Achieving Success In The Cloud Through The Hard Work Of Simplification

The cloud continuum presents one of the greatest opportunities to affect a lasting culture of innovation and enhanced customer experience. Mitel, as a major provider of Unified Communications as a Service, or UCaaS, has fully embraced and committed to exploiting this opportunity. Over the past few years, we have made the transition to cloud computing. We serve thousands of companies with UCaaS services for call centers, internet-based phones, and online collaboration tools.

Our transition to cloud-based technologies has allowed us—and our customers—to innovate faster and broader within the area of cloud services and products. We leverage cloud computing to gather data from diverse places, analyze it for less cost and effort, and use the information in more aspects of our daily work. That results in a new and better working experience for our staff, while delivering faster, more responsive, and empathetic experiences for our customers.

For example, one key metric we measure is service uptime. We recently achieved six consecutive months where our service uptime achieved “four 9s,” or capable of customer use more than 99.99% of the time. Our new goal is to extend the uptime metric to discrete services consumed by our customers through the introduction of new platform and infrastructure architecture and technology. 

While novel approaches to improving “uptime” are great, it’s only the start. Building and supporting a better innovation culture forced us to think critically about how cloud technology adds value and purpose to service delivery in other ways. We leveraged this knowledge to simplify the way all organizations within Mitel work together to improve the full lifecycle experience for our customers.

How cloud changes your job: From “things” to “experiences”

Cloud changes the conversations we have around enterprise technology. Before, we talked about technical needs, such as the number of servers that needed to be provisioned or whether the pool of virtual machines was the right size. Our focus was on hardware, software, and specifications. These considerations remain important for requirement compliance and traceability. Now, with the cloud, servers have become resources that can be scaled and automated to meet business requirements, allowing the conversational shift from provisioning things (like hardware) to enabling experiences (like the improved business outcomes for our customers). 

Now, instead of asking “what is the portfolio of technology we need, and how does this serve our business priorities?”, we can ask “do I have enough on-demand cloud resources to meet the performance needs for an improved customer experience?”

This new focus on enabling experiences, and not simply provisioning things, reverberates across other technology choices we have made at Mitel. For example, we discovered we were monitoring nearly 180,000 system alerts a month. That is far too many for any team to deal with productively, so we set about reducing that number. By identifying true stateful and actionable alerts, we were able to decrease the number of monthly alerts to around 2,000. This reduction helped us to focus on alerts that actually matter to running the customer’s business. As a result, we were able to improve our preventive monitoring (i.e., fault avoidance), helping us to achieve an 85% reduction in mean time to respond to issues—ultimately providing a better customer experience.

The previous firehose of non-critical alerts tended to indicate things like “this node is down,” signaling a hardware problem. Now, if we receive one of the 2,000 critical alerts, it tells us something critical, such as “we lost capacity due to the loss of this phone switch.” That is a focus on the experience the system was having and the effect on operations, instead of a bunch of sensor readings and why that matters to the technician reading the result (who should be thought of as the customer for the data). Creating that capability involved looking at all of the traditional data points in a new way and then simplifying the presentation in a way that focused on the customer experience. 

We extended and broadened our transition to include the use of KPIs for both “parent” and “shadow” KPIs. The shadow KPIs, or key performance indicators, helps us build our parent KPIs that are used as a measurement of how well we are delivering to our operational goals for improved customer experience. For example, “30 days after operational availability improved, how did it affect performance elsewhere?” We are evaluating ourselves more holistically, and on the experience the system is delivering to our customers.

Related: Forrester report: “New Technology: The Projected Total Economic Impact™ Of Google Cloud Contact Center AI”

How “elegant operation” experiences lead to more innovation

Simplification and standardization results in what I call an “elegant” means of operation. Elegance requires innovation, but it means teams must have the time, means, and authority to put forth more ideas with greater impact. Elegance shuns big numbers for their own sake—like how many multi-terabyte data lakes you have—and focuses on the quality of that data towards achieving understanding. 

As in nature, less done right is often more. For example, by taking a broader view of the system as an experience, we managed to lower the number of critical monitoring tools we use from 25 to four. We focus on stateful information, so we stay focused on the overall goals of system performance, service delivery, and customer experience. It took work to reach that level of simplification and standardization, while maintaining many aspects of the earlier tools in the new ones. In the end, it allowed us to focus on the “critical few;” thereby reducing and eliminating needless and non-productive work.

Extending the preventive monitoring discipline, which addresses symptoms before they become problems for our customers, included the evaluation and use of machine learning for customer retention, looking for the patterns that indicate customer satisfaction and better experience by our customers.

Here is how elegant operation works for one of our most important jobs at Mitel: system monitoring coupled with world-class cloud best practices. Working with Google Cloud, we are innovating from the typical forensic model, which examines data to analyze and eliminate past mistakes, to a predictive model that looks at present data to stop malfunctions before or as they happen. This can include something like carrying out a regression analysis on recent network performance to forecast where and when a fault might occur. This helps to improve the operational efficiencies of our SRE and vNOC teams by allowing for a learning and continual improvement culture, leading to enhanced best practices around critical tasks like incident, change, and problem management.

This new predictive model is a more elegant form of monitoring, as it relies on understanding the interactions of several key data points and operational actions. Engineering this requires a more complex understanding of the overall system experience. That is achievable by having standard behaviors (both system and practices) and common data points—the results of simplification—all along the way.

Back-end innovations like this affect the customer experience directly. Our recent Mean Opinion Score (MOS), which is used to measure the quality of experience, has improved significantly through our enhanced monitoring. We achieved a MOS score of > 4.03 (good to excellent) for nearly 99% of our call volume. This represented a 2% improvement over a six-month timeframe.

Related: Learn how to unlock strategy, resources, and technology with “Google’s Guide to Innovation.” Download now.

Taking the next steps

So, how do you get to this simplified, elegant, experience-oriented state? Here are a few suggestions to get you started: 

• Join internal IT user groups and join technical working groups outside your company. That way, you can begin to spot common patterns of frustration and complexities that need to be fixed, like too many low-information system alerts. 
• Speak with business leaders in your organization about the key results they want. Seek patterns among different groups, so people can start simplifying and unifying their data points. That has the added benefit of adding coherence to the reports senior management will read, since there are more common definitions throughout the organization.
• Work with your cloud service provider to see where you can take advantage of automation and scale, but remember that you still need to own key areas of your data, like user identity and corporate compliance, and the services you are using. 

The cloud represents a watershed change in how services are delivered and consumed by customers. The cloud combines and integrates diverse technologies in a manner that leads to greater efficiencies and innovations. As is the case with all technologies, it does not free us of our most important responsibilities; that is, delivering the best customer experience as possible. That includes the responsibility of elegance and simplification—never to an excess that hurts the corporate mission, but always in new ways that reflect a changing, dynamic, and modern technology organization. This is the Mitel of today.

Read this next: After an unprecedented 2020, IT leaders are trying to create new ways to operate and innovate. What are their biggest priorities in 2021? What are their biggest pain points? How do they measure up to yours? Learn from a survey of 2,000 Senior IT leaders conducted by IDG, in collaboration with Google Cloud. Get the report, “No turning back: How the pandemic reshaped digital business agendas.”

About the author

Rick Cirigliano is SVP of Global Cloud Operations at Mitel Networks.