5 Tips for a Successful AIOps Deployment

By Ranjith Lewis and François Habryn 

Successful technology modernization projects are driven by people and processes. Yet all too often, companies spend the majority of their time and resources acquiring digital tools rather than promoting organization-wide acceptance of the new technology.

Overlooking the human element inherent in digital innovation can undermine almost any transformation initiative. Roughly 70% of change programs fail to achieve their goals due in large part to employee resistance, lack of management support, and other “soft costs” of organizational change, according to McKinsey. As a result, investments in technology that are designed to improve customer satisfaction, enhance internal operations, or advance other corporate objectives tend to yield costly overruns, low adoption rates, and poor business outcomes.

There’s a better way.

When planning any large-scale digital innovation or a new way of working, think strategically about the nontechnical aspects of the initiative. Building awareness, gaining buy-in, and reinforcing change throughout your organization pave the way for a much smoother journey than focusing on the technology alone.

Experience offers proof. These five tips—learned through helping scores of customers deploy artificial intelligence operations (AIOps) across their organizations—can help put your company on the right path to successful technology implementation.

1. Humanize your approach. As an emerging technology, AI is often mistrusted and misunderstood. Some in the workforce fear it could make their roles obsolete—but it’s more likely to help people supercharge their capabilities. Your employees must trust the new technology before they can use it to make informed decisions or produce specific results.

Insights generated by AIOps can automatically predict emerging issues with a high degree of certainty. When your teams grow comfortable with machine-generated insights, they can investigate anomalies and resolve related issues before they experience service disruptions or other problems.

It’s a fundamental shift in operational thinking to move from a reactive model that measures mean time to detect and mean time to repair (MTTR) to a predictive strategy that gauges mean time before failure and mean time between failures. That’s why, when deploying initiatives like AIOps, you need to consider how it will affect your people and processes after implementation.

If you or your employees need assurance about the effectiveness of a technology, start your modernization journey with an experiment. With AIOps, you might continue responding to incidents after they arise, as you normally would, while using AIOps following remediation to investigate underlying causes of the disruptions. The resulting data from this use case shows how insights about some anomalies can help you predict changes in operational performance.

After formally implementing technology that’s backed by positive outcomes, seek out and incorporate employee feedback. Make a point to:

• Solicit comments and answer questions from users.
• Diagnose and address any post-implementation issues.
• Celebrate successes and team engagements.

Involving employees throughout the process and continually refining the new ways of working help foster goodwill and bolster adoption throughout your organization, leading to sustained progress.

2. Secure executive sponsorship. During the planning stages of technology modernization, your CIO, CTO, and other executives must align the transformation goals with your organization’s overall business strategy. This unified vision helps IT leaders engage their teams and prepare practitioners for change.

Once executive sponsorship is in place, identify advocates within your company who will drive change daily. These individuals need to champion modernization among their colleagues, highlighting the benefits the new technology can deliver.

In the AIOps case, three roles are particularly important:

• Site reliability engineer: Determines business needs, putting software and systems in place to promote adoption
• Operations leader: Ensures teams remain committed to the implementation strategy and full technology adoption
• Data engineer: Builds a flexible and scalable AIOps architecture 

After the implementation begins, your executive team needs to continue providing direction and support. Their ongoing leadership is vital to improving buy-in and usage rates throughout your company.

3. Communicate reasons for change. If employees at every level appreciate why your company is deploying new technology like AIOps, they may be more receptive to change. Put the journey in perspective by:

• Outlining operational value and business opportunities, such as enhanced customer experiences
• Showing how the change aligns with organizational and IT strategies
• Explaining why it’s the ideal time to implement enhanced technology
• Detailing ways innovation can improve their employee experience and productivity

Consistently promoting these messages helps increase employee buy-in and acceptance across your organization.

4. Implement changes incrementally. Employees will either embrace or resist modernization based on how useful they perceive the technology to be. A phased adoption helps demonstrate the value of transformation.

In the AIOps scenario, first apply the technology to a single use case. After achieving positive results with one data source and a defined area of scope, begin introducing AIOps in additional environments using multiple sources and larger amounts of data.

A methodical approach helps your team gain confidence in AIOps-enabled decision making and the technology’s ability to automatically deliver insights, after which the team can move to more complex implementations—while also coming to realize that AIOps is intended to augment employee capabilities, not supersede their expertise.

5. Measure and reward. During periods of major change, employees can be highly motivated by how they’re measured and rewarded. Results from numerous AIOps deployments confirm that technology adoption rates increase when project and organizational change teams and executive sponsors work together to:

• Demonstrate business impact at the departmental level, such as with a reduction in application downtime and customer incidents, which helps facilitate company-wide adoption.
• Develop incentive programs to recognize employees when they reach select milestones, such as a reduction in tickets and incidents, MTTR, or service and application availability.
• Measure post-implementation engagement and adoption rates across your organization as part of daily workflows.

No two technology modernizations are the same. However, successful AIOps deployments of every size and scope support one widely applicable lesson: regardless of your IT initiative, user-focused thinking drives greater business outcomes.

Learn how Kyndryl can help with your AIOps deployments and download our white paper Defining the Journey to AIOps.

Ranjith Lewis is chief technology officer at Kyndryl Denmark. François Habryn is associate partner for cloud, applications, data, and AI at Kyndryl Switzerland.