This quarter is about evidence: proving that SP Foundation can handle growth, contain bad releases, and be operated independently by a product pod.
Last updated September 3, 2026.
Sustain twice the highest five-minute average production requests per second measured during the previous 90 days for 30 consecutive minutes, while keeping p99 latency at or below three seconds and staying within the agreed availability and error-rate thresholds.
From production go-live through the end of Q3, zero Sev-1 or Sev-2 customer incidents are caused by traffic volume, resource exhaustion, infrastructure capacity, or scaling failure.
At least one product pod independently operates one real production service end to end—deployment, dashboard monitoring, alert response, diagnosis, and service restoration or rollback—without Core Platform taking an operational action.
100% of production services hosted on SP Foundation have startup health checks, release-health monitoring, and automatic deployment blocking or rollback configured for every defined bad-release condition.
At least 95% of qualifying bad releases are automatically blocked or reverted within five minutes, with 100% success across all three failure scenarios during a production-safe game day.
RPS means requests per second. p99 is the response time experienced by the slowest 1% of requests. Sev-1/Sev-2 are the two highest customer-incident severity levels. An SLO is an agreed service-level objective.
Confidence comes from controlled proof, configured protection, and real operating behavior reinforcing one another.
Production baseline, load-test result, latency, availability, and error-rate record.
Service configuration coverage plus detection and restoration timestamps.
Deployment, dashboard, alert-routing, game-day, and restoration records owned by the pod.
The migration work applies the same platform principles—isolated execution, recoverability, observable failure handling, and evidence.
SPPLT-17956 is one foundation slice in the broader individual-customer migration epic. It adds isolated background processing and downloadable failure reporting. The tracking foundation is complete; this next build slice currently depends on dedicated queue and worker infrastructure.
Jira status as of September 3, 2026: Story Creation · Priority: Medium · Unassigned. Jira remains the operational source of truth.
The Q3 platform objective proves SP Foundation. The compliance objective is related, but it has its own scope, evidence, accountability, and sign-off.
Approve, publish, assign, and launch acknowledgements for all required internal policies and guidance; begin the workforce walkthrough and training program; and establish an owner, due date, and evidence requirement for every identified GDPR gap.
100% of in-scope employees and contractors acknowledge all assigned requirements with no overdue items.
100% of the in-scope workforce completes required privacy, security, data-handling, and policy training.
All critical and high-priority gaps are remediated and evidenced; lower-priority items have an owner, date, and risk decision.
Walk through data requests, retention and deletion, privacy incidents, and evidence collection end to end, with every blocking finding resolved.
Legal, executive, or designated privacy leadership confirms that approved requirements in scope are implemented, operating, evidenced, and ready for ongoing monitoring.
Policy acknowledgements are an important milestone, not proof of complete GDPR compliance. Engineering supplies controls and evidence; designated organizational leadership provides legal or compliance sign-off.