Numerify’s Top 5 Best Practices For Incident Reduction

Are your employees bogged down by the sheer volume of tickets being logged every day? Isn’t it frustrating to see the same kind of incidents being raised repeatedly? What if we could find patterns, do root cause analysis, and prevent tickets from being raised in the first place?

Most sophisticated IT organizations invest heavily in incident management systems to enable smooth resolution processes. However, a lot of them struggle to meet Service Level Agreements (SLAs) because of the huge number of incidents being raised every day. Every incident that is raised in the system costs money and man hours. Employees are so stressed and busy resolving mundane incidents that there is little or no time left for innovation!

Even in a world focused on process automation and AI to cut costs and resolution time, you still need processes and mechanisms to reduce incident volume. Here, we take you through a few practices that can help you reduce your incident volume significantly – read on!

1 – Establish accountability

The first step to this is establishing accountability. Start by analyzing the data, categorize the incidents based on a predetermined parameter and identify leaders for every category. Incident volumes can be refined by having a standard metric across the organization and leaders handling them can be suitably rewarded for bringing down the numbers.

2 – Practice strategic Problem Management

The second most important step is to focus on problem management rather than incident management. To fix a hole in a pipe, would you replace the pipe or apply adhesive tape to close the hole? Break down the incidents based on top drivers, user locations, departments, etc. to identify problem areas and focus on resolving the problem. You’d be surprised to see how solving a specific problem can result in reduced incidents.

3 – Mine your data for root cause

Another best practice is to make use of data mining. Having huge volumes of data can be a boon and companies are leveraging techniques that go beyond traditional operational reporting. Looking at incidents alone may not give you the complete picture. Align incident data with adjacent data sources to obtain a 360-degree view of all incidents. You may already have a rich volume of data such as audit data, problems, or change log. Make use of these to better understand the root cause and drive incident reduction.

Text analytics is another important area that can be tapped into by an organization. Analyzing unstructured or free-form data like descriptions, comments, work notes, etc., and identifying keywords can help understand incident data. For example, say there are scores of incidents with the keyword ‘WebEx’. However, the resolution time for these incidents was found to be very low. Further analyses reveal that these incidents are informational in nature and could have easily been prevented with a knowledge base article!

4 – Stop chasing your tail

One more important step is to identify and eliminate processes and activities that are generating unnecessary incidents. It may be seen that most often, support engineers identify certain incidents as soon as they come in and these normally involve a simple fix. A bulk of these incidents usually comprise standard requests, auto generated false alarms, duplicate incidents, and reassigned incidents. Leaders must find ways to automate or prevent such incidents. Focus on improving your knowledge base to empower end users, so that such incidents are not logged.

5 – Prevent incidents caused by change

Lastly, driving better change management. Every change you make to your applications or infrastructure has the potential to cause incidents and disruption. E.g., deploying changes with insufficient test coverage can give rise to major incidents. As leaders, you must identify potential risks of a change and formulate ways to mitigate them. Analyze incidents caused due to changes, identify potential risk drivers associated with change pipeline, observe trends, and bucket these changes based on priority.

You can learn more about how a machine learning approach to change success can help reduce incident volume in this blog post.

Want to learn more?

The rewards of reducing incidents are significant – fewer business disruptions, lower IT costs, and the advantage of having your employees concentrate on innovation. And the practices required to be put in place for this to work are pretty simple. Finally, it is all about transitioning from incident management to better problem management!

Visit our resource center for more similar topics