A Workday outage on Friday evening, November 13, 2026, locked employees at many organizations out of the HR and payroll platform. Workday’s own notice, as relayed by Washington University in St. Louis IT, says "tenanted services within the DR Site unexpectedly connected to Production databases." In plain words, part of Workday’s backup site reached into live customer data during a test.
People could not sign in, clock in or out, see pay information, request time off or finish benefits enrollment, according to Engadget. Workday says on its home page that more than 11,500 organizations use it. It has not said how many were hit.
Status as of 11:50 p.m. ET, Sunday, November 15: service is back for most customers we could check. According to ABC10, a Workday spokesperson said service was restored by 11 a.m. EST Saturday. WashU IT marked its incident resolved at 7:49 p.m. ET Saturday. Workday’s own status page needs a customer login, so we could not read it.
The short version: this was not a cyberattack, per Workday. It was a disaster recovery test that touched production. If your organization uses Workday, check time punches and timesheets for November 13 and 14, check payroll inputs before the next run, confirm benefits deadlines, tell staff what to expect, and watch for outage-themed phishing. Then ask Workday how it keeps recovery tests away from live data.
What happened in the Workday outage
The first public signs came late Friday afternoon, Eastern time. Engadget reports that Downdetector reports spiked around 5:19 p.m. ET. UC Today reports that Emerson College said Workday took the system offline at about 22:15 GMT, which is 5:15 p.m. ET.
The cause comes from a notice Workday posted on its Community status page. Customers relayed it. WashU IT’s 8:05 p.m. ET update quotes it. Workday said that while validating weekly maintenance for its disaster recovery tenants, services at the DR site "unexpectedly connected to Production databases." It added that "this has left data in an inconsistent state that can cause errors accessing tenants." Workday then said it was "placing all WD1 Production tenants into Maintenance."
WD1 is the name customer IT pages use for one of Workday’s production data centers. Accounts of the scope differ. WashU’s relay names all WD1 production tenants. ABC10 reports that a Workday spokesperson said the disruption affected two of the company’s data centers. We could not reconcile the two from Workday’s own pages.
Workday’s public statements about the Workday outage were short. Per Just the News, Workday said: "We sincerely apologize for the challenges this is causing impacted customers." ABC10 reports Workday said the outage was not related to a cybersecurity incident. The spokesperson told ABC10: "we sincerely apologize for this outage."
Timeline of the Workday outage in ET
Times below come from WashU IT’s status page unless noted. WashU posts in Central time; we converted to Eastern.
| Time (ET) | What happened | Source |
|---|---|---|
| Fri 5:15 p.m. | Workday takes the system offline, per Emerson College | UC Today |
| Fri 5:19 p.m. | Downdetector reports spike | Engadget |
| Fri 8:05 p.m. | WashU relays Workday’s notice: DR site connected to production; WD1 production tenants put into maintenance | WashU IT |
| Fri 9:17 p.m. | “The vendor has taken the Workday service offline temporarily” | WashU IT |
| Fri 11:44 p.m. | Workday not expected back until after the overnight maintenance window | WashU IT |
| Sat 9:07 a.m. | Workday “still performing data validation activities” | WashU IT |
| Sat 11:00 a.m. | Service restored, per a Workday spokesperson | ABC10 |
| Sat 12:11 p.m. | Northeastern notes “limited recovery” | UC Today |
| Sat 12:24 p.m. | WashU: outage “has been resolved,” but updates between systems may be delayed | WashU IT |
| Sat 7:15 p.m. | “Workday’s core vendor systems have been restored” | WashU IT |
| Sat 7:49 p.m. | WashU marks the incident resolved | WashU IT |
Recovery from the Workday outage was not the same for everyone. UC Today notes that some universities showed only limited recovery hours after Workday’s restore time. Each tenant came back on its own schedule.
Why a disaster recovery test can break production
A disaster recovery (DR) site is a standby copy of a system. If the main site fails, the vendor switches to it. To trust it, vendors test it often. The test should never touch live data. Our disaster recovery planning explainer covers the basics.
Workday’s notice describes the one thing a DR test must not do. Services at the standby site connected to the live databases. Two sets of services may then have written to the same data. That is why Workday said the data was in an "inconsistent state." It also explains the long fix. Workday had to stop access and validate the data before letting people back in.
The lesson for IT teams is plain. A DR site needs a hard wall between it and production during tests. That means separate network paths, separate credentials and connection settings that cannot point at live systems by mistake. The same rule applies to your own systems. A test restore of your AMS or CMS database should never be able to write to the live one.
What HR and payroll staff should check after the Workday outage
The Workday outage hit on a Friday evening and lasted into Saturday. That window touched shift work, weekend timekeeping and open enrollment. Here is what to check this week.
-
Time punches for November 13 and 14. Staff who could not clock in or out may have gaps. Collect manual times from managers. Enter or approve corrections before the timesheet deadline.
-
Payroll inputs before the next run. Check that time, overtime and pay changes entered near the outage saved. Look for duplicate or missing entries. If your pay run is close, ask payroll for an extra review.
Two more checks cover requests and deadlines. -
Time-off requests. Requests made during the Workday outage may not have gone through. Ask staff to confirm their requests show up.
-
Benefits enrollment deadlines. Open enrollment season overlaps the Workday outage. UC Today reports that Emerson College extended its open enrollment to 17:00 GMT on Monday, November 16. If your window closed during the outage, decide whether to extend it and tell staff.
Integrations need their own check. -
Integrations and scheduled jobs. Workday’s notice warned of failed jobs and integrations. Check feeds to payroll providers, benefits carriers, single sign-on and your AMS or finance system. Rerun any that failed.
WashU IT warned that "updates between systems and within Workday may be temporarily delayed." Expect some lag in reports for a few days.
How IT should talk to staff and watch for phishing
Send one short note to staff. Say what was down, that Workday says it was not a security incident, and what they need to check. Give a contact for pay or time questions. Keep it plain.
Expect phishing that uses the Workday outage as bait. Common lures are "verify your direct deposit" or "re-enter your benefits choices." Remind staff that HR will not ask for bank details or passwords by email. Tell them to open Workday from a bookmark, not from a link in a message. Our post on what phishing attacks are has more on spotting these.
If you run a help desk, log calls tied to the Workday outage. Note who could not get paid information, who missed a deadline and which integrations failed. You will want that record for the next step.
What to ask Workday, or any SaaS vendor, about DR tests
The Workday outage is a good prompt to ask harder questions of every SaaS vendor that runs payroll, membership or donations. Workday has not published a root cause report yet. When it does, read it with these questions in mind.
| Question | Why it matters | Good answer looks like |
|---|---|---|
| How do you isolate DR tests from production? | This outage started when DR services reached live databases | Separate networks and credentials, with checks that block a live connection |
| Will you publish a root cause report? | You need facts for your own records and auditors | A written report with timeline, cause and fixes |
| Was any data changed or lost? | “Inconsistent state” can mean bad records | A clear yes or no for your tenant, with what was checked |
| Do service credits apply? | Your contract may set an uptime target | A direct answer under your contract terms |
| How will you tell us next time? | The status page needed a login | Email or text alerts to named contacts, plus a public page |
Workday’s own blog post on an earlier outage points customers to a status page on Workday Community, which needs a login. Make sure more than one person on your team has that login. Our post on data backup strategy basics covers keeping your own copies of key data, such as a recent export of pay rates and time records.
The Microsoft 365 outage was a separate event
The Workday outage came one day after a large Microsoft 365 outage that took down Outlook and other services. That was a separate event with a different cause, and nothing links the two. We covered it in the Microsoft 365 outage post and in our follow-up on Microsoft 365 outage recovery. Two big SaaS outages in two days is a fair reason to review your plan for when a cloud service you depend on goes down.
Frequently Asked Questions
Is the Workday outage over?
A Workday spokesperson told ABC10 that service was restored by 11 a.m. EST on Saturday, November 14. Some customers recovered later. Check your own tenant and Workday’s Community status page.
What caused the Workday outage?
Workday’s notice, as relayed by WashU IT, says services at its disaster recovery site connected to production databases during testing. That left data in an inconsistent state.
Was it a cyberattack?
No, according to Workday. A spokesperson told ABC10 the outage was not related to a cybersecurity incident.
Which customers were affected?
Workday has not said. WashU’s relay names all WD1 production tenants. Workday told ABC10 that two data centers were affected.
Was any data lost?
Workday has not said data was lost. Its notice says data was left in an inconsistent state. Check time, pay and benefits records entered near the outage.
What should payroll check first?
Time punches and timesheets for November 13 and 14, then any pay changes entered near the outage, before the next payroll run.
Should we extend open enrollment?
If your window closed during the outage, consider it. Emerson College extended its deadline to Monday, November 16, per UC Today.
Why could we not see Workday’s status page?
Workday’s status page sits on Workday Community and needs a customer login. Make sure more than one person on your team can sign in.
Is this related to the Microsoft 365 outage?
No. The Microsoft 365 outage on November 12 was a separate event with a different cause.