It’s Wednesday afternoon and your users are reporting system-wide lag and errors. You contact your infrastructure provider’s support channel, only to be placed in a ticket queue, met with a bot triage loop and bracing yourself for the long wait ahead.
Whether it’s a regional outage or performance hiccups during a product launch, the lack of timely and robust support is something most organizations face at one point or other.
But when every extra minute you’re left waiting to speak to an expert means your platform is performing poorly (or not performing at all), it’s more than time at stake; it’s revenue and reputation. In fact, the average cost of IT downtime is approximately $5600 per minute.
It’s a universal fact that technology will fail sometimes, but the lasting damage is rarely about that single botched disk or DNS error; it’s almost always about support failures.
Standard infrastructure as a service (IaaS) support models tend to break in predictable ways: bot triage loops slow down incident responses, opaque escalations leave customers in the dark, and vendor silos bounce clients between queue categories.
Often, those very same clients are then met by after-hours or weekend closures that leave them completely stranded when they most need help. Narrow that focus to the market giants and there’s an additional, more insipid, problem at play: unequal prioritization of accounts.
One of the most common misconceptions around hyperscale cloud providers is that they must, by default, come with comprehensive support. But, hyperscaler support isn’t rationed equally and only the biggest billing users (or those willing to pay for additional support tiers) get adequate care. For the rest, ‘break-fix only’ is standard, and it can take weeks to get a response after raising a ticket - no matter how significant an issue is to your business.
The net effect is that your incidents last longer, your launches carry more risk, and your team shoulders the very work that your hosting provider should absorb. As one Global Director of Technology at a leading iGaming platform told servers.com:
“You rarely get proper attention for things that matter to your business but are not critical for your [hyperscale] cloud provider. For example, a bandwidth interconnection between two regions is not important for them just because it's causing some delay. Even if this delay is causing you a loss of revenue, you are not going to get traction on understanding what happened and get the fix (or even a hot fix) for the problem.”
In short, hosting provider support teams simply do not serve as an extension of your team: this is where the problem lies.
For businesses that haven’t yet experienced the fallout after navigating infrastructure challenges without access to adequate support, it’s easy to think of it as just another cost center. But this is one of the biggest infrastructure hosting mistakes that businesses make.
When infrastructure support can’t be relied upon to get issues fixed quickly, it leads to a domino effect. For example, if you’re a business launching a new platform to market for the first time, unresolved incidents could delay your entire launch. If you’re forced to spin up more servers to compensate, there’s also additional spend incurred.
Similarly, if you have an established platform that goes down, being unable to fix the problem quickly leads to customer dissatisfaction, engineering drag (as your users’ own support tickets accumulate) and a negative hit to user retention. And if your issue concerns security, there’s also legal and compliance-based risk to consider, including potential fines and penalties. Quickly, what began as a support ticket, starts to hit your bottom line.
But what’s the real cost of a poorly supported incident? Let’s consider an example. Imagine you’re a live streaming platform. You suffered a 45-minute outage during your latest live sports event. Your revenue is driven by advertising and pay-per-view subscriptions and there’s a heavy social fallout when streams fail.
Let’s plug in some numbers to see how much this period of downtime might cost you:
Say your event generates an average hourly revenue of $200,000 and you suffer an effective outage of 0.75 hours. That’s $150,000 in lost revenue. If it takes 15 engineers and three managers to resolve the incident and they’re billed at $120 and $160 per hour respectively, that’s another $5700 on incident handling. The outage also caused a heavy social fallout, so your engineers spend an additional 60 hours on post-incident follow-ups costing $7200.
You owe $90,000 in customer refunds and credits, involving 120 hours of extra support staffing at $40/hour to manage, totaling $94,800 on customer remediation. You also owe your partner advertisers $30,000 in make-goods for lost impressions, spent $4200 on compliance and reporting (30 hours at $140/hour), and signed up for two months of excess compute capacity costing $40,000.
Lastly, you predict that you will incur $144,000 in probable churn (based on $12/month ARPA, an estimate churn delta attributable to the incident of 0.5 % per month and six-month average remaining paid term).
The total cost of the incident is $475,900.
Whilst these numbers serve only as an example, it demonstrates how quickly costs spiral when an incident goes unresolved (even for just 45 minutes) and the additional cost fallout that follows even after an incident is resolved.
If you’ve not conducted this type of exercise before it’s well worth plugging in your own numbers to see how much a similar outage could cost your business.
To work out the total cost of an unresolved incident, calculate the sum of the following:
Lost revenue = revenue per hour x business hours downLost productivity = number of affected staff x loaded rate per hour x hours disruptedIncident handling = role involved in incident handling x rate per hour x hoursCustomer remediation = value of refunds/ credits + extra support hours x support rateImpacted paying customers x average revenue per user (ARPA) x churn rate x remaining contract months
Making our customers happy – and keeping them happy – is what we focus on at servers.com. There are four support principles we follow:
When it comes to something as important as the infrastructure that underpins your business, support should be human and continuous. Due to the complexity and interdependence of IaaS environments, automated diagnostics alone cannot be relied on to diagnose issues. It’s important to have human experts on call to help interpret system-level interactions and more challenging problems.
Likewise, for mission critical workloads, an automated support response isn’t going to get you back online quickly after an outage – but rapid human engagement and clear next steps will. Because of this you should expect a range of support touchpoints: A human-first support team that responds to tickets in minutes, a dedicated account manager who stays with you for the duration of contract, and access to solutions architects and c-suite support as required.
For Nitrado it’s this human support that’s formed the foundation of our lasting and trusted partnership:
“We’ve had a very down-to-earth, human relationship with the servers.com team and the people we met in that very first meeting are still the people we deal with today,” shared Felix Oechsler, CTO of Nitrado.
Ultimately, the best infrastructure hosting provider support teams function as an extension of your own team. The best support is proactive as well as reactive and the best support teams don’t just fight fires (though this is, of course, important), they also help create a solid foundation to start with.
In other words, your infrastructure provider should be innovating and problem solving with your team from the point of initial onboarding and architecture design all the way through to supporting with new product launches and shifts in your business.
When your hosting provider knows your business inside and out, performance can be optimized and acute issues prevented in the first instance. It’s exactly what we’ve done with real-time streaming company, Ceeblue:
“We’ve collaborated to co-develop all the hardware that we currently deploy and many of our most recent cutting-edge advances have taken place on servers.com hardware,” said Jonas Blötz, VP of Engineering, Ceeblue.
Infrastructure requirements can also vary considerably between industries so a blanket approach to solution design and ongoing support is rarely effective. The best providers will be able to demonstrate a specific understanding of your industry and support you in optimizing your hardware in line with those unique industry requirements. Look out for account management teams with specific industry expertise.
“We’ve collaborated to co-develop all the hardware that we currently deploy and many of our most recent cutting-edge advances have taken place on servers.com hardware.”
Infrastructure doesn’t work regular business hours, meaning the benefits of 24/7/365 support in hosting can’t be underestimated. When time literally means money in the face of technical failures or outages, it’s essential that you can count on your provider’s support function to jump in at any time.
So, whether you have a simple query or a complex problem, you can (and should) expect access to support in minutes – even if that’s out of hours or at the weekend. For one of our fintech customers Ark Technologies, this direct ‘always on’ approach to support is a lifeline:
“Even at the weekends, the servers.com support team are always available, and action will be taken in minutes,” said Iyad Yasser, Co-founder and CTO, Ark Technologies.
The continuity of those support touchpoints also can’t be underestimated. You don’t want to be made to jump from one department to another to get the answers you need. Ensuring that you’ll have access to the same account manager for all your global contracts means fewer crossed wires and faster troubleshooting. Jonas continues:
“Unlike previous providers we’ve worked with, with servers.com our account manager handles all of our contracts globally which makes things much easier for us.”
“Even at the weekends, the servers.com support team are always available, and action will be taken within minutes.”
The size of your account should not impact the level of support you receive from your infrastructure hosting provider. No matter what you’re billing per month, you should receive the same quality and care. Admittedly, it can be hard to vet providers on this but there are signs to look out for.
One sign is to ensure a clear and transparent service level agreement (SLA) is agreed upon before sign-up. Similarly, you should seek testimonials from similar-sized companies who have utilized the vendors’ services.
As Isaac Douglas, CRO of servers.com puts it: “whether you’re running a single server or an entire fleet, world-class infrastructure demands world-class support and that’s how we do things at servers.com. There’s no place for favoritism - every customer is a priority and gets the same human-first service.”
Exceptional infrastructure support should be the default, but often the contrary is true. Impersonal ticketing systems, inefficient escalation processes, break-fix-only support, and unequal account prioritization are commonplace.
It means that, for businesses considering where to host their applications, choosing the right infrastructure hosting provider isn’t just about hardware specifications or price, it’s about people. 24/7/365 infrastructure support, a human-first approach, and equal care for every account are what truly protect your business from the hidden costs of downtime and form a solid foundation of resilience.
Frances is proficient in taking complex information and turning it into engaging, digestible content that readers can enjoy. Whether it's a detailed report or a point-of-view piece, she loves using language to inform, entertain and provide value to readers.
