07480 257117 Talk to us
For business ownersBy the numbers

The UK searched for the Microsoft outage at 8am on Friday

Microsoft said Outlook was failing on Thursday evening. UK searches for it were still near zero. They peaked at 8am on Friday, when the working day started, and few people asked what to use instead.

08:00 set large on a black background beside a column chart of hourly UK searches for the Microsoft outage on Friday 13 November, rising from 19 at 05:00 to a peak of 100 at 08:00 and falling to 26 by 13:00, above the words: When UK searches for the Microsoft outage peaked: Friday morning
08:00Friday 13 November, when UK searches for the outage peaked
4 timeshow much higher the UK search peak was than in the CrowdStrike week of July 2024
1peak reading for “microsoft 365 alternative”, on a scale where “microsoft 365 outage” reached 100
about a dayhow long interest took to fall back to near its normal level

On Thursday evening, with Outlook failing across the world, UK searches for “microsoft outage” sat at about three in a hundred of their later peak. At 8am on Friday they hit 100.

Microsoft acknowledged the problem at 17:54 UTC on Thursday 12 November. We wrote up what stopped and what kept working. This post asks a different question: what did people type into Google, and when? The figures are from Sunday 15 November, so they cover the outage and its first day of aftermath, nothing longer.

Thursday evening: the world looked, the UK was at home

Worldwide, searches jumped in the hour after Microsoft’s first post and stayed high through the night. In the UK the same hours were quiet. Thursday evening was no time to be checking email, and most people were not at work.

Searches for “microsoft outage”, hour by hour

Each line is scaled to its own peak (100). UK time and UTC are the same in November.

  • Worldwide
  • United Kingdom
05010018:0000:0006:0012:0018:0000:0006:0017:54 Thursday: Microsoft’s first post

Source: Google Trends via the DataForSEO API, hourly, Thursday 12 November 17:00 to Saturday 14 November 06:00, pulled 15 November 2026

The worldwide line is our check on the clock. It jumps at 18:00, the hour after Microsoft’s post, so the hours in this data line up with the real event.

Friday 8am: the working day starts

UK interest began to climb at 05:00 and hit 100 at 08:00. Hourly readings ran 19, 41, 75 and 100 from 05:00 to 08:00, then 89, 73, 66, 44 and 26 to 13:00. By 17:00 it was 7. In the same hour, on the same scale, “outlook down” reached 66 and “is microsoft down” reached 30.

Our reading, which the data does not prove: many UK businesses found out when they opened a laptop. Thursday’s failure had been sitting there overnight. The first person to try sending an email on Friday became the first person to search.

For a small business this is the awkward part. Your customers and staff may learn about an outage at the same moment you do. That is the start of the day, with nobody yet in charge of working out what is wrong.

The biggest week since at least January 2024

We also looked back to January 2024 for the same phrase in the UK. The week of the outage scored 100. The week of the CrowdStrike fault in July 2024 scored 24. Every other week scored 4 or less.

UK searches for “microsoft outage”, busiest weeks

Weekly index, January 2024 to 15 November 2026. 100 is the busiest week in that span. Labels show the day each week starts.

1008 Nov 20262414 Jul 2024426 Apr 202646 Jul 2025326 Oct 2025

Source: Google Trends via the DataForSEO API, weekly, United Kingdom, pulled 15 November 2026

Teams barely registered. “teams down” peaked at 6 in the same request where “outlook down” reached 56, about a tenth as much. That fits what we reported before: Teams chat mostly kept working, and email did not.

What people asked next

The related searches read like a story told in questions. First came “is microsoft down” and “outlook down”. Then “is outlook still down” and “when will outlook be fixed”. Then “what caused microsoft outage”. By Friday there were searches for a status page, and one for “ofsted email issues microsoft outage”, which suggests public bodies felt it too.

A second group never mentioned Microsoft at all. On Friday “email not working” rose from about 12 to 100 and “email down” from about 4 to 74. These people knew something was wrong with their email and did not know whose fault it was.

Searches about money appeared on the same day. “microsoft sla” and “microsoft compensation” both peaked on Friday, having barely registered in the three weeks before. We have not read what service credits Microsoft’s terms give, so we have no view on what anyone is owed.

What people did not search for was a way out. On the same scale, “microsoft 365 outage” reached 100 on Friday. “microsoft 365 alternative” reached 1.

Friday 13 November: asking what happened, not what next

UK daily index from one request, so the columns compare with each other. The usual level for 1 to 11 November was 4 for “email down”, 9 for “google workspace”, 1 for “backup email” and 0 for the other two.

100365 outage74email down11google workspace2backup email1365 alternative

Source: Google Trends via the DataForSEO API, daily, United Kingdom, pulled 15 November 2026. Full terms: “microsoft 365 outage”, “email down”, “google workspace”, “backup email”, “microsoft 365 alternative”

Two small signs are worth a mention. Searches around “google workspace” included “is google workspace down” and “protonmail”, both rising. Some people may have been checking whether other providers had failed too. The volumes were small, so we would not build much on it.

By Saturday, daily searches were back to between 0 and 5 on the same scale. That was the day Microsoft said service had returned for the last customers. Interest fell back within about a day.

What a small business can take from it

None of this says what you should buy. It does say something about timing and habit.

What this data cannot tell us

Google Trends gives an index, not a count of searches. Low values are rounded, so a small term can show as nothing. It covers Google only, and only up to the day we pulled it. We could not reach forums or comment threads, so we cannot say what people said to each other, only what they typed. Microsoft had not published its full incident report when we wrote this, so the cause is not settled either.

Last, our readings of why searches moved are readings. The numbers are what the tool returned. If you can see a better explanation, we would like to hear it.

Where the numbers come from

  1. Google Trends interest over time and related queries for the United Kingdom and worldwide, pulled on 15 November 2026 through the DataForSEO Google Trends API. Hourly figures for 12 to 14 November, daily figures from 1 October, weekly figures from January 2024. Values are an index from 0 to 100 relative to the peak inside each request, not counts of searches.
  2. Microsoft's first public acknowledgement at 17:54 UTC on Thursday 12 November, and the reports from 74 countries and territories: Associated Press and Wikipedia, as listed in our earlier post on the outage.
  3. Microsoft Learn, How to check Microsoft 365 service health, read on 14 November 2026: the public status page for when you cannot sign in to the admin centre.
  4. The week of the CrowdStrike fault is our reading of the date, 19 July 2024, inside the weekly figures. We have not matched the three other busy weeks to news reports.

The zxn.uk team. Written together.