animal Welfare Charter: Animal Health Benefits | |
| SE006 | Halter | Halter's Animal Welfare Charter: Training Animals | |
| SE007 | Halter | Halter's Animal Welfare Charter: Leadership and Governance | |
| SE008 | Halter | Halter's Animal Welfare Charter: Legislative Compliance | |
| SE009 | Halter | State of regulation in Australia | |
| SE010 | Halter | From escape rooms to agritech: Meet Dylan, Software Engineer at Halter | |
| SE011 | Halter | Software Usage Policy for Rural Professionals | |
| SE012 | Google Play | Halter - Apps on Google Play | |
| SE013 | Apple App Store | Halter App - App Store | |
| SE014 | TechCrunch | Peter Thiel's big bet on solar-powered cow collars | |
| SE015 | DTN / Progressive Farmer | Halter's Satellite-Connected Cattle Collars Could Expand US Beef Market Coverage | |
| SE016 | Farm Progress | Virtual fencing gains ground in U.S. despite high costs | Ranchers need to consider return on investment before fully committing on their operations. |
| SE017 | Justia Patents | Patents Assigned to HALTER USA INC | |
| SE018 | Dairy Australia | Virtual fencing and herding option now available to all dairy farmers | |
| SE019 | VetSalus | Halter! Who goes there? | |
| SE020 | BusinessWire | U.S. Ranchers Create 11,000 Miles of Virtual Fencing With Halter’s Smart Cattle Collars | |
| SE021 | WireMeshNews | Virtual Fencing for Cattle — Costs, Vendors, Trials & 2026 Outlook | |
| SE022 | Halter | Halter's Animal Welfare Charter: Virtual Fencing Background | |
| SE023 | Halter | Halter's Animal Welfare Charter: References | |
| SE024 | Halter | Halter's Animal Welfare Charter | |
| SE025 | Halter | Reduce Farm Workload | Halter® | |
| SE026 | Halter | Harvest More Pasture | Halter® | |
| SE027 | Halter | Improve Mating Results | Halter® | |
| SU001 | Halter | Halter Rancher Stories | Halter® | |
| SU002 | Halter | Halter Farmer Stories | Halter® | |
| SU003 | Halter | Drovers features Grace Magruder's story of five generations of women ranching in California | |
| SU004 | Halter | WyoFile explores how Halter's virtual fencing is transforming Wyoming ranching | |
| SU005 | Halter | 3 ways ranchers are using Halter for winter grazing and cattle management | |
| SU006 | Halter | Two Smart Ways Beef Ranchers Are Using Virtual Fencing (That You Might Not Know About) | |
| SU007 | Halter | 7 Ways Dairy Farmers Are Improving Profitability with Virtual Fencing | |
| SU008 | Halter | Halter brings world-leading virtual fencing technology to U.S. ranchers | |
| SU009 | Halter | Halter expands '1 Year Free' competition winners, awards four standout farmers | |
| SU010 | Halter | Halter partners with ranchers and conservation groups on Rio Grande del Norte National Monument | |
| SU011 | Halter | Ranch Management Resource Initiative | Halter® | |
| SU012 | YouTube | Halter's virtual fencing saved $210K: How one California rancher built 7 miles of fence | |
| SU013 | Apple App Store | Halter App - App Store | |
| SU014 | Google Play | Halter - Apps on Google Play | |
| SU015 | AppBrain | Halter: Free Android Business App - APK Info & Stats | |
| SU016 | BusinessWire | U.S. Ranchers Create 11,000 Miles of Virtual Fencing With Halter’s Smart Cattle Collars | |
| SU017 | Animal AgTech | Halter’s Breakout Year: From Market Penetration to Industry Validation | |
| SU018 | DCVC | Halter’s invisible fencing drives an inevitable shift in ranching | |
| SU019 | Farm Progress | Virtual fencing gains ground in U.S. despite high costs | |
| SU020 | VetSalus | Halter! Who goes there? | |
| SU021 | Stuff | Petition opposes virtual fencing | |
| SU022 | WireMeshNews | Virtual Fencing for Cattle — Costs, Vendors, Trials & 2026 Outlook | |
| SU023 | DTN / Progressive Farmer | Halter's Satellite-Connected Cattle Collars Could Expand US Beef Market Coverage | |
| SU024 | Halter | Virtual Fencing 101: How it works, and why it’s a game-changer for farmers | |
| SU025 | Halter | NSW Farmers travel to State Parliament to press their case for virtual fencing | |
| SR001 | Halter | Halter raises $220M in Series E to accelerate global expansion of virtual fencing | Halter serves more than 2000 ranchers and farmers across New Zealand, Australia, and the U.S., with a million of its solar-powered collars now sold. |
| SR002 | Business Wire | Halter Raises $220M in Series E to Accelerate Global Expansion of Virtual Fencing | Halter plans to hire more than 200 people - its largest-ever hiring effort - with a focus on product, engineering, and customer roles at its Auckland HQ. |
| SR003 | AgFunderNews | Halter says it's not an agtech company on the heels of $220m Series E | Halter shouldn’t be limited by the typical boundaries associated with agtech investment or identity. |
| SR004 | TechCrunch | Peter Thiel's big bet on solar-powered cow collars | TechCrunch | Halter’s collar is on more than a million cattle across more than 2,000 farms in New Zealand, Australia, and the United States, where the company operates in 22 states. |
| SR005 | Halter | Halter launches world-first virtual fencing via satellite | Halter’s internal modelling estimates direct-to-satellite capability expands coverage of the U.S beef cattle market by 2.5x. |
| SR006 | High Plains Journal | Halter uses satellite technology for virtual fencing | Using Starlink, the new technology enables ranchers to manage cattle anywhere they can see the sky. |
| SR007 | Halter | Halter partners with the Foundation for America’s Public Lands and the Bureau of Land Management to expand joint access in California | The partnership includes $2.7 million in funding to support ranchers using Halter on BLM-managed land. |
| SR008 | Halter | Halter partners with ranchers and conservation groups on Rio Grande del Norte National Monument | |
| SR009 | Halter | Careers | Halter® | Halter is not your normal 9 to 5, and we understand this is not for everyone. |
| SR010 | Halter | Farm Advisory Tools for Farm Consultants and Vets | Halter® | Through the Halter app, you can view real-time data on daily farm and paddock APC, live feed wedge, daily growth rates, and live stock numbers. |
| SR011 | Halter | Animal welfare is central to Halter | The maximum strength of a single pulse is 0.45 joules, which is significantly less energy than the shock received from a typical mains-powered electric fence. |
| SR012 | Halter | Animal Welfare Charter: Virtual Fencing Research | |
| SR013 | Animal Welfare Committee / GOV.UK | AWC opinion on the welfare implications of using virtual fencing for livestock | This opinion considers whether virtual fencing can be used without detriment to livestock health and welfare in the UK; focussed principally on its use for cattle, but sheep and goats are also considered. |
| SR014 | New Zealand Veterinary Association | Position Statement: Virtual Fencing and Virtual Herding for Cattle | The NZVA also acknowledges that current literature is mostly short-term and often industry-funded, with limited independent, long-term research. |
| SR015 | Dairy Australia | Virtual fencing and herding option now available to all dairy farmers | With recent legislative changes in New South Wales, South Australia and Victoria, these innovations are now legal in all six dairying states. |
| SR016 | U.S. Department of Agriculture | USDA, DOI Move to Boost Support for American Ranchers, Help Lower Prices for Consumers | The agreement promotes targeted grazing to reduce wildfire risk, supports reopening vacant allotments, and encourages adoption of innovative technologies such as virtual fencing. |
| SR017 | U.S. Small Business Administration Office of Advocacy | BLM Proposes Revisions to Grazing Regulations | The BLM acknowledges that the proposed rule may significantly affect small ranching and livestock operations that hold grazing permits or leases on federal public lands. |
| SR018 | Congressional Research Service | The 2026 Farm Bill (H.R. 7567): Comparison with Current Law | |
| SR019 | University of Idaho | Virtual fencing study targets public land grazing conflicts | They are using virtual fencing technology to guide the grazing of cattle. |
| SR020 | North Dakota State University | Grazing with Virtual Fence | Halter ... $4,500/tower ... $50/month internet fee/site ... $1,000 delivery fee ... $6/collar/month ... $11.08. |
| SR021 | PERC | Virtual Fencing for Conservation | Virtual fencing must work for producers, make sense for their bottom lines, and become widely adopted before it can become a conservation tool at scale. |
| SR022 | Colorado State University | How does virtual fencing work for cattle? | Several collars fell off, some seemed to malfunction after being exposed to too much moisture, and in others the battery died quicker than he was expecting. |
| SR023 | Rangelands Gateway / University of Arizona | Virtual Fence Vendors Basic Comparison (March 2025) | Virtual fence components from different manufacturers are generally not interoperable or interchangeable. |
| SR024 | Gallagher | eShepherd Virtual Fencing | Gallagher United States | |
| SR025 | Merck Animal Health | Vence - Merck Animal Health USA | |
| SR026 | WireMeshNews | Virtual Fencing Gains Ground in the U.S. — Costs, Vendors, Trials & 2026 Outlook | |
| SR027 | Farm Progress | Virtual fencing gains ground in U.S. despite high costs | There is a standard cost of $72 per year per collar with a required three-year commitment. For the required towers, there is a one-time $4,500 price per tower. |
| SR028 | Otago Daily Times | Halter package price cut 37% | A virtual fencing, remote shifting, pasture management, heat detection and health monitoring package will drop to $9.90 a cow, per month — back 37%. |
| SR029 | Agtech Industry Examiner | Halter’s $220 Million Bet on a World Without Wire | So yes, virtual fencing can work. But it is not plug-and-play magic. |
| SR030 | New Zealand Companies Office | HALTER LIMITED (6063230) company extract | Ultimate holding company: Halter USA inc. |
| SR031 | iStart | Halter ramps up hiring after Thiel-backed US$220m raise | Halter is hiring for 220-plus roles across New Zealand, Australia, and the US in coming weeks. |
| SV001 | Halter | Halter raises $220M in Series E to accelerate global expansion of virtual fencing | Halter serves more than 2000 ranchers and farmers across New Zealand, Australia, and the U.S., with a million of its solar-powered collars now sold. |
| SV002 | Business Wire | Halter Raises $220M in Series E to Accelerate Global Expansion of Virtual Fencing | Halter plans to hire more than 200 people - its largest-ever hiring effort - with a focus on product, engineering, and customer roles at its Auckland HQ. |
| SV003 | AgFunderNews | Halter says it's not an agtech company on the heels of $220m Series E | Halter shouldn’t be limited by the typical boundaries associated with agtech investment or identity. |
| SV004 | TechCrunch | Peter Thiel's big bet on solar-powered cow collars | TechCrunch | Halter’s collar is on more than a million cattle across more than 2,000 farms in New Zealand, Australia, and the United States, where the company operates in 22 states. |
| SV005 | SmartCompany | NZ agtech startup Halter raises $315 million at $2.9 billion valuation | Halter was founded in 2016 by CEO Craig Piggott, and its virtual farm fences are now used across more than 2,000 cattle farms in New Zealand, Australia and the US. |
| SV006 | Rural News Group | Halter Raises NZ$377M to Expand Virtual Fencing Globally | Halter now serves more than 2000 farmers and ranchers across New Zealand, Australia, and the United States, with one million of its solar-powered collars now sold. |
| SV007 | Farmers Weekly | Halter eyes fresh pastures after huge capital raise | Halter is hiring for 220-plus roles across New Zealand, Australia, and the US in coming weeks. |
| SV008 | Halter | Halter raises $165M in funding to help farmers boost productivity | Halter has raised $165M in a Series D fundraising round, valuing Halter at $1.65billion (USD $1 billion). |
| SV009 | Halter | Independent Research Confirms Profit Gains from Halter Technology on NZ Dairy Farms | Farms recorded an average 13.2% increase in profit before tax (EBIT). |
| SV010 | New Zealand Companies Office | HALTER LIMITED (6063230) company extract | Ultimate holding company: Halter USA inc. |
| SV011 | Farm Progress | Virtual fencing gains ground in U.S. despite high costs | There is a standard cost of $72 per year per collar with a required three-year commitment. For the required towers, there is a one-time $4,500 price per tower. |
| SV012 | Otago Daily Times | Halter package price cut 37% | A virtual fencing, remote shifting, pasture management, heat detection and health monitoring package will drop to $9.90 a cow, per month — back 37%. |
| SV013 | North Dakota State University | Grazing with Virtual Fence | Halter ... $4,500/tower ... $50/month internet fee/site ... $1,000 delivery fee ... $6/collar/month ... $11.08. |
| SV014 | Agtech Industry Examiner | Halter’s $220 Million Bet on a World Without Wire | So yes, virtual fencing can work. But it is not plug-and-play magic. |
| SV015 | Tech Funding News | Founders Fund leads Halter’s $220M round at $2B valuation for AI cow collars in one of agtech’s largest fundings | This brings the valuation to $2 billion, making it the largest in the global agtech sector. |
| SV016 | CompaniesMarketCap | Merck (MRK) - Market capitalization | As of May 2026 Merck has a market cap of $293.21 Billion USD. |
| SV017 | Macrotrends | Merck Revenue 2011-2025 | MRK | Merck revenue for the twelve months ending September 30, 2025 was $64.235B. |
| SV018 | CompaniesMarketCap | Deere & Company (John Deere) (DE) - Market capitalization | As of May 2026 Deere & Company (John Deere) has a market cap of $146.44 Billion USD. |
| SV019 | Macrotrends | Deere Revenue 2010-2025 | DE | Deere revenue for the twelve months ending July 31, 2025 was $44.433B. |
| SV020 | CompaniesMarketCap | AGCO (AGCO) - Market capitalization | As of May 2026 AGCO has a market cap of $8.13 Billion USD. |
| SV021 | Macrotrends | AGCO Revenue 2010-2024 | AGCO | AGCO revenue for the twelve months ending December 31, 2024 was $11.662B. |
| SV022 | CompaniesMarketCap | Trimble (TRMB) - Market capitalization | As of May 2026 Trimble has a market cap of $13.14 Billion USD. |
| SV023 | Macrotrends | Trimble Revenue 2012-2025 | TRMB | Trimble revenue for the twelve months ending September 30, 2025 was $3.601B. |
| SV024 | Macrotrends | Merck Price to Sales Ratio 2012-2025 | MRK | |
| SV025 | Macrotrends | Deere Price to Sales Ratio 2012-2025 | DE | |
| SV026 | Macrotrends | AGCO Price to Sales Ratio 2010-2024 | AGCO | |
| SV027 | Macrotrends | Trimble Price to Sales Ratio 2010-2025 | TRMB | |
| SV028 | Merck Animal Health | Vence - Merck Animal Health USA | |
| SV029 | Gallagher | eShepherd Virtual Fencing | Gallagher United States | |
| SV030 | U.S. Securities and Exchange Commission | Merck & Co., Inc. 2025 Form 10-K |