Automated rejection

An automated rejection occurs when a configured rule or automated decision flow rejects a candidate without a new manual action for each case.

Also called: automated decline

Separate the decision from the message

An automated rejection can follow a defined rule. A knockout question, for example, can process an answer against a pre-set criterion. Another configuration can send a rejection after a defined period.

This differs from automatically sending an email after a person has made the decision. The dispatch automates communication. It does not decide who should be rejected.

Define the rule before using it

The rule needs to be intelligible for the particular job. A team should be able to identify the trigger, check that the criterion belongs to the role, and assign responsibility for changes. A vague impression from a CV or an unexplained score is not a defined rule.

Avoid general assumptions about reversibility. Whether a decision can be changed, and what follows, depends on the workflow and software configuration.

The data-protection boundary

Under Article 22 of the General Data Protection Regulation (GDPR), people generally have the right not to be subject to solely automated decisions with legal or similarly significant effects. The exceptions and safeguards are set out in the statutory text reproduced by CNIL. Whether a recruiting workflow meets that boundary depends on the case. See GDPR in recruiting for the broader framework.

What to examine

Keep the trigger, timing, template, and later correction visible to the responsible team. An email delay, undo control, or general consent checkbox does not on its own establish compliance.

Sources and references

Talk to Join