Start with the platform that failed
A failed Instagram post, TikTok video, Slack message, and Reddit submission can all look like a generic publishing error in a product UI, but the cause is usually platform-specific. Instagram failures often happen around media containers, TikTok can block privacy or spam-risk combinations, Slack can return ok false in the response body, and Reddit can reject posts because of community rules.
Map the error to the right action
Most failed posts fall into one of four actions: reconnect the account, fix the input, retry later, or wait for the platform. Token, scope, revoked-access, and permission failures belong in the reconnect path. Caption limits, unsupported media, missing fields, duplicate content, and invalid URLs need a payload or content change. Rate limits, processing delays, timeouts, and 5xx responses need a delayed retry.
Keep reconnects rare and specific
A reconnect is useful when the platform no longer trusts the token, the user changed permissions, a required scope is missing, or the account can no longer access the target page, channel, profile, or location. It will not fix an expired media URL, a rejected video format, a policy block, a duplicate post, a community rule, or a temporary platform outage.
Retry only when the same request can succeed later
Retry temporary platform failures, processing delays, timeouts, 5xx responses, and rate-limit errors after the platform allows it. Do not immediately retry validation, permissions, unsupported media, account restriction, or policy failures. The same payload usually needs to change before it can publish successfully.
Use the platform pages for exact checks
Choose a platform below to search native error messages, normalized Bundle error keys, likely causes, and recommended retry behavior. Each page ties the failure to the account, publish stage, media requirement, permission state, and next user-facing action.