Engineering Atlas
☰ Menu
PortfolioAtlasDeploySQLData ViewsEnvBites
PT
Engineering Atlas / Shit That Bites You in Production / RaiseError Can Skip the Send but Still Show Up in Tracking Metrics
medium impactSTUDYAMPscript

RaiseError Can Skip the Send but Still Show Up in Tracking Metrics

Tracking/report counts include recipients for which RaiseError prevented the email from being sent.

Context Pack →Open Workspace →
Trust & provenanceSTUDY
Evidence tierSTUDYOfficial sources1Other sources0Last verifiedNot recorded

Evidence tier describes my hands-on relationship with the content. Official sources describe platform contracts/documentation. Production observations should not automatically be treated as official limits.

Why it bites

Salesforce documents that messages are built/preprocessed before sending, and tracking/reporting metrics can include emails not sent because RaiseError() was called.

What to check
  1. Was RaiseError used for genuine error handling or as segmentation logic?
  2. Are business metrics treating tracking rows as proof of actual send?
Make it cheaper next time
  1. Use audience queries/exclusion lists for segmentation.
  2. Reserve RaiseError for exceptional conditions and document its reporting effect.
ampscriptraiseerrortrackingsegmentation
Connected knowledge

Keep following the context.

Explicit and inferred relationships from the Atlas model.

View this node in graph →
Snippet
RaiseError() for a Broken Recipient, Not Segmentation

Abort only the current recipient when a genuine rendering invariant is violated.

Shared context: ampscript, raiseerror
Snippet
CASE-Based Business Segmentation

Translate explicit business rules into a readable segment label.

Shared context: segmentation
Snippet
Fast Approximate Split with _CustomObjectKey Modulo

Use the hidden row key for cheap approximate segmentation of a standard Data Extension.

Shared context: segmentation
SQL Guide
Audience Splitting: Precision vs Throughput

Use NTILE() when you need balanced random buckets. Use _CustomObjectKey modulo when you need a very cheap, approximately even split on large standard Data Extensions and exact balance is not required.

Shared context: segmentation
Sources / verification
Salesforce — RaiseError() ↗