Every Ninety Days Our Service Desk Had the Same Bad Week
For about a year my service desk had a bad week roughly once a quarter. Calls doubled on the Monday, the contact centre started the early shift short of agents, and by Thursday everything had calmed down again. Each review blamed something different and named nothing in particular. It was a team leader in the contact centre who saw the pattern, because she kept a diary of the mornings she started…
For about a year, the company's service desk experienced a bad week roughly once every ninety days. During these weeks, calls to the service desk doubled on Mondays, the contact centre started with a shortage of agents, and by Thursday, the situation had improved. Various reviews attempted to identify the cause, but no consistent explanation was found. A team leader in the contact centre noticed the pattern, as her short-handed mornings coincided with the bad weeks, occurring roughly thirteen weeks apart.
The root cause was discovered to be a project completed and celebrated in the spring of 2024. The company had moved all accounts into a new directory over the course of one weekend. On the following Monday, all users were required to set new passwords. The company's password policy expired passwords after ninety days, so setting, expiring, and resetting passwords on the same day created a ripple effect.
This issue affected both the service desk and the contact centre. When agents changed their passwords on the desktop, the softphone and customer system retained the old password, causing accounts to lock within minutes. Each lockout was genuine and properly handled. However, nobody realized that a single weekend had initiated a chain reaction across every account in the company.
To address the issue, the company discontinued password expiration for accounts with a second factor, which is now the recommended practice. Passwords are changed only when there is a legitimate reason to believe they have been compromised. Self-service password resets can now be initiated directly from the sign-in screen for anyone enrolled in the system.
Following a large migration, the service desk now asks what applications were synchronized during the process, in addition to identifying what changes were made. This proactive approach helps prevent similar issues from arising in the future. The quarterly bad week has ceased, and reset calls have decreased by approximately two-thirds.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.