Another outage on a platform that thousands of organisations depend on. This time, the problem appeared in the system designed to recover from a disaster.

On Friday 13 November, Workday left some of its customers unable to access their workforce management tools. Clocking in, checking payslips, choosing benefits or managing time off became a problem. The company serves more than 11,500 organisations worldwide. The exact number affected has yet to be confirmed.

The technical explanation is particularly disturbing. Workday was testing its disaster recovery environment when services in that environment unexpectedly connected to production databases. Some data became inconsistent, and the company put the affected systems into maintenance mode while it worked out how to recover them. The service went down rehearsing how to get back up.

The consequences are far from abstract. Brown asked its staff to record their hours elsewhere. Emerson extended the deadline for choosing employee benefits. Ashby, another provider, reported that its connections to certain Workday environments had stopped working. FAU warned that scheduled processes and integrations could fail. The failure crossed the boundaries of the company that suffered it and reached other connected tools.

As of 14 November, there have been no confirmed widespread failures to pay wages or a confirmed number of delayed salary transfers. That distinction matters. What is already documented is serious enough to ask how long an organisation can function when it loses access to the systems that record its daily activity.

Microsoft on the 12th, the FAA on the 13th and now Workday: three in two days, all three around a technical intervention, according to published reports. It proves nothing. It is the pattern described in Too much in too little time.

There is also a shared dependency. We have concentrated essential tasks on platforms that let us work better, faster and on a scale that was once impossible. That concentration also allows a localised error to interrupt the processes of many organisations at once. And integrations extend its effects to providers that are working perfectly well.

For a company's management, the question becomes very concrete. Can we record hours, reconstruct the necessary data and prepare payments if we lose access tomorrow? For how long? Has anyone tested that procedure?

Outsourcing a function means retaining the ability to respond when the provider fails. Workers will still need to be paid. Deadlines will keep running. Someone will have to decide how to carry on.

Main sources

Opinion article. Sources accessed on 14.11.2026. The timing does not establish a causal link between the outages. The customer notices reviewed remained open; full service recovery has not been confirmed.

Text by Roberto Ortega, prepared with AI editorial assistance. The Norwegian and English editions are editorial adaptations of the Spanish original.