In 1969, ARPANET – short for the Advanced Research Projects Agency – was released. This project, create by JCR Licklider, is considered by many to be the inception of cloud computing. That’s right: cloud, at least the basic precepts and architecture that form its basis, is 50 years old. Technologies are built off earlier ones, and they are also advancements at meeting the same core needs. One of the key ones that is considered valuable in a technology is its capacity to enhance reliability and reduce downtime.
The threat of downtime is vague until you have numbers, though. Calculating downtime can be achieved by using a basic formula that adds major costs – lost revenue, lost employee productivity, and money paid out directly to customers. You can then expand your understanding by incorporating downtime frequency over time.
Formula to calculate downtime
To calculate downtime, you can start with this basic formula:
Cost of downtime = Revenue + Employees + Customers
COD = R + E + C
Note that you may want to add an additional variable, M, miscellaneous. That category would include the cost for any consultants that you use for data recovery or related services. The below goes into more detail on each of these key variables.
Note that this plan should be shared with your customers to the extent they are impacted. Here is how to determine how much revenue is lost to downtime:
- Figure out and list the revenue-generating parts of your business.
- Determine what the per-hour revenue is on average for each of them.
- Determine the significance of IT downtime. A 100% ecommerce company might lose all revenue, while one that is half brick-and-mortar would be 50% impacted.
- Add up all the lost revenue from each revenue-generating component.
Employee productivity loss
Your productivity is determined through a formula that reuses some of the variables from the above COD one but with different meanings, as follows:
Cost of downtime to employee productivity = number of impacted employees x percent to which the outage impacts them x per-hour cost of average employee x the number of hours of downtime
Productivity cost = E x % x C x H
An important aspect of downtime is what you will need to pay customers. You need to pay out whatever is mandated in your service level agreement (SLA) contracts. Plus, include in that figure the costs of any apology money that you pay – such as when an airline gives out flight vouchers or hotel rooms.
Looking at downtime broadly: frequency
As indicated in the introduction, technology changes, but many of the core considerations with technology remain the same. Hence, the way to approach downtime is fundamentally the same as it has been for years, as indicated by this plan that is 15 years old and still relevant – incorporating frequency:
1.) Determine what elements of your business you are considering related to downtime. Those might be any of four types: data, systems, property, or people. Often data and systems are considered.
2.) Figure out what it is that you are safeguarding. You must be concerned first and foremost with protecting your core competencies. Your value in the marketplace is determined by those ultra-valuable core competencies.
3.) Focus on business functions. You want to put approximately 80% of your attention on the 20% of your data and applications that are most critical. You might include email, customer-facing apps, and your primary operational platform.
4.) Classify different lengths, frequencies, and types of downtime. You might have a failure related to a single data center, or could suffer a national, regional, or branch outage, for example.
5.) Determine full costs of downtime. Those extend beyond the above to the damage to your brand, potential litigation, etc.
Considering all costs will get you to your hourly cost of downtime.
Then, finally, you are able to use this formula to figure out downtime over time:
Frequency x Duration x Hourly Cost = Total Cost
How to convey cost of downtime to management
Three steps can help when you want leadership at your organization to understand the critical value of uptime and why downtime is so important to avoid. First, establish the problem using broad data. Second, create a visual. Third, communicate your internal numbers to those who can invest in downtime prevention.
Know the industry numbers.
Establishing that downtime is an issue generally can help before you drill into specifics at your organization. The cost of downtime continues to rise over time as digital interactions grow more and more important to how businesses operate and bring in money. Andrew Lerner of Gartner noted in 2014 that the average downtime cost for a company was $5600 per minute, adding up to $336,000 per hour.
Really, though, there is far more diversity in downtime costs than that, noted a study cited in Channel Futures. Per-minute cost actually ranged from $137 to $17,244 among the firms studied in that analysis.
However, it is important to note that there is a range of different downtime amounts that would apply to different organizations. Size and industry greatly impact the amount. An hourly range from Avaya suggests typical hourly costs of $140,000 to $540,000.
Use a picture.
You need to translate your technical understanding of downtime into something that is easy to grasp for a nontechnical audience. Get away from the jargon by translating it into an image of how you understand the situation.
For example, as in this image posted in CIO, you have a wide area network (WAN) that is for a staff of 1000 working at 5 remote offices. You need to install a router at your home office where all the infrastructure for your operational business software resides, along with routers for each of the 5 remote locations.
If you lose a router, your productivity for that portion of your staff could plummet.
Convey your own numbers.
The other thing you want to do is to make your numbers known. Granularity can help immensely in this effort. You need to know the cost of a failure of 1 router.
Here is how you figure that out:
1.) Decide, roughly (through HR data perhaps), how much an employee at a particular location makes on average per hour.
2.) Determine what the amount is that people are being rendered unable to produce because of the downtime. That number is called the productivity impact factor and is a percentage.
3.) Figure out what your productivity impact is through the following: Cost of downtime to productivity = Impacted employees x Productivity factor x average hourly salary.
Using the above example, the formula would work like this:
1000 employees involved x $20 per hour on average x 0.5 productivity impact (i.e. 50%) = $10,000 per hour
Since just the cost to your productivity is $10,000, and if you estimate the cost of a backup router installation at $8000, you can make a case for putting in that router with under an hour of downtime.
Uptime and high performance for your organization
Are you in need of a high-performance system so that your company does not suffer downtime? At Total Server Solutions, our infrastructure is so comprehensive and robust that many other top-tier providers rely on our network to keep them up and running. See our platform.