Incident Post-Mortem — Transactional Email Delivery Delay
Date: 16 June 2026
Duration: Approximately 40 minutes
Status: Resolved
Issue Summary
On 16 June 2026, our transactional email delivery service (covering both API and SMTP relay sending) experienced a delay of approximately 40 minutes. During this window, transactional emails submitted to the platform were accepted and safely queued, but their delivery to recipients was temporarily held while the issue was being addressed.
No emails were lost. All affected messages were successfully delivered once the service was restored.
Customer Impact
- Affected service: Outbound transactional email delivery (API and SMTP relay).
- What customers experienced: Transactional emails were accepted by the platform but delivery to recipients was delayed for the duration of the incident.
- Scale of impact: At its peak, up to 100% of transactional email traffic was affected — approximately 4 million emails were delayed by around 40 minutes.
- Email loss: None — all messages were queued safely and delivered once the service was restored.
- Services not impacted: Marketing email campaigns, SMS, WhatsApp, contact management, dashboard access, automations, and all other platform features remained fully operational.
Timeline (all times in UTC)
- 09:00 UTC — Incident began. Transactional email delivery started to slow down.
- 09:04 UTC — Issue detected by our internal monitoring systems.
- 09:26 UTC — Engineering team began applying mitigation steps.
- 09:43 UTC — Service fully restored. Queued emails began delivering.
- 09:52 UTC — Full recovery confirmed. All metrics returned to normal baseline.
Root Cause
The incident was caused by a temporary configuration issue within our internal email-processing infrastructure that occurred during a routine maintenance operation earlier in the day. This issue led to a buildup of queued transactional emails, which eventually impacted the system component responsible for moving emails through the sending pipeline. Once that component reached its operational limit, transactional email delivery was temporarily paused while the underlying cause was being resolved.
Resolution
Our engineering team carried out the following steps to restore service:
1. Identified the impacted system component and increased its operational capacity, which immediately unblocked the sending pipeline.
2. Restored the underlying configuration that had been affected during the earlier maintenance operation.
3. Closely monitored email delivery throughput, queue depth, and error rates until the entire backlog had drained and all metrics held steady at their normal baseline.
Preventive Measures
To prevent a similar incident from happening in the future, we are implementing the following improvements:
1. Stronger guardrails on routine maintenance operations to ensure that production configurations cannot be inadvertently affected during unrelated infrastructure work.
2. Improved traceability on automated maintenance jobs so any change can be immediately identified and rolled back if needed.
3. Capacity and headroom review of the transactional email pipeline to ensure sufficient buffer for unexpected traffic spikes, and earlier alerting before any component reaches its operational limit.