Transform user support with proactive and predictive analytics

In today’s mobile and digital workplace, employees expect and demand IT support that’s fast, reliable and easy to use. Many companies have already tried to reduce call volumes or queue times, offer call-back options, provide self-service and even more personalized, convenient ways to support employees. But what happens when companies use analytics to predict issues and implement proactive user support?

The importance of proactive support services 

No matter how good they are, standard user support processes are largely reactive, providing assistance only after a problem or disruption happens. Because accessing support services takes too long, many employees choose not to report situations and instead, suffer in silence. The cost to business productivity and user morale and confidence is difficult to quantify, but it’s considerable.

Adding elements such as mobility, outsourcing, software as a service and cloud-delivered services, there is an even greater chance for the technology to break down. This puts pressure on organizations to find more efficient and automated ways to deliver support services.

Companies can no longer run in reactive mode. Today, the most productive and satisfied employees are those who have uninterrupted, responsive, reliable access to the systems and information they need. (See Figure 1.) The old legacy processes, such as ITIL, service desk incident and root cause analysis, are reactive responses. Companies need to identify and diagnose developing issues early. In a proactive mode, companies can find delayed services and failing devices and predict which users will have issues.

Figure 1. From a reactive to a proactive IT user support model

Benefits of a proactive support approach

A well-designed proactive and preventive service model will significantly improve overall IT performance and user satisfaction, and it can reduce inbound calls by 20% to 30%.¹ Using a combination of 24×7 monitoring technologies, companies can analyze events to show the real-time state of devices, user activity, locations, applications and business services. It’s now possible to analyze and act on these live, user-centric metrics immediately. And companies can prioritize actions to minimize employee disruption. 

Imagine the delight when users receive a call, instant message or email alerting them to potential trouble and giving them instructions on how to avoid or react to it. Even better, systems can now automatically resolve the issue on the user’s behalf. By generating business intelligence through analytics, companies can use data that helps them pinpoint precisely where to reach out in  he organization with recommended changes. This dramatically improves the support team’s performance, which reduces negative user experiences. Table 1 provides a comparison of reactive, proactive and predictive approaches.

Table 1. Reactive to proactive and predictive

How predictive analytics and machine learning help your business

Predictive analytics can help organizations realize significant improvements in system performance, avoid more common outages and keep employees productive. Predictive analytics look at probabilities and trends, measuring variables to predict how something might behave. Advanced analytics capabilities include ad hoc statistical analysis, predictive modeling, data mining, text analytics, optimization, real-time scoring and machine learning. Through predictive analysis, organizations can analyze past outages to determine root causes and proactively develop strategies to prevent reoccurrence and future downtime. 

Machine learning lets companies more accurately and efficiently identify where technological assets are underperforming and find faults that could lead to downtime. Machine learning can determine “normal” behavior or performance levels. Predictive analytics can then identify patterns or anomalies across the systems on the corporate or even on public networks to predict when an issue may occur. The more analytics that are accumulated and analyzed, the better the predictive capabilities become.

Four steps to predictive analytics

Here are four steps to make predictive analytics work:

1. Clarify the organization’s goals and identify the most relevant data.
2. Prepare the data and determine the predictive model. Once the data is consolidated from multiple sources, the organization can select the algorithms that best match its goals. Test the algorithms with small subsets of the data until the best model has been determined.
3. Deploy the model using all the data collected.
4. Monitor the model’s effectiveness for ongoing analysis.

Over time, the data fed into predictive analysis systems can degrade in accuracy. Adjust the model by collecting and entering additional data or adjusting the algorithms.

By using predictive analytics to forecast the likelihood of outages, companies can proactively plan for operations that run more smoothly. Downtime can cost millions, and when that happens, valuable resources are diverted from more valuable activities such as improving the overall performance of IT systems and support teams.

Six benefits of proactive and predictive user support

Proactive IT support and solutions provide a far greater return on your investment than reactive support. Predicting and preventing problems from occurring — at a fraction of the cost and without the disruption of outages or failures — just makes business sense.

Companies have to expect they will encounter IT issues. However, proactive user support lets businesses foresee problems,  reduce costs and implement fixes before the problems lead to significant losses in productivity and results.

Proactive user support delivers benefits in six areas:

1. Decision making. Gives businesses the insights needed for effective planning and analysis.
2. Budgeting/planning. Helps businesses gain deeper insights through monitoring, continuous health checks and more effective use of scarce IT resources.
3. IT support-cost savings. Saves money and makes employees more efficient, increasing morale and confidence.
4. Uptime. Identifies areas of vulnerability and fixes them before they become serious disruptions.
5. Data security and compliance. Sets security policies and monitors shadow IT as well as undesirable or unauthorized data stores and activities that may make a device or user vulnerable.
6. Response times. Identifies issues before the user calls, and takes action quickly.All issues are priority issues and receive better attention because of more automated and effective resolution.

The bottom line

Proactive and predictive user support, continuous 24×7 health checks and smart diagnostic capabilities improve dependability and offer a more efficient route to outcomes and use of IT support resources. Taking a proactive approach to user support with effective user-experience monitoring and actionable analytics saves money and makes employees more productive.

About the author

Carol Rizio, is the offering manager for the Digital Support and Site Support business in DXC’s Workplace and Mobility portfolio. She leads the go-to-market strategy and brings innovative and market-pleasing support services to DXC customers. Prior to this role, Carol led portfolio and strategic innovation as a director in the Office of Strategy and Technology, managed global operations as account delivery executive for Fortune 100 clients, and led Global Strategic Alliances at Hewlett Packard Enterprise.

Watch the Proactive Support video

1 “The Next Monumental Shift in CX: Combining Big Data and Proactive Customer Support,” Execs In The
Know, Jennifer Turner, February 2020.