A customer who contacts support twice about the same case is often telling the team that the first answer did not settle the question. The second contact may arrive minutes after the first, while the case is still being handled by another agent. Without a cooldown rule, the case can be reassigned to a new owner who repeats the same checks. A short cooldown keeps the second contact with the first owner long enough to finish the work. The rule should be visible, testable, and easy to override when the customer raises something new.

Name the repeat-contact problem

Repeat contacts are not automatically failures. A customer may send a receipt after the first reply, or ask a related question that belongs in the same case. The problem starts when the new contact triggers a fresh assignment and a fresh promise. The customer then receives two answers that may disagree. The team should treat that split as the event the cooldown is designed to prevent.

Track how often a case receives a second contact inside the same hour. Compare that count with the number of reassignments that follow. A high count with few reassignments suggests the queue is already absorbing the contact. A high count with many reassignments suggests the routing rule needs a cooldown. Record the count by channel so the pattern is visible.

Define the cooldown window

A cooldown is a period in which a new contact on an open case returns to the current owner instead of the general queue. The window should be short enough to match the work, such as thirty or sixty minutes. It should not be long enough to trap a customer who has a genuinely new problem. Write the window in the routing configuration and in the agent guide.

Start the clock at the first outbound reply, not at case creation. The first reply is the point at which the customer has something to respond to. A case that has not been answered yet needs triage, not a cooldown. Record the start event in the case history so the rule can be audited. The clock should use one time zone and one system source.

Separate a repeat contact from a new issue

Define the markers that keep a contact inside the cooldown. A reply on the same thread, a mention of the same order, or a reference to the same refund all count. A contact that names a different order or a different person should start a new case. The agent should have a one-click way to override the cooldown when the content clearly differs.

Write two examples for the team so the boundary is concrete. A customer who forwards a second email about the same delay is a repeat contact. A customer who asks about a second shipment in the same message is a new issue attached to the old thread. The examples should be short enough to remember during a busy shift. Review them when the contact reasons change.

Put the cooldown on the case record

Add a visible field that shows the cooldown status and its end time. The field should be readable by the agent, the queue lead, and the quality reviewer. A hidden rule produces inconsistent behavior because each person guesses. The field should also record the owner who will receive the return contact.

Keep the field out of the customer-facing message. The customer does not need to know the internal routing window. The customer needs a clear answer and a realistic next step. Use the internal field to protect the answer, not to explain the process.

Decide when a reassignment overrides the cooldown

Some cases must leave the original owner. An agent may go offline, lose access, or lack the authority to resolve the issue. The cooldown should yield to an explicit reassignment with a named reason. Record the reason beside the new owner so the next reviewer can see why the rule was broken. A silent override is the failure mode to watch.

Give the queue lead a single override action instead of a set of informal workarounds. The action should require a reason and should notify the new owner. It should also keep the original promise visible so the new owner does not reset it. Review overrides weekly to see whether the cooldown is too rigid.

Brief the queue leads

Run one short session with the leads before the rule goes live. Walk through a repeat contact, a new issue, and an override. Ask each lead to bring one real case from the last week. The discussion will surface the exceptions faster than a policy memo.

Publish the rule in the same place as the routing guide. Include the window, the override path, and the field name. Keep the note to one page and date it. A rule that lives only in a meeting will drift within a week.

Review the cooldown after a month

Compare reassignment counts before and after the change. Check whether repeat contacts now stay with the original owner. Look for cases where the cooldown delayed a legitimate transfer. The review should name one adjustment, not produce a new policy.

Ask agents whether the field helped or added noise. A rule that agents ignore is worse than no rule because it hides the real routing behavior. Update the window or the override path if the evidence supports it. Record the change and the date so the next review has a baseline.

A cooldown is a small routing control, not a fix for every repeat contact. It works when the window is short, the boundary is clear, and the override has an owner. The case record should show the rule in action so the team can learn from it. That visibility turns a quiet routing setting into a managed promise.