The Five Things You Need to Do to Avoid Network Automation Failures

Areas that have been automated must be checked and updated to ensure they are continually evolving with the network and properly performing their intended function.

Ravi Chandrasekaran

July 17, 2020

5 Min Read
The Five Things You Need to Do to Avoid Network Automation Failures
(Source: Pixabay)

Today’s networks are too complex and important for them to remain manual, but network automation remains one of the most challenging tasks for enterprises to achieve. Automating the network brings numerous benefits as increases in data and devices start outpacing IT capabilities. Gartner found that organizations that automate 70% of their network change activities reduce outages by 50% and deliver services 50% faster.

Automation tools have evolved from operating point products into tools that manage policy and orchestration. For automation to be effective, it needs to encompass the entire network to streamline management. There are five key areas to avoid costly automation failures that enterprises should focus on.


First, IT needs to create end-to-end visibility into all the existing devices on the network for secure and successful automation. This includes the automatic discovery of switches, routers, access points, controllers, and appliances. Achieving this level of visibility requires working to update smaller segments of the infrastructure at a time, which could be as simple as updating the software. IT should be working in stages to update different segments along this transformation journey. It’s important to be cautious of variations in the generations of hardware and software, as you want to be sure each segment is transitioned properly before moving to the next to achieve true end-to-end visibility.


The purpose of implementing intent into your infrastructure is to hide its complexity while creating the benefits of agility, simplicity, and improved security and management. This also relieves the load on the network, boosting performance and preventing other issues. Having to regularly make manual changes often leads to configuration errors and inconsistencies. By defining how the network should operate, automation can continuously monitor and translate the intent to drive operations to deliver the desired outcome.

Here’s how – start by determining and defining how the network should operate and if the infrastructure is able to do intent based on the current devices and software version. Then look at different processes or procedures that, once automated, will help improve the agility of the network. Start simple and small with procedures that you’re comfortable handing control over to a controller to perform the automation. It’s essential to make sure that the controllers are performing their function properly, and the desired results are attained before moving to the next procedure.


To ensure that the network and its automation are always running properly, IT must ensure the network and controllers are running the right software version. Keeping the controller software version up to date ensures access to current capabilities and optimal performance. Also, as part of the software or intent updates, it’s essential to run pre- and post-tests to test their impact on the network. This will help mitigate the risk that an update that’s pushed through creates a failure somewhere in the network, leaving it vulnerable or creating outages. This is another case where pushing out updates should be completed in smaller segments at a time to ensure everything is functioning properly.

Once intent is created and automated via the controllers, the deployment of new or updated hardware and software should be automatically managed. This will enable automatic discovery of the new devices being added to the network or existing ones that undergo updates so that they're assigned the proper configuration to ensure proper capabilities, security, and user experience. This will significantly reduce the IT resources needed as well as prevent errors.

Change Management

To help teams transition to automation, it's important to invest in the proper training and tools to learn how to standardize automation. This also helps teams understand that automation isn’t there to replace them, but to help them improve the network’s function so that it can help the business achieve better performance.

The network is constantly changing with the addition of new devices, patches, software and security updates, and other tasks. Automation can help significantly reduce the costs and number of tasks for IT and can help ensure fast and accurate deployment and configuration. Reducing the number of tasks for IT with automation provides them with greater bandwidth to tackle new projects for improving business outcomes and agility, enabling IT to play a broader role in the business transformation. This transforms IT from being a cost center to a strategic role.


Lastly, integrating automation with other parts of IT can even further enhance the user experience, reduce management burdens on IT, and drive rapid business transformation. For example, by integrating with ITSM systems and automating IT service management capabilities like monitoring and troubleshooting the network or devices, issues can be pinpointed and resolved quickly, vastly limiting business impact. When starting to extend automation to other areas, focus on transitioning the areas that will help improve the organization's business output level, and help it scale.

By addressing these five key areas as automation is implemented across the network, organizations can effectively address their network automation needs. This will reduce network costs and the number of management tasks that enterprise IT teams are dealing with as they manage a constantly growing and changing network, allowing them to work on other important activities for improving the infrastructure.

It’s important to remember that areas that have been automated need to be checked and updated to ensure they are continually evolving with the network and properly performing their intended function. While the automation journey is never-ending, focusing on these five areas will drive the organization toward a better digital outcome and improve its agility.

About the Author(s)

Ravi Chandrasekaran

Ravi Chandrasekaran is Senior Vice President at Enterprise Network, Intent-Based Networking Group at Cisco. He is responsible for product development across Cisco's switching, routing, wireless, IoT, network automation, and analytics portfolios. He leads the engineering effort on the Cisco Digital Network Architecture (DNA). Previously at Cisco, Chandra served as Vice President of Software for the Enterprise Networking Group, where he created a cohesive software-plus-services architecture for the enterprise market segment. Before that, he was Vice President and General Manager of the Services Routing Group when it delivered Cisco's flagship WAN edge router and WAN Optimization platforms as well as products for the Cloud Intelligent Network. He also spent the first part of his career at Cisco, where he developed a significant part of the Cisco BGP and MPLS forwarding protocols and co-authored six RFCs.

Stay informed! Sign up to get expert advice and insight delivered direct to your inbox

You May Also Like

More Insights