Claim
Vendor claim, demo, announcement, or public signal.
Methodology
How Humanoids Watch turns evidence into buyer signals.
We classify public claims, deployment signals, readiness, risk, and missing proof so buyers can see what is worth watching, tracking, or evaluating.
Evaluation Path
From public claim to buyer decision.
Evidence
Customer signal, commercial agreement, pilot context, operating metric, or independent validation.
Assessment
Evidence quality, readiness, use-case fit, missing proof, and buyer risk.
Buyer signal
Ignore, watch, track, contact, or evaluate for a scoped pilot.
Score System
Scores are signals, not certifications.
Scores help compare public evidence and buyer readiness. They are not technical certification, procurement approval, or vendor endorsement.
Reality Score
0-100How real and substantiated is the robot?
- High score means
- Real hardware, repeated demos, customer signals, and stronger technical or commercial evidence.
- Low score means
- Concept, render, limited proof, or unclear hardware capability.
Deployment Readiness
0-100Can a buyer realistically evaluate it for a defined use case?
- High score means
- Named customer, clear workflow, commercial path, support model, and operating evidence.
- Low score means
- No buyer access, unclear task fit, no operating evidence, or missing commercial structure.
Evidence Quality
Weak / Moderate / Strong / Very StrongHow strong is the public evidence behind the claim?
- High score means
- Customer-confirmed, operating context, commercial signal, metrics, or independent validation.
- Low score means
- Vendor-stated, demo-only, vague partnership, or unverified claim.
Validation Ladder
Validation levels use individual L-values.
The ladder shows how close a public signal is to operating evidence without overstating what the evidence proves.
Unverified claim or concept.
Vendor-stated capability.
Controlled demo or lab signal.
Joint test or evaluation signal.
External pilot signal.
Customer-confirmed operating signal.
Commercial agreement signal.
Repeated deployment signal.
Operating metrics disclosed.
Independently validated deployment evidence.
Public Claims
Not every announcement is evidence.
Vendor announcements, demos, and partnerships can be useful signals, but they do not prove operating deployment unless the customer, workflow, scale, metrics, or commercial context is clear.
Useful signal, but not operating evidence.
Relevant if customer, task, and operating context are named.
Stronger when workflow and commercial context are visible.
Stronger when uptime, intervention rate, cycle time, or cost data is disclosed.
Buyer Recommendation
How buyer recommendations work.
Too speculative or weakly evidenced.
Relevant, but not buyer-actionable yet.
Meaningful signals exist, but proof remains incomplete.
Specific workflow or customer signal may justify direct engagement.
Strong enough for a scoped buyer discussion.
Red Flags
Red flags we track.
Missing uptime data
Unknown intervention rate
Unclear autonomy boundaries
Vague customer claim
No commercial model
Unclear safety boundary
Weak support model
Stale claim
Update Process
How assessments stay current.
- New public signal appears.
- Source and context are reviewed.
- Evidence quality and validation level are updated.
- Scores or buyer recommendation change if justified.
- Last reviewed date is updated.
This is buyer intelligence, not certification.
Humanoids Watch does not certify robots, approve deployments, or replace technical diligence. It helps buyers separate public evidence from claims and identify what still needs verification.
Use The Methodology
Use the methodology with the data.
Apply the scoring logic to deployment signals, robot profiles, use cases, and vendor watchlists.