Customer-controlled monitoring for BESS sites when periodic review is no longer enough.
Use Horizon when periodic review is no longer enough. It keeps the same diagnostic logic close to the asset, on infrastructure the operator controls.
If scope is still unclear, start with Verify and let the site earn continuous coverage.
Best fit when repeated operating pressure or hard data-boundary rules make local signal worth carrying.
Thermal anomaly watch
Surface thermal and chemistry drift earlier, before the site tells the story the hard way.
Resistance trend tracking
Keep resistance stress visible across cycles and operating windows.
Controlled deployment boundary
Choose connected or isolated deployment without moving response out of your existing safety stack.
Continuous monitoring when the site truly needs it.
Earlier warning, local control, and a disciplined rollout path
Continuous local signal for sites where repeated pressure makes it worth carrying.
Earlier signal detection
Keep drift visible before it becomes an outage, safety event, or dispute.
The problem shows up late.
Deployment control
Run on infrastructure you control, connected or isolated.
Permanent cloud dependence even when local control matters.
Commercial fit
Start where repeated pressure proves continuous monitoring is worth it.
The platform is sold before the operating case is proven.
Use Horizon when the site keeps needing the same answer.
Horizon is not the default upsell. Use it only when repeated availability, revenue, warranty, or data-boundary pressure makes continuous local signal worth the operating load.
Merchant or availability pressure keeps recurring
Use Horizon when revenue swings, availability drift, or repeated underperformance make continuous signal more valuable than occasional review.
The site keeps re-entering technical review cycles
If lenders, insurers, principals, or warranty stakeholders keep revisiting the same asset questions, Horizon keeps a cleaner evidence trail between events.
Multiple stakeholders need one operating view
Portfolio operators, owner teams, and remote principals benefit from one on-premise signal source rather than another spreadsheet relay.
Data custody or critical-power rules are hard constraints
If cloud dependency is a non-starter, Horizon gives you continuous analysis inside an operating boundary the site can defend.
Default rule
If the asset only needs an occasional diligence, warranty, insurer, or operating answer, stay with Verify. Horizon should follow only when ongoing coverage is justified.
Choose the operating boundary once the use case is clear
The same diagnostic logic can run inside a connected or isolated operating model. The right choice depends on who needs visibility and what the site boundary allows.
Connected deployment
On-site monitoring with a secure upstream path.
Use this model when merchant or portfolio-operated sites need local analysis plus a centralized view for remote operators or asset managers.
Isolated deployment
No cloud dependency. No routine data egress.
Use this model for critical-power, sovereign, or tightly controlled sites where routine data egress is off the table.
Verify is the fixed entry point. Horizon starts from a scoped pilot when continuous coverage is justified.
Verify is S$4,800 for the standard scope. Horizon pilots typically start from S$35,000 and stay scoped around telemetry path, deployment boundary, and reporting needs.
Horizon pilot
Pilot deployment for one site or one operating environment.
Scoped pilot starting fee
Typical one-site pilot starts here. Final scope depends on telemetry path, deployment boundary, and reporting needs.
- Builds on Verify findings and pilot goals
- Monitoring design for site topology and duty cycle
- Continuous evidence trail for operators, principals, and technical reviewers
- Trend tracking across charge cycles
- Connected or isolated deployment planning aligned to site controls
- Structured pilot scope for one site or asset block
- Clear expansion path to a broader rollout if justified
Target Profile
Method evidence
The monitoring story should point back to the underlying method.
Horizon is strongest when buyers can see it as the continuous extension of the same benchmark-backed logic used in Verify.
Disciplined public claim boundary.
Evidence limit
Method stack
ICA and chemistry drift
Horizon is only credible if the continuous monitoring story is visibly tied back to the chemistry-fingerprint layer that Verify uses in deeper review.
Dataset
Oxford Battery Degradation Dataset
Evidence reference
Oxford ICA Peak Shift
Resistance and precursor logic.
Precursor sequence
Operating stress
Resistance and anomaly logic
Continuous monitoring matters because DCIR, abnormal transitions, and precursor behaviour often move before the dashboard story catches up.
Dataset
NASA PCoE Battery Dataset
Evidence reference
NASA Resistance Transition
Disciplined public claim boundary.
Evidence limit
Claim boundary
Monitoring is stronger when the evidentiary limits are explicit
Horizon should read like the ongoing extension of a forensic method stack with visible public anchors and disciplined claim limits.
Dataset
Oxford + NASA public reference stack
Evidence reference
BESS Mastery Method Chart Atlas
Principal briefing:
Scope, method, and deployment posture
A concise principal-led briefing on what the work can support, how the method holds up, and where Verify and Horizon fit.
Covers telemetry limits, main degradation and operating-risk patterns, plus how Verify and Horizon fit the engagement path.

Focus
BESS-First Review
Scope first
Defined review scope
Boundary, telemetry window, and mandate question are pinned down before conclusions move.
Encrypted handling
Protected review workflow
Review traffic and operating data are handled with encrypted transfer and controlled access.
Customer boundary
Customer-controlled deployment
Managed, private, and isolated deployment paths are available when the environment requires them.
Direct accountability
Principal sign-off
Technical accountability stays close to the method rather than disappearing into a generic workflow.