For a small law firm, a lost inbox is a lost client list. That is the practical case behind a proposed Canada-wide class action filed against Microsoft after the Microsoft 365 outage of 12 and 13 November 2026, an incident that left most users unable to reach their mailboxes in Outlook and disrupted Teams, OneDrive, SharePoint and Copilot worldwide.
CanadianSME Business Magazine reported on November 17 that Diamond and Diamond Lawyers filed the suit in the Ontario Superior Court of Justice against Microsoft Canada Inc. and Microsoft Corporation, alleging breach of contract, negligence and unfair practices, including failure to meet a 99.9% monthly uptime standard. The representative plaintiff is Metus Lykos Debt Law, a Markham, Ontario business that says its staff could not respond to time-sensitive client emails, hold meetings or open business files for approximately 24 hours, and claims lost opportunities, unproductive staff time and recovery costs.
The filing lands while Microsoft is still explaining what went wrong. 365Fanatic reported on November 16 that the company's preliminary post-incident report, a seven-page document dated 15 November 2026 and tracked in the admin center as MO1550821 and EX1550543, puts the impact between 16:20 UTC on 12 November and 16:10 UTC on 13 November. Microsoft listed Exchange, Copilot, Teams, SharePoint, OneDrive, Defender, Purview, Fabric, Intune, Planner, Power Apps, To Do, Power BI and Universal Print among the affected services, and said the cause was a software defect in its own networking gateway, not a cyberattack.
The claim also lands at an awkward moment for cloud procurement. Enterprises have spent a decade moving mission-critical workloads onto suites that promise uptime but rarely promise consequences, and the filing asks what those promises are worth when a failure runs a full business day.
Key Facts
The action seeks to represent individuals and businesses across Canada that paid for affected Microsoft services, and it asks for aggregate damages plus punitive and aggravated damages. It is proposed, not certified: no court has yet decided that the claims share enough common ground to proceed as a class.
Darryl Singer, head of class actions at Diamond and Diamond, framed the filing around operating dependency. "A 24-hour email blackout isn't a minor inconvenience when your business runs on it," he said in remarks carried by CanadianSME Business Magazine. "There are lost customers, missed deadlines and employees you still have to pay. Microsoft sold reliability. This lawsuit asks who pays when that reliability fails." BriefGlance reported on November 16 that independent telemetry logged the event as Downdetector's largest outage of the past year, with more than 7 million reports and a spike of 129,000 simultaneous Outlook failure reports.
The suit's own arithmetic gives a sense of scale. Canada has roughly 1.37 million employer businesses, and an estimated 1 million of them could use Microsoft 365. Under an illustrative scenario in which 80% were significantly disrupted and half a typical day's revenue went unrecovered, potential economic losses could exceed $100 million. That number is a modelling assumption attached to a pleading, not an audited loss estimate, and its usefulness depends on whether losses across very different businesses can be measured in common.
Microsoft's preliminary account describes a three-part failure. Expert Insights reported on November 16 that a latent defect came from a fix built specifically for Microsoft 365 networking gateway servers, first deployed in July 2026, run worldwide for more than three months without observed problems, and did not apply to broader Windows deployments. Lock contention inside an internal process overwhelmed gateway servers under certain loads, and lab testing missed it because test environments used different traffic and hardware mixes and fewer cores than production hardware. About 15 hours before the outage, Teams failed over from an alternative gateway to the primary gateway shared with Exchange Online, raising load. An unrelated issue the previous day had already taken nearly 14% of gateway capacity out of service.
Together the three conditions produced a cascade. Automatic rerouting pushed traffic to healthy endpoints, concentrating the same load on fewer servers, and within two hours the defect had hit the fleet. At the lowest point, more than 80% of gateway servers were unresponsive. Bulkheads meant to contain failures between regions did not stop the spread because of a separate, undisclosed issue, so the outage went global. Recovery dragged because returning servers were overwhelmed, automated recovery was not built for a failure that broad, and the defect was too old to roll back. Microsoft applied an update restoring the previous code path at 07:20 UTC on 13 November and cycled servers in large batches, sometimes 50% or more of capacity at a time. Consumer Law Group has announced its own investigation into a potential class action, and Victoria Peak Litigation Funding is separately weighing whether to fund one.
Analysis
What this really means is that the fight is less about one bad Thursday than about the architecture of cloud contracts. Microsoft, like most enterprise software providers, advertises a 99.9% monthly uptime commitment whose standard remedy is a service credit rather than compensation for lost revenue. Commercial agreements typically cap liability and exclude indirect, incidental, punitive or consequential damages such as lost profits, which is precisely why the Ontario claim pleads breach of contract, negligence and unfair trade practices rather than pointing at the service level agreement and asking for a refund.
BriefGlance reported on November 16 that limitation of liability clauses are generally enforceable in Canada, especially between sophisticated commercial parties, but that a total failure of mission-critical infrastructure could push courts to revisit how those clauses are applied. That is the opening the plaintiffs need. A 99.9% monthly standard tolerates only a sliver of unavailability, and this incident consumed something close to 24 hours across a gateway that many customers never knew existed, which complicates any argument that the risk was priced into the deal.
The bigger picture here is that certification, not liability, is likely to decide the case's value. Canadian class action rules require a common issue and some workable method of aggregate loss modelling across very different industries, and the proposed class spans sole practitioners, manufacturers and retailers with wildly different exposure. Proving a dental clinic and a freight broker each lost half a day of revenue is one problem; proving the counterfactual is another. Microsoft can also be expected to argue that customers who kept paying after the incident accepted the contractual remedy.
Still, the pleadings have value beyond the courtroom. Filing is cheap relative to the exposure, it generates coverage, and it gives a funded claimant standing with regulators. The gap between what customers think they bought and what the contract delivers is the real subject of the dispute.
Why It Matters
Most of the affected organisations have no legal department and no leverage. ABC News reported on November 16 that RMIT Associate Professor Mark Gregory said compensation is normally available only to businesses under service level agreements, which often require the service to be fully operational for 99.49% of the year, and that where targets are missed the remedy is a percentage of the monthly or annual fee, which "doesn't normally extend to lost productivity." Deakin IT lecturer Fariha Jaigirdar said the agreements cover only the loss of service, not knock-on effects, and that any credit amounts to a small discount on the next bill.
That asymmetry is what the Canadian claim is designed to expose. A credit worth a fraction of one month's subscription is a rounding error against a day of idle payroll, missed filings or abandoned sales calls, and going to court reflects a view that the contractual channel no longer matches the operational risk small firms carry. If the class proceeds even partially, every supplier that depends on Microsoft 365 inherits a question about who documents downtime and who absorbs the cost.
There is a policy dimension as well. Jaigirdar described platforms of this kind as essential national infrastructure, no different from telecommunications and water, and urged a national alert system for outages. That framing, if it gains traction, moves outages out of the private contract column into the public interest column, where liability caps are harder to defend.
Next Up
The immediate step is certification. Diamond and Diamond must persuade the Ontario Superior Court of Justice that the proposed class shares common issues, and Microsoft will argue that individualised loss defeats that requirement. Consumer Law Group's separate investigation and Victoria Peak's funding review could produce competing or consolidated proceedings, and funding matters because class counsel needs a backer for years of motion practice.
Microsoft, meanwhile, said its final review was due later that week, and the difference between its preliminary and final accounts will be read closely by customers weighing a claim. If the final report adds detail on how a defect introduced in July survived three months of production use, the plaintiffs gain a narrative about foreseeable failure. If it narrows the causal chain, Microsoft gains a narrative about an exceptional cascade that no contract could reasonably price.
Comments (0)
Log in or sign up to leave a comment.
No comments yet. Be the first to share your thoughts.