Robotics Why Now: Deployment KPI Catalyst v1

Date: 2026-06-12 Owner: Hugo / Genius Team Research owner: Finance / Charlie AGT-002 Status: public-safe research artifact for slide 3 / site section Visibility: PUBLIC Related files:

• knowledge/robotics-research-harness.md
• knowledge/robotics-why-now-ai-model-catalyst-v1.md
• knowledge/robotics-why-now-hardware-cost-catalyst-v1.md
• knowledge/robotics-figure-ai-deep-dive-v1.md
• knowledge/robotics-agility-digit-evidence-v1.md
• knowledge/robotics-ubtech-walker-evidence-v1.md

Public-safety: yes. No Hugo private portfolio data. No trade recommendation.

0. One-line answer

截至 2026-06-12，“Why robotics now” 不应只讲 AI 模型或低成本硬件；第三个高质量 catalyst 是：humanoid / embodied robotics 开始出现可量化的客户现场 KPI、商业部署语言、收入 / 销量披露，而不是只停留在 demo video。

最有用的公开证据链：

1. Figure AI：BMW Spartanburg 11 个月部署，10-hour shifts Monday-Friday、90,000+ parts loaded、1,250+ runtime hours、contribution to 30,000+ BMW X3 vehicles；BotQ 披露 350+ Figure 03 delivered、cadence from 1/day to 1/hour、EOL first-pass yield >80%、battery-line first-pass yield 99.3%。🟢 Figure official posts dated 2025-11-19 and 2026-04-29; checked in existing knowledge base 2026-06-10.
2. Agility Robotics：GXO multi-year RaaS agreement after proof-of-concept pilot; later disclosed Digit moved 100,000+ totes at GXO Flowery Branch facility; Mercado Libre and Toyota Motor Manufacturing Canada added commercial-agreement / RaaS language. 🟢 Agility official releases dated 2024-06-27, 2025-11-20, 2025-12-10, 2026-02-19; checked 2026-06-10.
3. UBTECH / 优必选：2025 annual report disclosed full-size embodied intelligent humanoid robot products/services revenue RMB 820.557m, 41.1% of 2025 revenue, +2,203.7% YoY; 1,079 full-size embodied humanoid units sold in 2025; annualized full-size humanoid capacity over 6,000 units at end-2025. 🟢 UBTECH Annual Report 2025, disclosed 2026-04-15; checked 2026-06-11.

Current conclusion: this is S4-heavy deployment / commercial-application evidence, not S5 scaled-commercial-economics proof. The missing S5 chain remains customer-confirmed ROI/payback, uptime, intervention rate, repeat orders, contract value, service cost, robot revenue durability, and humanoid-specific gross margin. 🟠 Charlie synthesis as of 2026-06-12.

1. Core question

Why does deployment KPI matter as a “why now” catalyst?

Because robotics cycles usually overheat when investors extrapolate from demos, TAM slides, or product specs. The current cycle is more interesting because at least three evidence types are becoming measurable:

• customer-site task output: parts loaded, runtime hours, totes moved;
• commercial structure: RaaS / commercial agreement / named deployment site;
• financial filing anchors: humanoid revenue, unit sales, capacity and margin context.

This changes the research question from:

“Can a humanoid do a cool task on video?”

To:

“Can a robot repeat valuable work in a customer environment, with measurable utilization and economics?”

That is a better investment-research question, but it still does not justify trade recommendations.

2. Evidence table: deployment KPI as a catalyst

3. Why this is a real catalyst, not just newsflow

3.1 It changes the unit of evidence

Old weak evidence unit:

• one edited demo video;
• one customer logo;
• one TAM chart;
• one “soon mass production” statement.

Stronger 2024-2026 evidence unit:

• task volume: 90,000+ parts loaded; 100,000+ totes moved;
• runtime: 1,250+ hours;
• shift schedule: 10-hour shifts Monday-Friday;
• production process: first-pass yield, battery-line yield, actuator output;
• filing metrics: revenue, units sold, production capacity.

Interpretation: the sector is moving from narrative-only tracking into KPI tracking. 🟠 Charlie synthesis, as of 2026-06-12.

3.2 It creates falsifiable follow-up questions

A demo can always be followed by another demo. A KPI creates a trackable chain:

1. Can the same customer repeat / expand deployment?
2. Can a second and third customer show similar task KPI?
3. Does uptime improve and intervention rate fall?
4. Does deployment time fall from bespoke engineering to repeatable playbook?
5. Does revenue convert into gross margin after service / warranty / support cost?

This is why deployment KPI is more useful than hype: it can be falsified.

3.3 It separates three different paths to commercialization

The paths are complementary, not rankings. Figure may show stronger task KPI; Agility may show clearer RaaS language; UBTECH may show stronger public-company financial disclosure. None alone closes the S5 chain.

4. Signal vs noise

Signals

1. Named customer + task KPI + duration: Figure BMW 11 months, 1,250+ hours, 90,000+ parts loaded. 🟢 S4.
2. Commercial/RaaS language after pilot: Agility GXO and TMMC. 🟢 S4.
3. Throughput milestone: Agility Digit 100,000+ totes moved at GXO Flowery Branch. 🟢 S4.
4. Manufacturing KPI: Figure 350+ robots, 1/day to 1/hour cadence, >80% EOL FPY, 99.3% battery-line FPY. 🟢 S4.
5. Filing-backed humanoid revenue / unit sales: UBTECH RMB 820.557m and 1,079 units in 2025. 🟢 S4.

Noise unless upgraded

1. Customer logo without robot count, task KPI, runtime, or economics. 🟢/🟠.
2. “Commercial agreement” without deployment schedule, volume, contract value, or KPI. 🟢/🟠.
3. Production capacity without actual output / accepted deliveries. 🟢/🟠.
4. Revenue spike without repeat-order and margin durability evidence. 🟢/🟠.
5. Vendor-disclosed KPI treated as customer-confirmed ROI. 🟢 for vendor disclosure; 🟠 for ROI inference.

5. What would change our mind

Upgrade toward S5 if:

• Figure or customers disclose repeat deployments with robot count, customer-confirmed ROI/payback, uptime, intervention rate, and repeat-order terms. 🟢/🟠 depending on source.
• Agility discloses deployed robot count, RaaS pricing / contract value, uptime, intervention rate, and gross-margin or contribution-margin proxy. 🟢/🟠.
• UBTECH discloses sustained 2026 humanoid revenue growth, humanoid-specific gross margin, customer repeat orders, service/warranty cost, and production-use KPIs. 🟢.
• Multiple companies converge on comparable KPI disclosure: runtime, task count, intervention rate, payback, service cost, gross margin. 🟢/🟠.

Downgrade if:

• Figure BMW remains the only deeply quantified customer KPI after 6-12 more months. 🟠.
• Agility RaaS deployments stay narrow, high-support, and economics-undisclosed. 🟠.
• UBTECH 2025 humanoid revenue proves lumpy and does not repeat in 2026, or cash burn worsens despite revenue growth. 🟢/🟠.
• Customer-side sources fail to confirm productivity, payback, or safety/reliability after vendor claims. 🟡/🟠.

6. Common misconceptions

1. “有客户现场 KPI，所以 humanoid 已经商业化成功。”

   • Correction: 客户现场 KPI 是 S4 强信号；S5 还需要 ROI/payback、低干预、复购、毛利率和服务成本。🟠.

2. “RaaS agreement 等于高质量 recurring revenue。”

   • Correction: RaaS 语言说明商业结构可能更接近服务化，但没有价格、利用率、服务成本和 churn / renewal 前，不能判断 revenue quality。🟢/🟠.

3. “UBTECH 有 RMB 8.206 亿 humanoid 收入，所以经济性已被验证。”

   • Correction: 报表收入和销量是强证据，但还缺 humanoid-specific margin、客户 ROI、复购和售后成本。🟢/🟠.

4. “Figure / Agility / UBTECH 可以直接排成 winner ranking。”

   • Correction: 三者证明的是不同证据层：deployment KPI、RaaS workflow、filing-backed revenue。公开输出应做 evidence map，不做 winner ranking。🟠.

7. Think Deeper questions

1. 如果 deployment KPI 是最重要的 “why now” 证据，行业应该统一披露哪些最低指标：runtime、task count、intervention rate、payback、还是 repeat order？
2. 早期 humanoid 的商业化会更像硬件销售，还是像 RaaS / managed service？哪一种更容易被服务成本吞噬毛利？
3. UBTECH 的报表披露会不会迫使 Figure、Agility、Tesla、Unitree 等公司开始披露更可比的商业指标？
4. 如果客户现场 KPI 增长但 ROI/payback 不披露，市场最容易高估哪一环：技术成熟度、需求强度、还是单位经济性？
5. 哪个 S5 信号最先出现会真正改变行业判断：repeat customer order、customer-confirmed ROI、low intervention rate、还是 humanoid-specific gross margin？

8. Public site section draft

Why robotics now: deployment evidence is becoming measurable

The robotics cycle is more interesting now because the evidence is no longer only demo videos. Some companies are starting to disclose operating metrics.

Figure gives one example: its BMW Spartanburg deployment ran for 11 months, with 10-hour shifts Monday-Friday, 90,000+ parts loaded, 1,250+ runtime hours, and contribution to 30,000+ BMW X3 vehicles. Agility gives another: GXO signed a multi-year RaaS agreement after a pilot, and Digit later moved 100,000+ totes at GXO’s Flowery Branch facility. UBTECH adds a public-company reporting anchor: its 2025 annual report disclosed RMB 820.557m of full-size embodied humanoid robot products/services revenue and 1,079 units sold.

This is not proof that humanoid robotics has reached scaled commercial economics. It is evidence that the right questions are becoming measurable. The next stage is not more videos; it is repeat orders, customer ROI, uptime, intervention rate, service cost, gross margin, and durable revenue.

Required footer:

Evidence map only. No winner ranking. No trade recommendation. Deployment KPI is stronger than demo, but S4 does not equal S5 scaled-commercial economics.

9. Slide-ready compression

Title:

Why robotics now: the evidence unit changed

Main visual:

Old evidence unit              New evidence unit
Demo video                     Runtime hours
Customer logo                  Task count
TAM slide                      RaaS / commercial agreement
Production claim               First-pass yield / units sold
Concept revenue                Filing-backed humanoid revenue

Current state: S4-heavy deployment evidence, not S5 scaled economics.

Three-card version:

1. Figure = customer-site KPI

   • BMW: 1,250+ runtime hours, 90,000+ parts, 10-hour shifts Mon-Fri. 🟢
   • Missing: ROI, contract value, intervention, margin. 🟠

2. Agility = RaaS / workflow KPI

   • GXO multi-year RaaS; 100,000+ totes moved. 🟢
   • Missing: robot count, pricing, uptime, margin. 🟠

3. UBTECH = filing-backed revenue / units

   • RMB 820.557m humanoid revenue; 1,079 units sold in 2025. 🟢
   • Missing: humanoid-specific margin, repeat-order economics, customer ROI. 🟠

10. Source list

Primary / official sources:

• Figure, “F.02 Contributed to the Production of 30,000 Cars at BMW”, 2025-11-19, checked 2026-06-10: https://www.figure.ai/news/production-at-bmw 🟢
• Figure, “Ramping Figure 03 Production”, 2026-04-29, checked 2026-06-10: https://www.figure.ai/news/ramping-figure-03-production 🟢
• Figure, “Figure Signs Agreement with Catalyst Brands to Scale Humanoid Operations”, 2026-05-26, checked 2026-06-10: https://www.figure.ai/news/figure-signs-agreement-with-catalyst-brands 🟢
• Agility Robotics, “GXO Signs Industry-First Multi-Year Agreement with Agility Robotics,” 2024-06-27, checked 2026-06-10: https://agilityrobotics.com/content/gxo-signs-industry-first-multi-year-agreement-with-agility-robotics 🟢
• Agility Robotics, “Digit Moves Over 100,000 Totes in Commercial Deployment,” 2025-11-20, checked 2026-06-10: https://agilityrobotics.com/content/digit-moves-over-100k-totes 🟢
• Agility Robotics, “Mercado Libre and Agility Robotics Announce Commercial Agreement to Deploy Humanoid Robots,” 2025-12-10, checked 2026-06-10: https://agilityrobotics.com/content/mercado-libre-and-agility-robotics-announce-commercial-agreement 🟢
• Agility Robotics, “Agility Robotics Announces Commercial Agreement with Toyota Motor Manufacturing Canada,” 2026-02-19, checked 2026-06-10: https://agilityrobotics.com/content/agility-robotics-announces-commercial-agreement-with-toyota-motor-manufacturing-canada 🟢
• UBTECH Robotics Corp Ltd Annual Report 2025, disclosed 2026-04-15; checked 2026-06-11. Official IR CDN copy: https://owebsite-cdn.ubtrobot.com/resources/file/2026/04/15/795097683329093.pdf 🟢

Internal synthesis sources:

• knowledge/robotics-figure-ai-deep-dive-v1.md, dated 2026-06-10. 🟠 synthesis from primary sources.
• knowledge/robotics-agility-digit-evidence-v1.md, dated 2026-06-10. 🟠 synthesis from primary sources.
• knowledge/robotics-ubtech-walker-evidence-v1.md, dated 2026-06-11. 🟠 synthesis from primary sources.

11. Public-safety check

Safe to package publicly if used as written:

• No Hugo private portfolio weights or private rationale.
• No buy / sell / hold language.
• No paid-report excerpts.
• No private-channel claims.
• Source grades and as-of dates included for material claims.
• Deployment KPI, commercial agreement, revenue, capacity and economics are separated.

Do not publish:

• “Humanoid commercialization is solved.”
• “Figure / Agility / UBTECH winner ranking.”
• “RaaS agreement proves high-quality recurring revenue.”
• “UBTECH group gross margin equals humanoid gross margin.”
• Any trade recommendation or implication.