Increased dependency on data centers to drive businesses has led to increased demand for 24/7 accessibility — and data center downtime can have disastrous consequences. How costly is data center downtime? Studies from APC by Schneider Electric have examined the costs of data center downtime by industry: from $1,544.80 per hour in hospitality, to $9,774.80 per hour in retail, to a whopping $14,836.80 per hour in insurance.

To avoid downtime — and the big price tag that comes with it — here are four important areas to address.

1. Make Availability the Top Priority

In today’s connected environment, availability is highly important to end users. Despite this expectation, 68 percent of respondents in a recent Ponemon Institute study believed that their organizations were sacrificing availability in favor of increasing efficiency or lowering costs. But the costs of an outage could far outweigh any savings from these sacrifices. Making availability the number one priority was one of the key identifying factors for organizations the Ponemon study classified as high-performance — the ones that experienced the least amount of data center downtime.

Two of the most popular answers to the survey question of what could be done to prevent future outages were “investment in new or improved equipment” (49 percent) and “increased data center budgets” (26 percent), both of which could improve availability.

2. Get Senior Management On Board

Prioritizing availability will require the support of senior management. Convincing higher-ups of Item 1, making availability a top priority over energy efficiency or costs savings, may take some convincing. It’s also important to make them realize outages are possible and to calculate what they would cost to your organization. Show that planning and careful management now can prevent big problems in the future.

3. Utilize DCIM

Data Center Infrastructure Management (DCIM) software enables tracking and monitoring of assets, power usage, energy efficiency, environmental conditions, and more, creating a holistic view that can help senior management’s understanding of the data center’s performance and function. DCIM users typically experience less data center downtime than non-DCIM users.

4. Conduct Regular Testing

Routine testing of systems and equipment such as generators and switchgear can help to spot vulnerabilities before they become problems. It can also help establish procedures to be followed in actual outages. It’s far better to test your data center’s outage response under controlled conditions than to discover problems during a crisis. Another key equipment category to test and monitor is the uninterruptible power supply (UPS) systems, which are the culprit in many data center outages. UPS technology continues to evolve to help avoid data center downtime, and intelligent systems such as the Smart-UPS series from APC by Schneider Electric offer features such as advanced battery management.

Businesses rely on data centers to power e-commerce and other revenue-generating business operations. Unplanned outages pose a major threat to that revenue. While data center downtime may never be entirely eliminated, management can take steps to greatly reduce its impact and protect that bottom line.