email home / Playbook
Email operations / practical steps
Diagnose delivery before sending more
A practical checklist for permission-based business email, authentication and useful delivery evidence.
STEP 01
Separate the failure types
Distinguish an invalid recipient, a rejection, a temporary deferral and a message delivered to spam. Keep the provider’s response code and timestamp. “Not received” is a symptom, not a diagnosis, and increasing volume can make an unresolved sending problem worse.
STEP 02
Check the actual sending path
List the services that send as your domain, including support, transactional and marketing tools. Review their official authentication instructions and inspect a test message’s authentication results. Do not replace DNS records blindly or assume one service’s setup covers every sender.
STEP 03
Send to people who expect the message
Use clear subscription expectations and a working opt-out for recurring marketing. Remove invalid recipients and respect unsubscribe requests. Avoid bought lists, deceptive subjects or attempts to work around provider restrictions.
STEP 04
Test changes in a controlled way
Use a small set of authorised test recipients and compare the exact message before and after a change. Record the sending service, From domain, authentication results and delivery response. Follow the receiving provider’s current sender rules rather than old deliverability folklore.
Choose the right next step
- Hard rejection: investigate the response code before retrying the same address.
- Temporary delay: follow the provider’s retry guidance and examine the sending pattern.
- Spam placement: review authentication, expectations and content together; no setting guarantees inbox placement.
Worked example
A team’s CRM messages arrive but its newsletter fails. It checks the newsletter provider’s sending configuration separately, rather than assuming the CRM’s authentication proves everything is set up. It records one controlled test and the resulting headers before changing DNS.