I am not particularly superstitious. But I am writing this at four in the morning in an airport hotel, a few hours before I fly to San Francisco for Microsoft Ignite 2026, and Friday the 13th of November has done its best to change my mind.
In the space of about 18 hours, the Microsoft 365 estate fell over, my workshop went in the bin, and the US issued a nationwide ground stop. Here is how the day went, and the one lesson I am taking from it.
I woke up to a Microsoft ecosystem in bits
The trouble had actually started on Thursday teatime, UK time. By the time I woke on Friday, it was a global outage. Outlook, Teams, Copilot, OneDrive and SharePoint were all affected. Downdetector received reports from 74 countries and territories within about an hour. DW coverage · The Guardian live coverage
Microsoft said it had found no evidence of a cyberattack. It traced the outage to an issue introduced during a network capacity upgrade that surfaced after a later network change. Services came back gradually through Friday, with email queues approaching normal levels by the afternoon. Microsoft has promised a full post-incident report, and I will be reading it with interest. Microsoft’s incident updates, reproduced by Keystone · Friday’s email recovery update
The timing is hard to ignore. This was the Friday before Ignite, the week Microsoft tells the world what its platform can do next.
The workshop that went in the bin
I had a workshop planned for Friday. Copilot Studio, Power Platform, and a set of exercises that lean on Teams and Outlook. You can probably see where this is going.
That plan went in the bin early on Friday morning. There is no clever workaround when the two tools your labs depend on are the two tools the whole world is waiting for.
It was not all gloom. I am part of a networking group of charity and non-profit IT leaders, and the group chat was in fine form. Plenty of fun, plenty of banter, and a fair amount of gallows humour from people whose entire organisations had suddenly gone very quiet.
Then the United States grounded its flights
I made my way to the airport on Friday evening, ready for a Saturday morning flight to San Francisco. A bit of dinner in the hotel, a quick check of the news before bed, and there it was. Departures across the US had been halted.
The Federal Aviation Administration issued a nationwide ground stop after communications problems at regional air traffic control centres. It lasted more than four hours and was the first nationwide ground stop since January 2023. FlightAware had already recorded more than 10,000 delays by Friday evening in the US. Associated Press coverage · ABC News flight figures
The stop had been lifted by the time I sat down to write, but delays and cancellations are expected to run through the weekend. When I checked before writing this, my flight was showing on time. I am not counting on anything until the wheels leave the tarmac.
The lesson: your plan B cannot live on the same platform as your plan A
Two unrelated failures, one day, and the same lesson. Both incidents showed how quickly one shared dependency can bring everything built on it to a halt.
I talk to IT leaders about this all the time, and Friday was a reminder that I need to apply it to my own business. If your continuity plan is stored in SharePoint, and the people who run it are contacted through Teams, you do not have a continuity plan. You have a document you cannot open.
Three questions worth asking on Monday morning:
- If email and chat both went down for 24 hours, how would you contact your own staff?
- Which of your critical processes have no offline or alternative route?
- When did you last test the plan, as opposed to writing it?
That matters more as we hand more work to agents. If an agent depends on a service that goes down, it needs a way to pause, hand the task back or use a fallback. Otherwise, it stops.
Ready for Fabric Data Agents at Microsoft Ignite 2026
None of this has dented my enthusiasm. This is my first Ignite, and I am very excited about it. The conference runs from 17 to 20 November at the Moscone Center in San Francisco. Event details
My focus for the week is Microsoft Fabric Data Agents, including my own session, ‘Can You Trust Your Fabric Data Agent? Test It and Prove It’. After a Friday like this one, a session about trust and proof feels well timed. My Ignite session times and details
Fingers crossed the rest of the trip from London goes rather more smoothly than it has so far.
Follow my Ignite week
I will be blogging regularly throughout the week, with a practical take on what the announcements mean for data teams and IT leaders, minus the hype. Keep an eye on the gethynellis.com blog for daily updates from San Francisco.
PakarPBN
A Private Blog Network (PBN) is a collection of websites that are controlled by a single individual or organization and used primarily to build backlinks to a “money site” in order to influence its ranking in search engines such as Google. The core idea behind a PBN is based on the importance of backlinks in Google’s ranking algorithm. Since Google views backlinks as signals of authority and trust, some website owners attempt to artificially create these signals through a controlled network of sites.
In a typical PBN setup, the owner acquires expired or aged domains that already have existing authority, backlinks, and history. These domains are rebuilt with new content and hosted separately, often using different IP addresses, hosting providers, themes, and ownership details to make them appear unrelated. Within the content published on these sites, links are strategically placed that point to the main website the owner wants to rank higher. By doing this, the owner attempts to pass link equity (also known as “link juice”) from the PBN sites to the target website.
The purpose of a PBN is to give the impression that the target website is naturally earning links from multiple independent sources. If done effectively, this can temporarily improve keyword rankings, increase organic visibility, and drive more traffic from search results.