Troubleshooting

Common errors

The table below outlines common errors that may occur during account availability or ownership evaluation using Socure Account Intelligence (Standard or Premier). These issues can result from missing fields, input formatting problems, or module configuration. Understanding these patterns can help fine-tune workflows, reduce false positives, and determine when escalation or fallback logic is appropriate.

Issue	Explanation	Suggested fix
Missing or invalid `accountNumber` / `routingNumber`	Required bank account fields were not provided or failed basic validation.	Ensure both fields are present, numeric, and properly formatted before submission.
Unexpected `status_code` in enrichment block	A non-200 response indicates enrichment was not executed successfully.	Review the `status_code` and `response` fields. Retry if transient, otherwise open a support ticket.
Account Intelligence Premier not triggered	The `accountintelligencepremier` module was not called.	Confirm the `modules` array in the request included `"accountintelligencepremier"`.
Low `availabilityScore` despite valid account	External data sources may show conflicting signals on account status.	Confirm account details with the user. Escalate to review if the score is inconsistent with expectations.
Low `ownershipScore` with correct inputs	Provided name or entity did not align with bank records.	Validate `firstName`, `surName`, or `entityName`. Check for formatting errors (e.g., casing, spacing, suffix).
Ambiguous business vs. individual ownership	Request did not clearly specify whether account belongs to an individual or a business.	For businesses, include `entityName` and `ein`. For individuals, include `firstName`, `surName`, and `dob`.
Reason codes repeated (e.g., I998, I992–I994)	Multiple fallback signals may be returned.	Implement hierarchy or suppression logic to avoid duplicate handling of redundant reason codes.
Scores close to threshold (e.g., 0.50–0.70)	Probabilistic model generated uncertain signals.	Tune routing logic to treat mid-range scores as `REVIEW` instead of `PASS` or `FAIL`.

How to debug

To debug Account Intelligence Standard or Premier behavior in RiskOS™:

Use the eval_id to trace execution in the Developer Console or RiskOS logs.
Check the status_code in each data_enrichments[] block.
- A 200 indicates success.
- Any other status should be flagged.
Validate which enrichment modules were triggered. If Premier was expected but not called, confirm that required inputs were present:
```
"modules": ["accountintelligencepremier"]
```
Review reasonCodes for enrichment-level explanations (e.g., fallback logic, missing inputs, data mismatches).
Inspect account.availabilityScore and account.ownershipScore:
- High scores (≥ 0.90) indicate strong matches.
- Low scores (≤ 0.50) indicate weak or no correlation.
Confirm that sourceAttribution is present. Missing attribution may indicate insufficient coverage.

Fallback strategies

If Account Intelligence does not yield a confident result:

Retry with normalized inputs → Reformat account holder name, remove special characters, and ensure routingNumber follows ABA format.
Escalate to review → Use tags or review_queues to route uncertain results to human analysts.
Enable additional enrichments → Chain to other RiskOS modules (e.g., Document Verification or Watchlist) for additional confidence.
Accept partial matches in low-risk flows → For example, accept availability confirmation even if ownership is uncertain.

Escalation path

For production-impacting issues or debugging edge cases:

Reproduce the issue — Do this in sandbox or staging if possible.
Capture relevant identifiers — Collect:
- eval_id
- Full request payload
- referenceId from the Account Intelligence enrichment (if applicable)
Open a support ticket — you can email [email protected] for additional assistance or contact your Solutions Consultant team and include:
- Workflow name
- Input fields submitted
- Expected vs. actual decision
- Logs or screenshots (if applicable)
Urgent escalation — For high-severity cases, escalate via:
- Slack (if integrated)
- Your designated Socure escalation channel

Known issues

United States-only — Account Intelligence Standard and Premier are limited to U.S. accounts. They cannot be used in global flows.
Probabilistic scoring — Instead of binary yes/no, the system returns availabilityScore and ownershipScore on a 0–1 scale.
Reason code overload — Some transactions may return multiple fallback-related codes. Use a hierarchy or suppression logic.
Coverage gaps — Thin-file or recently opened accounts may return low confidence due to limited data source availability.
Special character normalization — Names or entity strings with diacritics/unicode may cause mismatches if not normalized.

📘
Note:
Use the Developer Console to view raw request/response logs in real time for debugging or validation.