Website downtime: causes, dangers, ways to fight
A website is the face of any Internet project, so visitors want as little time as possible between clicking on a link and its appearance on the screen. According to generally accepted statistics, most users expect a website page to load in no more than two seconds. A delay of just one second can lead to a huge drop in sales.
If even fractions of a second can be so critical, it’s easy to imagine how the site will be affected by its complete inaccessibility during the so-called downtime period. If a website of any subject regularly “downs” or takes a long time to load, you can surely say goodbye to the chances of a stable audience, not to mention loyal customers and constant sales.
In this article, we will talk about what regular problems with site availability can threaten owners with, highlight the main causes of downtime from a technical point of view, and give practical recommendations for preventing and combating this dangerous phenomenon.
What is downtime
Downtime or downtime (from English downtime) is when production or business processes stop due to the unavailability of applications, technical failure, network outage, or natural disaster. The term is usually applied to networks and servers but can refer to any process involving machines.
During downtime, the computer system, server, or network is offline or unavailable, and Internet-based enterprise employees cannot perform production tasks or provide customer service. Such situations can paralyze the work of an organization critically dependent on Internet tools and, in some cases, even cause irreparable damage to it.
The opposite term to downtime is uptime or uptime. Uptime is a measure of the reliability of a system, usually expressed as a percentage of the time a machine remains operational or available (for example, 99.99% uptime).
Downtime types
Their administrators provide planned downtime for sites (planned downtime) in advance for routine maintenance or inspection, hardware/software upgrades, repairs, or testing. Typically, such downtimes occur on holidays, weekends, or after hours, which allows you to maintain a relatively uninterrupted operation of current business processes and not reduce employee productivity.
Planned downtimes are essential to keep systems, applications, and servers up to date-and running at optimal performance. IT can also run tests during planned downtime to identify potential threats and fix them before they become real problems.
Unplanned downtime is unintentional, unforeseen, and can happen at any time. Many factors, including hardware or software failure, human error, cyberattacks, or natural disasters, can cause unplanned downtime. Since unplanned downtimes are always a surprise factor, preventing them can be challenging.
The main reasons for downtime
cyber security breach
A sharp increase in the load on the network or server resources of the site does not always mean that it has attracted increased attention from visitors. Often, such consequences are provoked by the activity of advertising or search bots, as well as the activities of cybercriminals.
A distributed denial of service or DDoS (Distributed Denial of Service) attack is among the most popular intentional violations of website cybersecurity. With the help of distributed networks of infected machines (botnets), attackers overload the target resource with fake traffic and make it inaccessible. Similarly, ransomware hacking and phishing attacks can negatively impact a site.
Low-quality hosting
Miscalculations in the hosting selection process are among the most common reasons for website downtime. Investing in advanced design, modern UX, or site speed optimization is only possible if the hosting site of the resource matches its technical needs and the size of the potential audience.
Most often, downtime occurs if shared (“common”) or virtual / web hosting is chosen. And this, even though the providers of these services are the largest players in the industry. The terms “web hosting” and “downtime” are often used together in the very name of this type of hosting. Each site on virtual hosting shares resources with neighboring resources, which can be up to several hundred on one node. Any problems that occur on other sites, from security holes to increased load due to traffic, can negatively affect the performance and uptime of the neighbors.
Human factor
Regardless of the cause, human error is among the most common causes of unscheduled downtime. An employee who inadvertently deletes corporate data, accidentally unplugs a cable, or forgets to follow standard security protocols can cause costly downtime for a company’s website.
Hardware or software failure
Outdated hardware or software increases the likelihood of application and computer system failures. The same reason makes individual employees’ work less efficient, inevitably lowering their overall productivity.
Man-made and natural disasters
Natural Disasters: Hurricanes, floods, earthquakes, and major human-caused accidents can disrupt power and communications or damage the server hardware that hosts the site. Events like the 2021 fire that destroyed the server storage of Europe’s largest hosting provider, OVH, could be catastrophic for an enterprise of any size.
Incorrect device configuration
Device misconfiguration is another major cause of unplanned downtime. Configuration errors can create security holes in your network, making it vulnerable to cyberattacks. To avoid misconfiguration errors, you can automate the process instead of setting the parameters manually. Test configurations in a lab before making changes to your system.
Bugs in the server’s operating system can also affect performance and lead to security issues. If patches are not applied promptly or without proper testing, applications can be corrupted, and the server may crash.
Traffic redundancy
What site owner wants his web project to become popular? However, when it suddenly happens, only some, even the most advanced sites and services, are ready for the technical consequences.
An abnormally high flow of visitors or their activity can lead to a 500 Internal Server Error or, worse, bring the site to a complete halt. Take the famous selfie of Oscars host Ellen DeGeneres in 2014 as an example. The photo was retweeted over 3 million times in less than 24 hours, bringing down the popular resource.
How much can downtime cost
The cost of downtime may vary depending on the size and nature of the Internet resource, as well as the duration of the downtime. According to a Statista survey of site owners worldwide, 25% of respondents reported that the average cost of their server downtime is between $301,000 and $400,000 per hour.
A Fortune 1000 study by IDC, a global IT infrastructure company, found that the average cost of a global IT infrastructure outage is $100,000 per hour, with the average total cost of unplanned application downtime per year ranging from $1.25 billion to $2.5 billion.
According to a survey conducted by IB and DRaaS (Disaster Recovery as a Service) provider Infrascale among CIOs and CTOs of American small and medium-sized businesses, the cost of downtime of their Internet resources ranges from 50 to 10 thousand dollars per hour. At the same time, regular losses from downtime for such IT giants as Amazon can be more than $ 220,000 per second.
How to calculate the cost of website downtime
The following step-by-step algorithm will help determine the real cost of downtime for an enterprise of any size.
- Determine the lines of business that generate income.
- Calculate the average income per hour: (per week / 40 hours) or (per month / 30 days).
- Determine what percentage of revenue-generating areas depend on site uptime. For example, an e-commerce business is 100% dependent on uptime, while this figure may be only 20% for a brick-and-mortar store.
- Calculate how much revenue is lost per hour in each business area during the downtime.
You can also use one of the following formulas to accurately calculate the performance loss in the event of a downtime:
- Lost productivity = (number of affected users) x (performance impact [percentage]) x (average salary per hour) x (downtime).
- Loss of productivity = (number of affected users) x (performance impact [percentage]) x (average profit per employee) x (downtime).
- Loss of productivity = (transactions per hour) x (percentage of affected transactions) x (average profit per transaction) x (downtime).
What else is dangerous downtime
In addition to the obvious financial losses, there are other downtime consequences that organizations dependent on Internet presence have to deal with. Depending on their frequency and duration, they can have both short-term and long-term business impacts.
Loss of performance
Critical enterprise systems become unavailable when a hardware failure or a network outage causes downtime. With essential applications, systems, and network services, employees can do their jobs and unwittingly idle. The longer the downtime lasts, the more the company’s productivity decreases.
Missed Opportunities
In today’s interconnected digital world, where businesses rely heavily on uptime and application availability, a moment of downtime can harm business growth. If customers cannot access products or services, and employees cannot support them without the right tools, many current and potential customers can be lost.
Image Damage
Customers always expect fast and flawless service in a dynamic business landscape. Even a single site downtime at a critical moment can jeopardize the credibility and reputation of an entire organization. And repeated cases of unavailability of a web resource are guaranteed to cause customer dissatisfaction and increase the number of negative reviews that directly affect a business’s reputation.
Data loss
Downtime affects not only your business but also your customers. Due to cyberattacks and server or network failures, data corruption, damage, or theft can accompany downtime. Unforeseen downtime, such as a server shutdown, can reveal valuable information to attackers or create security holes that cybercriminals can use to gain unauthorized access.
Downtime Prevention and Mitigation
Downtime is unpredictable and can happen to any site, regardless of its profile and size. So how do you prepare for the unexpected? Here are some tips to help increase website availability and minimize the potential impact of downtime on your business.
Choose stable and secure hosting.
The sustainable availability of a website at a basic level depends on how the hosting service provider can ensure the stability and security of the resource it hosts. When choosing a hosting provider, you should pay attention to the speed and quality of work of its technical support specialists, which reviews on specialized resources can assess. Equally important is the level of hosting protection against cyberattacks.
Deploy an anti-DDoS solution.
Of all types of information security threats, DDoS attacks pose the greatest threat to website availability, as they can disable the website of a large company or government institution in a matter of minutes. Any enterprise whose activity or reputation is critically dependent on an Internet resource must protect it in advance from such cyber attacks by installing a modern system that can cope with the most powerful and prolonged DDoS attacks at different levels of the OSI network model.
Organize ongoing monitoring
The introduction of network monitoring tools, server software, and hardware allows you to establish constant monitoring of the state and performance of the site that depends on them. Tools like these, including the free ones, add an extra layer of protection and help you detect problems early and fix them before they crash your system.
Set up a “service” page on the site.
During the planned downtime, the site should be configured to display a stub page with the relevant information. This will notify visitors that there are no failures in the operation of the web resource and that it will resume work soon. A “site maintenance” or “service unavailable” page with a short message shows that you are in control and helps maintain customer trust.
Optimize download speed
In the fight to reduce downtime, you can work backward by finding ways to make your site run faster and thus increase your uptime.
Create and check backups regularly.
During long downtime, pre-saved site backups (backups) can become an extreme starting point for restoring its full functionality. With regular and well-organized backups of critical data, you can quickly restore them and start working normally, even in a devastating disaster. You should regularly check your site backups’ integrity to ensure the data recovery is working.
Train employees
The human factor will always be one of the main causes of site failures, but the ability to avoid such errors is greatly increased if work is done to improve the IT education of staff. All employees with access to corporate information systems should be aware of common cyber threats such as malware, ransomware, social engineering techniques, and phishing attacks. It is also necessary to convince employees to strictly follow the standard information security policies and procedures established in the organization.
Conclusion
The loss of site availability is difficult to predict. Still, every owner and administrator of a web resource can well prepare for this unpleasant but, unfortunately, common occurrence in the modern IT landscape. Using the above recommendations, you can avoid financial and reputational losses from downtime and minimize the likelihood of it occurring.
