Platform Availability, Performance, and Support Commitments
Last Updated: May 19, 2026
This Service Level Agreement ("SLA") defines the availability commitments, performance targets, support response times, and customer remedies for the AI Bizlutions platform. This SLA applies to all customers on paid plans (Professional, Agency, Enterprise).
Sandbox and free-trial accounts are not covered by this SLA.
The following services are covered by the 99.9% uptime SLA:
The following are NOT counted as downtime:
| Metric | Target | Measurement |
|---|---|---|
| API response time (p95) | < 3 seconds | Measured at API Gateway |
| API response time (p50) | < 500ms | Measured at API Gateway |
| Frontend load time | < 3 seconds | Time to interactive (CloudFront edge) |
| Email delivery | < 60 seconds | Time from API call to SES acceptance |
| Scheduled jobs | Within 5 minutes of schedule | EventBridge trigger to Lambda execution |
Performance targets are best-effort goals, not SLA commitments with remedies.
| Priority | Description | First Response | Resolution Target |
|---|---|---|---|
| Critical (P1) | Platform completely unavailable | 1 hour | 4 hours |
| High (P2) | Major feature unavailable, no workaround | 4 hours | 24 hours |
| Medium (P3) | Feature degraded, workaround available | 8 business hours | 72 hours |
| Low (P4) | Minor issue, cosmetic, feature request | 24 business hours | Best effort |
Business hours: Monday-Friday, 9:00 AM - 6:00 PM Eastern Time (US), excluding federal holidays.
Critical and High priority issues are monitored 24/7.
If monthly uptime falls below the 99.9% commitment, affected customers are eligible for service credits:
| Monthly Uptime | Service Credit (% of monthly fee) |
|---|---|
| 99.0% - 99.9% | 10% |
| 95.0% - 99.0% | 25% |
| 90.0% - 95.0% | 50% |
| Below 90.0% | 100% |
Service credits do not apply when downtime results from:
| Event | Notification |
|---|---|
| Unplanned outage detected | Within 30 minutes via status page |
| Outage ongoing > 1 hour | Email to affected customers |
| Outage resolved | Resolution notice within 1 hour of recovery |
| Post-incident report | Within 5 business days of resolution |
| Scheduled maintenance | 72 hours advance notice via email |
| Component | Durability | Backup Method | RPO |
|---|---|---|---|
| DynamoDB tables | 99.999999999% (11 nines) | Point-in-Time Recovery (PITR) | Near-zero (continuous) |
| S3 objects | 99.999999999% (11 nines) | Versioning enabled | Near-zero (versioned) |
| Application code | N/A | Git version control | Zero (immutable deploys) |
Recovery Time Objective (RTO): 15-30 minutes for full service restoration from backup.
| Feature | Professional | Agency | Enterprise |
|---|---|---|---|
| Uptime SLA | 99.9% | 99.9% | 99.9% |
| Support hours | Business hours | Business hours | 24/7 |
| First response (P1) | 1 hour | 1 hour | 30 minutes |
| Dedicated account manager | No | No | Yes |
| Custom SLA terms | No | No | Negotiable |
| Status page access | Yes | Yes | Yes |
AI Bizlutions monitors service availability using:
Monthly availability reports are available upon customer request.
This SLA may be updated with 30 days advance notice. Changes that reduce service commitments require 90 days notice. Customers who do not accept modified terms may terminate without penalty within the notice period.
This SLA is incorporated by reference into the Terms of Service. In the event of conflict between this SLA and the Terms of Service, the Terms of Service shall govern except for uptime percentages and credit amounts specified herein.
This section documents the formal Business Continuity and Disaster Recovery (BC/DR) plan for the AI Bizlutions platform, satisfying SOC 2 A1.2, ISO 22301, and ISO/IEC 27001 Annex A.5.29/A.5.30 requirements.
| Metric | Target | Mechanism |
|---|---|---|
| Recovery Point Objective (RPO) | Near-zero (< 1 minute) | DynamoDB PITR (continuous), S3 versioning |
| Recovery Time Objective (RTO) | ≤ 30 minutes | Infrastructure-as-Code redeploy (Serverless Framework) |
| Maximum Tolerable Downtime (MTD) | 4 hours | Aligned with P1 incident resolution target |
| Component | Resilience Strategy | Failover |
|---|---|---|
| Compute (AWS Lambda) | Stateless, multi-AZ by default | Automatic (AWS-managed) |
| Database (DynamoDB) | Multi-AZ replication, 11-nine durability | Automatic (AWS-managed) |
| Object Storage (S3) | Multi-AZ, versioning, 11-nine durability | Automatic (AWS-managed) |
| API Gateway | Multi-AZ, edge-optimized | Automatic (AWS-managed) |
| CDN (CloudFront) | Global edge network, origin failover | Automatic (AWS-managed) |
| DNS (Route 53) | 100% availability SLA, global anycast | Automatic (AWS-managed) |
| Email (SES) | Multi-AZ within region | Regional — secondary region via manual config |
| Scenario | Impact | Recovery Procedure | Expected RTO |
|---|---|---|---|
| Single Lambda failure | Partial service degradation | Automatic retry + redeploy from main branch | < 5 minutes |
| DynamoDB table corruption | Data loss for affected table | PITR restore to pre-corruption timestamp | 15-30 minutes |
| S3 bucket object deletion | Media/document loss | Restore from S3 versioning (object-level recovery) | 5-15 minutes |
| Full region outage (us-east-2) | Complete service unavailable | Cross-region redeploy: (1) Deploy stack to us-east-1 via Serverless Framework, (2) Restore DynamoDB from on-demand backups, (3) Update DNS via Route 53 | 2-4 hours |
| Compromised deployment | Security breach via malicious code | Git revert to last-known-good commit, force redeploy, rotate secrets | 15-30 minutes |
| Third-party API outage (Gemini, Replicate) | AI features degraded | Model fan-out with fallback candidates, graceful feature degradation | Automatic (0 minutes) |
| Test Type | Frequency | Scope |
|---|---|---|
| DynamoDB PITR restore validation | Quarterly | Restore single table to verify data integrity |
| Full stack redeploy (staging) | Monthly | Deploy from clean state to verify IaC completeness |
| Secret rotation drill | Semi-annually | Rotate all API keys, verify service continuity |
| Cross-region failover tabletop | Annually | Walk through regional failover procedure with team |
| Role | Responsibility |
|---|---|
| Incident Commander | Declares disaster, coordinates recovery, communicates with stakeholders |
| Platform Engineering | Executes recovery procedures, validates service restoration |
| Security Lead | Assesses compromise scope, manages credential rotation |
| Customer Support | Manages customer communication per §6 notification timelines |