1. DEFINITIONS

The following capitalized terms shall have the definitions set forth below for the purpose of this SLA:

• Service: The ReFresh Workforce Dynamics Intelligence API, Vector Database, and related reporting dashboards provided to the Customer.

• API Availability: The percentage of the Total Monthly Minutes during which the Core API is fully operational and capable of processing requests.

• Core API: The primary ReFresh API endpoint responsible for vector generation, data ingestion, and semantic retrieval.

• Monthly Uptime Percentage (MUP): Calculated as (Total Minutes in Month - Unscheduled Downtime) / Total Minutes in Month

• Scheduled Downtime: Unavailability of the Service due to ordinary maintenance, which is communicated to the Customer at least forty-eight (48) hours prior to such unavailability.

• Unscheduled Downtime (UDT): Any period when the Service is unavailable to Customer outside of Scheduled Downtime or Exclusions.

• Normal Business Hours: 9:00 a.m. to 5:00 p.m. Eastern Standard Time (EST) Monday through Friday, excluding public holidays.

2. SERVICE COMMITMENT (API AVAILABILITY)

Monthly Uptime Target

• ReFresh commits to making the Service available with a Monthly Uptime Percentage (MUP) of at least 99.5%.

• Uptime Calculation

  • API Availability is measured on a calendar month basis. Time is counted in whole minutes.

• Uptime Tier and Monthly Uptime Percentage (MUP):

  • Tier 1 (Target): ≥ 99.00%

  • Tier 2: < 98.00% but ≥ 99.00%

  • Tier 3: < 99.00%

3. SUPPORT & INCIDENT MANAGEMENT

Incident Priority Levels

• ReFresh shall use commercially reasonable efforts to classify all reported Incidents according to the following criteria:

• Priority Level, Description, and Impact

  • P1 (Critical): The Core API is completely unavailable (0% success rate) or suffering catastrophic data loss/corruption. Service is unusable. Complete halt of business processing.

  • P2 (High): The Core API is available but suffering severe, continuous degradation (e.g., >50% error rate or high latency preventing real-time use). Significant business function impairment.

  • P3 (Medium): A non-critical function is impaired (e.g., secondary dashboard data is slow, a non-core integration fails). Workaround is usually possible. Minor business impairment; core service operational.

  • P4 (Low): General questions, documentation issues, feature requests, or minor cosmetic bugs. Minimal or no impact on core service.

• Response & Resolution Targets

  • Response times are measured from the moment an Incident is logged in ReFresh's support system.

• Priority Level, Initial Response Time Target (IRTT), Mean Target Resolution (MTR)/Workaround Time

  • P1 (Critical): 1 Hour (24/7/365) (IRTT), 8 Hours (MTR)

  • P2 (High): 8 Hours (During Normal Business Hours) (IRTT), 24 Hours (MTR)

  • P3 (Medium): 2 Business Days (IRTT), Best Efforts (MTR)

  • P4 (Low): 3 Business Days (IRTT), Best Efforts (MTR)

4. EXCLUSIONS

Unscheduled Downtime shall not include any performance or availability issues arising from:

• Scheduled Downtime.

• Force Majeure Events (e.g., acts of God, war, acts of terror, widespread internet outages outside of ReFresh's control).

• Customer’s Data Sources: Any issues or unavailabilities caused by the Customer’s integrated third-party systems (e.g., Slack, Google Workspace, Jira).

• Customer’s Environment: Equipment, software, or network failure on the Customer's side.

• Customer Misuse: Use of the Service in a manner inconsistent with the documentation or in breach of the Agreement.

• Beta (Pilot) Services: Performance of any features designated as beta or experimental.

• Denial of Service: Issues resulting from a sustained or malicious denial of service attack on the Service.