Date: February 17–18, 2026 Duration: ~15 hours Severity: Minor
A portion of Pinterest posts scheduled through Buffer failed to publish due to an issue on Pinterest's side. The problem resolved itself as Pinterest restored normal API behaviour. No action was required from customers, and no data was lost.
This was a third-party issue caused by instability in Pinterest's API. Pinterest's servers were returning errors for publishing requests, which caused affected posts to fail. The issue was not related to any change on Buffer's end.
Approximately 100–200 Pinterest posts failed to publish during the incident window. Customers whose posts were affected would have seen an error on their scheduled posts. The issue gradually subsided as Pinterest's API recovered, and by the following morning no further failures were occurring.
Our monitoring systems detected the elevated failure rate and alerted our team. After confirming the issue was on Pinterest's side, we monitored error rates closely as they trended downward. Once no errors had occurred for several hours, the incident was confirmed resolved. There were no actions available to us to speed up the resolution, as the root cause was entirely on Pinterest's infrastructure.
As this was a third-party outage, there was nothing Buffer could have done to prevent it.
Our monitoring detected the issue promptly, allowing us to keep track of its progression and communicate with customers without waiting for reports to come in.
