When the Cloud Goes Dark: Data Center Claims And Specialized Adjusting Expertise

Data centers don’t make headlines until something goes wrong. Then, suddenly, everyone notices.

These huge facilities keep the internet working, cloud services running smoothly and AI programs going strong. They are some of the most important and valuable parts of our modern world, even though most people don’t think about them much. Building one big data center can cost hundreds of millions or even billions of dollars. The computers and equipment inside run all day and night, every single day, with almost no room for mistakes or breaks. Any stop in service can cause major problems.

When something breaks, the problems start right away. A report by Emerson Network Power and the Ponemon Institute says the average cost of downtime for a major data center has climbed to $8,851 per minute, up 38% from 2010. Yes, that’s per minute. At that rate, just a few hours offline can lead to losses as big as what some companies make in a whole year.

A Loss That Never Stays Simple

A data center can go down because of a fire, flood, power outage or a cooling system breakdown on a very hot night. These things sound scary, and they can cause real damage to the building or equipment. But physical fixes are often not the hardest part of dealing with the problem.

As soon as the systems stop working, other losses pile up quickly, becoming a business interruption or extra expense. Companies that use the data center have contracts promising certain levels of service. When those agreements are broken, penalties kick in, customers demand compensation, reputations suffer and lawyers might even get involved. By the time everything is fixed and back online, the original damage to the building or machines might only be a small part of the total money lost. The extra costs from lost business and penalties can be much bigger.

Real-world incidents show just how fast losses can spiral. Consider what happened to Delta Airlines in August 2016. An electrical equipment failure knocked out the airline’s systems, grounding about 2,000 flights over three days. Delta was obligated to issue refunds to every affected passenger, and the cascading costs from cancelled flights, compensation, and recovery efforts, added up fast. The airline reported $150 million in total losses, more than 200 times the industry’s average per-incident cost at the time.

The physical cause was a single piece of failed equipment. The financial damage came from everything that followed. Getting everything running again is not easy or quick. Data centers use special high-power systems, custom-built equipment and unique cooling setups. These parts are not easy to find in a store. Even in normal times, it can take weeks or months to get them replaced. Right now, with so many new data centers being built for AI and cloud services, the wait times are even longer and there is high demand for equipment. While waiting, the costs continue adding up each minute the center is down.

On top of physical problems, cyberattacks are becoming more common. Hackers can shut things down or cause damage digitally. Now, physical damage and online attacks often mix, making the situation harder to fix. Insurance policies from years ago were not written to handle these issues, so recovery gets even more complicated.

The Coverage Is More Complicated Than It Looks

Most people who own buildings know the basics of their insurance. But for data center owners, the basic coverage often is not enough.

Regular property insurance policies usually do not cover data or digital information very well. Since most of the value in a data center is in the digital information (like servers full of important information), there can be big gaps in what gets paid for. Business interruption insurance has its own rules, like waiting periods before it starts paying, and specific ways to calculate losses. Insurers look closely at whether the operations team did everything possible to stop extra damage from happening. If the carrier thinks more could have been done, they might pay less or even deny the claim.

The OVHcloud fire in March 2021 makes this point plainly. A blaze destroyed the company’s SBG2 data center in Strasbourg, France, taking around 65,000 customers offline. OVHcloud’s IPO filing projected the expected cost at €105 million (roughly $122 million USD), while the company’s insurers committed to covering €58 million of that figure. More than 140 customers subsequently joined a class-action lawsuit seeking more than €10 million in additional compensation. For any data center operator reading those numbers, the message is simple: having insurance is not the same as being covered.

Without experts who know both the tech side of data centers and how insurance works, claims can take months or years to settle. Sometimes they end up with much less money than they need, or they get turned down completely. For anyone running a data center, those results are unacceptable, as the data center business depends on fast recovery.

What Good Adjusting Actually Requires

Dealing with a big data center claim requires someone who really understands what they do and the difficulty to get them running again.

A good adjuster works with IT experts and repair teams to check the damage carefully. They make detailed records that show the full picture of losses, including exact calculations for business loss, prices for broken equipment, and extra costs to keep services running while repairs are underway.

They will also need to explain clearly to the insurance company why a key part might take six months to arrive or why a quick temporary fix adds its own expenses and dangers. This kind of knowledge does not come from just doing regular insurance claims. It comes from people who have worked inside data centers and know the challenges.

Get Ahead of It

The smartest time to bring in a nominated adjuster is before a claim, during the policy buying or renewal process. That way, there is time to read the policy carefully, find any weak spots in coverage, and build a strong relationship with the data center’s risk managers and operations team.

When a real problem does happen, all that preparation pays off immediately. The nominated adjuster can move faster, with stronger paperwork and proof, and the whole claim process goes smoother, avoiding delays that cost even more time and money.

Data centers are getting bigger and more complicated every year. The huge growth in AI, more cloud services, and all the new digital information being stored means we need more of these facilities, and they are becoming even more important. The problems and claims that come from them should get the same high level of skill and care from the nominated adjuster that went into designing and building them in the first place.

In short, data centers are the quiet backbone of our connected world. When something goes wrong, speed and expertise matter. Working with a nominated adjuster helps organizations respond faster, minimize disruption and get critical operations back online with confidence.

Murchek is an executive general adjuster with Sedgwick. He has more than 15 years of experience handling major and complex property losses nationwide. His expertise spans commercial property, builder’s risk, boiler and machinery, inland marine, equipment breakdown, earthquake, catastrophe and residential property claims.