Skip to content

Trust Center

Service Level Agreement

Effective date: 2026-06-01 · Applies to all HelmXP subscription tiers

Standard SLA

99.9%

monthly uptime target

Enterprise SLA

99.95%

monthly uptime target

Max credit

50%

of monthly fee, per subsystem

1. Uptime commitment

HelmXP guarantees the following monthly uptime for the platform's core services. “Monthly uptime” is calculated as the percentage of minutes in a calendar month during which the core services are available, excluding scheduled maintenance windows (see Section 4).

TierMonthly uptime target
Starter, Core, Growth99.9%
Enterprise99.95%

Uptime is measured per subsystem. The following subsystems are covered:

SubsystemDescription
apiThe platform API and portal
authAuthentication and session services
databaseAurora/RDS Postgres availability
storageS3 object storage
emailOutbound email delivery
ragRAG retrieval and grounding pipeline
pdf_exportBriefing PDF generation
scimSCIM provisioning sync

2. Service credit schedule

If HelmXP's monthly uptime for any covered subsystem falls below the applicable target, the subscriber is eligible for a service credit against the next invoice. Credits are calculated on the subscriber's monthly subscription fee. Credits are not refunded as cash.

Credit formula: 10% of the monthly fee per 0.1 percentage point of uptime below the applicable target, capped at 50% of the monthly fee regardless of actual outage duration.

Standard tiers

Monthly uptimeCredit
99.8% – 99.9% (SLA met)0%
99.7% – < 99.8% (0.1% below SLA)10%
99.6% – < 99.7% (0.2% below SLA)20%
99.5% – < 99.6% (0.3% below SLA)30%
99.4% – < 99.5% (0.4% below SLA)40%
< 99.4% (0.5% or more below SLA)50%

Enterprise tier

Monthly uptimeCredit
99.85% – 99.95% (SLA met)0%
99.75% – < 99.85% (0.1% below SLA)10%
99.65% – < 99.75% (0.2% below SLA)20%
99.55% – < 99.65% (0.3% below SLA)30%
99.45% – < 99.55% (0.4% below SLA)40%
< 99.45% (0.5% or more below SLA)50%

3. Credit eligibility and request process

To be eligible for a service credit, the subscriber must:

  1. Submit a credit request to support@helmxp.com within 30 days of the end of the affected calendar month.
  2. Include in the request: institution name and hub slug, the calendar month for which the credit is claimed, and a brief description of the service impact observed.

HelmXP validates claims against the status_events table, which records start time, end time, and component for every platform incident. Downtime is computed from these records. Credits are issued within 15 business days of a validated request.

4. Exclusions

The following do not count toward downtime for purposes of this SLA:

  • Scheduled maintenance announced at least 7 days in advance via the HelmXP status page and email notification to the CEO seat.
  • Customer-caused outages — including incorrect SAML configuration, blocked IP ranges in the hub's IP allowlist, or deliberate suspension of service due to non-payment.
  • Force majeure — events outside HelmXP's reasonable control, including natural disasters, acts of government, widespread internet failures not limited to HelmXP's cloud infrastructure, or actions of third-party providers (for example, cloud-provider service events that affect multiple customers globally).
  • Third-party integrations not operated by HelmXP, including Plansmith, NCUA data feeds, or any other data source the subscriber connects.
  • Trial periods — the free 30-day trial for Starter and Core tiers is provided as-is, without uptime guarantees.

5. Measurement methodology

Downtime is measured from status_events.started_at to status_events.resolved_at for events with status = 'down' or status = 'degraded'.

  • Events with status = 'down' count fully toward downtime.
  • Events with status = 'degraded' count at 50% weight (the subsystem was impaired but functional).
  • Events not resolved within a calendar month carry the open duration through to the timestamp at which they are resolved.
uptime_pct = 1 − (downtime_minutes / total_minutes_in_month)

For the standard 99.9% SLA, a 31-day month has 44,640 minutes. The maximum allowable downtime is 44.6 minutes (approximately 44 minutes 36 seconds).

6. Enterprise SLA terms

Enterprise subscribers receive the 99.95% uptime target and may negotiate additional terms, including:

  • Dedicated support contacts
  • Custom maintenance windows
  • Named SLO targets for specific subsystems

These are documented in the subscriber's order form or enterprise addendum, which supersedes this general SLA where they conflict.

7. Changes to this SLA

HelmXP will provide at least 30 days' notice of material changes to this SLA. Continued use of HelmXP after the effective date of a change constitutes acceptance.

Questions about this SLA: support@helmxp.com