Back

Workday Outage, November 13 2026: How a Disaster Recovery Check Locked Employees Out of Pay and Clock In

For most of a Friday night and Saturday morning, Workday users couldn’t clock in, request time off or finish benefits enrollment. Workday says its disaster recovery site connected to production databases by mistake. Here’s what’s confirmed, where the reports split, and the lesson for anyone with a backup copy of their own app.

Jagdish Patil

•

Shortly after 5 p.m. ET on Friday, November 13, nearly 3,000 people had told Downdetector the same thing: Workday wouldn’t let them in. Employees couldn’t clock in or out, request time off or open a payslip, according to LiveNOW from FOX. Emerson College ended up pushing its open enrollment deadline to Monday because staff simply couldn’t reach the benefits forms.

Workday’s explanation came in a notice to customers, which Washington University’s IT team passed along and UC Today reported. While Workday was “validating the weekly Maintenance Window for Disaster Recovery tenants,” services in its DR site “unexpectedly connected to Production databases.” That left data “in an inconsistent state.”

A Workday spokesperson says service was back by 11 a.m. ET on Saturday. This outage breakdown sticks to what Workday and its customers have confirmed, and calls out every spot where the reporting doesn’t line up.

Incident Overview (TL;DR)

  • What failed: Workday access for some customers. That blocked clocking in and out, time off requests, payslips, training and open benefits enrollment. Workday’s notice also mentions jobs, integrations and tasks failing.

  • When it started: Friday, November 13, 2026. Emerson College says Workday took its system offline at about 5:15 p.m. ET (22:15 UTC). One roundup places the first reports earlier in the afternoon (more on that below).

  • Where: Mostly US customers, including a long list of universities. UC Today also saw StatusGator reports from India. Workday hasn’t said how many customers or data center pools were involved.

  • Peak reports: Nearly 3,000 Downdetector reports shortly after 5 p.m. ET (22:00 UTC), per LiveNOW from FOX. We haven’t found a later peak figure.

  • Confirmed cause: Disaster recovery tenants unexpectedly connected to production databases during a maintenance window validation, per Workday’s customer notice. Workday told ABC10 it was not a cybersecurity incident.

  • Status: Resolved. Workday says service was restored by 11 a.m. ET Saturday, November 14 (16:00 UTC). We haven’t found a published incident report from Workday.

What Happened

Workday runs HR, payroll, timekeeping and finance software for large employers and a lot of universities. Its own investor materials put the customer count around 11,500, as Sesame Disk notes.

On Friday afternoon, people trying to log in got a message saying Workday was unavailable because of a service interruption. Workday told LiveNOW from FOX that “some of our customers are experiencing a service disruption” and apologized. According to UC Today, Workday then placed its WD1 production tenants into maintenance and posted updates roughly every 30 minutes.

What Workday’s own notice described: errors accessing tenants, and jobs, integrations and tasks failing with “missing transaction” error messages.

What users and employers reported, which Workday never itemized itself: failed clock in and clock out, blocked time off requests, payslips that wouldn’t load, training modules that wouldn’t open, and benefits enrollment stuck mid form.

Recovery was uneven. Emerson College marked the incident resolved at 10:27 a.m. ET Saturday. Washington University called it resolved at 12:24 p.m. ET but warned that updates between systems might lag for a while. The University of Chicago didn’t report normal operation until the afternoon.

Timeline

All times are ET and approximate. Each one comes from a named outlet or a customer status page as reported.

  • Fri Nov 13, about 3:24 p.m.: The outage is first flagged, according to Applescoop’s November roundup.

  • Shortly after 5 p.m.: Downdetector shows nearly 3,000 reports (LiveNOW from FOX).

  • About 5:15 p.m.: Workday takes Emerson College’s system offline, per Emerson’s IT status page as cited by FourWeekMBA.

  • Friday evening: Workday’s notice blames the DR site connecting to production. WD1 tenants go into maintenance (UC Today).

  • Overnight: Washington University tells users Workday isn’t expected back until after an overnight maintenance window (UC Today).

  • Sat Nov 14, 10:27 a.m.: Emerson College marks the incident resolved.

  • 11 a.m.: A Workday spokesperson says service was restored by this time (ABC10, via UC Today; AP reported the same time).

  • 12:11 p.m.: Northeastern reports “limited recovery” (UC Today).

  • 12:24 p.m.: Washington University marks it resolved, with a warning about delayed syncs (UC Today).

  • 2:35 p.m.: The University of Chicago reports Workday “operating normally” (UC Today).

  • Mon Nov 16, noon: Emerson’s extended open enrollment deadline.

Where the Sources Disagree

When it started. Downdetector and Emerson put the visible start right around 5 p.m. ET. Applescoop says it was first flagged at 3:24 p.m. ET. Reddit posts quoted by UC Today go further and claim the DR site first touched production at 8:50 a.m. PT (11:50 a.m. ET), with the first customer alert at 2:12 p.m. PT (5:12 p.m. ET). UC Today says it couldn’t verify those Reddit times, and neither can we.

How many were hit. The LiveNOW from FOX headline says employees “across some 11,500 businesses” were affected. Workday only said “some of our customers.” UC Today, Sesame Disk and WindowsForum all point out that 11,500 is Workday’s total customer base, not an impact count. The same goes for big names like Walmart and Target: FOX listed them from Workday’s customer page, and nobody has confirmed they were affected.

Which pool. The notice names WD1 tenants. Sesame Disk, citing a Georgia Tech update relayed through StatusStack, quotes a Workday update about remediation for “approximately 96% of WD5 customers.” Workday hasn’t cleared up which pools were affected.

When it ended. Workday says 11 a.m. ET. Emerson says 10:27 a.m. Other universities kept reporting problems into the afternoon. Both UC Today and Sesame Disk noticed Workday’s own community page still showed the maintenance notice on Saturday morning after some customer pages had already marked the incident resolved.

What to call the trigger. Press headlines say “DR test.” The SF Chronicle, quoted by FOX, said a problem was found “while testing backup systems.” Workday’s notice says it was “validating the weekly Maintenance Window for Disaster Recovery tenants.” Those aren’t quite the same thing, and Workday hasn’t said more.

Who Got Hit, and How

Employees. The most visible damage. People couldn’t clock in, request time off, see pay details or finish training. Workday hasn’t said payroll calculations were wrong, and WindowsForum makes a fair point: a portal that won’t load doesn’t prove pay processing failed.

Employers and HR teams. Deadlines collided with the outage. Emerson extended open enrollment to Monday at noon ET. Any HR team with a Friday cutoff had to make a call without knowing when Workday would be back.

Universities. They were the most public about it, because their IT departments run public status pages. Washington University, the University of Chicago, Northeastern, Saint Joseph’s, Brown, Emerson and Georgia Tech all posted updates.

Developers and downstream apps. This is the quiet part. Workday’s notice explicitly says jobs, integrations and tasks could fail with missing transaction errors. Anything that pulls from Workday on a schedule, like an outside payroll processor or an account provisioning script, could have failed or synced stale data during the window.

What wasn’t reported. No source links this outage to the Microsoft 365 failure that started the day before or to the FAA ground stop the same afternoon. WindowsForum looked for a shared cause and found none. Also keep the Downdetector caveat in mind: it counts complaints, not affected customers, and most Workday users never file a report.

Root Cause: What Workday Has Confirmed

Here’s everything Workday has said on the record, as reported.

  • Customer status notice (quoted by Washington University IT, reported by UC Today): Workday was “validating the weekly Maintenance Window for Disaster Recovery tenants” when “tenanted services within the DR Site unexpectedly connected to Production databases.” That “left data in an inconsistent state that can cause errors accessing tenants or jobs/integrations/tasks to fail with missing transaction error messages.”

  • Spokesperson to ABC10 (via UC Today and FourWeekMBA): two data centers were affected, it was not a cybersecurity incident, and service was restored by 11 a.m. EST Saturday.

  • Statement to AP (via FourWeekMBA): “We recognize the critical services we provide to our customers and their workforces,” and “we sincerely apologize for this outage.”

What Workday hasn’t said yet: why a DR environment could reach production databases at all, how many tenants or pools were hit, whether any data written during the window needs correcting, and what customers get back under their service agreements.

Our Read of the Symptoms (Interpretation)

This section is our interpretation, not anything Workday has confirmed.

The first question with any big outage is whether something shared broke underneath it. We checked the obvious candidates against the evidence. Microsoft 365 was still recovering from its own incident, but Workday’s notice points at its own databases, not at sign in, and nobody official connects the two. AWS had trouble in its US East region, but that was Sunday, November 15, two days after Workday went down. Nothing about the timing or the stated cause fits an outside dependency.

The internal explanation fits much better. When a backup copy of a system starts writing into the live database, you end up with two sources of truth. Engineers call it split brain. The fix isn’t a restart. Someone has to work out which writes landed where, reconcile them, and only then let users back in. That lines up with Workday putting WD1 tenants into maintenance instead of rolling something back, and with recovery taking most of a day.

One more data point worth knowing. FourWeekMBA found a copy of Workday’s standard service agreement in a city council packet that promises 99.9% monthly availability and a 12 hour recovery time objective for disaster recovery. Customer contracts can differ, but by FourWeekMBA’s math this outage ran well past both.

What This Means for Small Teams and Solo Founders

You probably don’t run Workday. But the safety net causing the outage is the part that applies to you. Most freelancers and solo founders have a staging copy, a backup server or a restore script somewhere, and those are rarely held to the same rules as production.

A few things you can do this week:

  • Make sure your staging or backup copy can’t write to production. Give it separate database credentials and block it at the network level. If your staging config can point at the live database, one wrong environment variable recreates this exact incident.

  • Test restores in isolation. Restore into a fresh database on a separate host, check the row counts, then throw it away. A backup you’ve never restored is a guess.

  • Know where your vendors post updates. Workday’s community status page needs a login, which is why so much of the public detail in this outage came from universities’ own IT status pages. Bookmark the real status page for every tool your business runs on and subscribe to it.

  • Have a manual fallback for anything with a deadline. If a client runs payroll, invoicing or enrollment through a vendor, know the cutoff and the paper route before Friday afternoon turns into Saturday.

  • Watch your own integrations. If you sync data out of a vendor on a schedule, check that the sync actually ran and produced sane output, rather than assuming the vendor would tell you.

For a different flavor of the same lesson, our Microsoft 365 outage breakdown from the day before shows how one vendor change rippled into everything built on top of it.

How to Protect Your Site: Alarms, Not Notifications

Workday went down on a Friday evening and stayed down through the night. That’s the exact window where a push notification fails you. It arrives, your phone is on silent, and you read it over breakfast while your clients already know. A notification you sleep through isn’t an alert.

We built SIOPS around that problem. When a check fails, SIOPS rings a Critical Alarm on your iPhone or Android that cuts through silent mode and Do Not Disturb, using iPhone Critical Alerts and the Android alarm channel. It fires the second a check fails, not after a retry timer. You can also get a voice call, Slack, email, push or a webhook, so the alarm is the loud layer and the rest are backups.

For a failure like this one, monitor the URLs your users actually hit: your login page, your main app route and any health endpoint your integrations depend on. Each monitored site gets a public status page on a siops.app subdomain, so clients have somewhere to look instead of your inbox. The free Hobby plan covers 5 monitors with one alarm and 5 minute checks. The flat priced Serious plan moves you to 60 second checks with unlimited alarms, and Pro goes to 30 seconds or faster with escalation.

One honest limit. An external check sees what a visitor sees. It would have caught Workday’s login page failing on the very next check, but it can’t see a backup copy quietly writing into your production database. For that you need database level checks and logs on your own side.

Workday Outage FAQ

Was the Workday outage a cyberattack?

No. A Workday spokesperson told ABC10 it was not a cybersecurity incident.

How long was Workday down?

From about 5:15 p.m. ET Friday, November 13, until 11 a.m. ET Saturday, November 14, by Workday’s account. That’s roughly 18 hours, though some customers recovered earlier and others later.

What caused the Workday outage?

Workday says tenanted services in its disaster recovery site unexpectedly connected to production databases while it was validating a weekly maintenance window, which left data in an inconsistent state.

Did the outage affect my paycheck?

Workday hasn’t said pay runs were affected. If anything on your payslip looks off for that period, check with HR, since some integrations failed during the window.

Is Workday down right now?

Check Workday’s community status page (it needs a login) or your employer’s IT status page. Customers reported normal service by Saturday afternoon, November 14.

Was it connected to the Microsoft 365 or FAA outages?

No public evidence links them. Workday’s stated cause is internal to its own systems.

We’ll update this outage breakdown when Workday publishes its incident report.

Download SIOPS before your site is the next headline.

SIOPS rings a real alarm that cuts through Do Not Disturb, so you fix it before your client notices.

Similar Stories

Similar Stories