New Zealand’s largest mobile carrier, One NZ, adheres to its central value—customers. To give nearly 1,000 users a real-time, tailored view of customer data, it upgraded to Microsoft Fabric Real-Time Analytics (RTA) within less than two weeks. The new platform elevated One NZ’s customer service and added depth to its analytics, transforming raw data into actionable insights. Empowered teams can now make accurate, data-driven decisions and respond to customer inquiries nearly twice as fast as before.
One NZ, a New Zealand-owned telecom giant with nearly 5,000 employees, serves 2.4 million customers, commanding 38% of the mobile market. Despite its vast reach, the company remains steadfast in its focus on sustainability to preserve the astounding New Zealand nature. "We partner with REMOBILE and Brightpoint to refurbish old mobile devices and recycle 95% of our network waste," says Strathan Campbell, Channel Environment Technology Lead at One NZ. It also unites customers in charitable causes through the One Good Kiwi app, donating a portion of NZD100,000 each month.
But no matter the initiative, customer experience is always central to One NZ’s endeavors, and technology is a magic wand that helps enhance them. "Good service is crucial," emphasizes Chris Fletcher, Small and Medium Enterprise (SME) and Consumer Director at One NZ. “In a market of 5 million people, growth opportunities are limited. So, we aim to reduce friction for our customers by using data and information to improve their experience," Strathan explains.
An express upgrade from raw data to meaningful insights
For some time, the company relied on Microsoft Power BI to monitor call center performance and visualize data for service agents across New Zealand, India, and the Philippines. But as Strathan explains, "The system fulfilled our needs for a while. However, our increasing data volumes started leading to delayed refresh rates in what should have been real-time Power BI dashboards. We needed a solution that would help us take our data-driven performance to the next level.”
Building a new solution from scratch required multiple product licenses and setting up the necessary security architecture within One NZ’s existing infrastructure. What it needed was a seamless solution that would be easy to implement and manage with a small internal team. The solution was Microsoft Fabric Real-Time Analytics (RTA). “What drew us to it was that it was an all-in-one solution. Since we didn’t need to buy new components and were already embedded with Power BI, putting the architecture and security in place was quick and easy,” shares Steven Easton, BI Channels Specialist at One NZ.
In fact, the transition was so seamless that it happened in record time according to Strathan, “We went from a concept to a delivered product in just two weeks. That’s what Fabric has empowered us to do. We can add all the services we need with just a little bit of code.”
The rapid shift came with its lesson. "A pleasant surprise was that we underestimated the user uptake, which strained our Power BI capacity,” shares Steven. “More people were using it because it ran much faster." The high usage was a good problem, which Strathan and Steven were happy to solve, stress-testing the system in real-world conditions and becoming a business user with the most queries in Microsoft Fabric worldwide. “We received valuable advice and assistance from Microsoft. We only had to make one code change, and now, the system is robust and stable,” Steven says.
The winning setup for real-time response to data
One NZ uses Fabric event streams (custom source) within Fabric RTA to seamlessly ingest data from Amazon Connect and Kinesis data streams into Microsoft Fabric. Analytics on top of real-time streaming data is achieved using the KQL Database. KQL (Kusto Query Language) replaced SQL as a more intuitive and accessible alternative to analyze real-time data streams. It allows every analyst to write queries and combine analyses with dimensional data quickly and easily. As the last step, visualizations are delivered to end-users through Power BI, available via desktop PCs and mobiles.
Altogether, this revamped setup has allowed One NZ to evolve from tracking call center metrics to generating actionable insights. "Metrics such as the longest waiting contact, answer rates, and handle times have always been tracked. The emphasis now is on using data to create actions," Steven elaborates.
One NZ's data now informs immediate actions—reallocating resources or preemptively solving customer issues. "We've created three reports—agent, queue, and readerboard—that are all individually customizable to suit the users' specific needs. Currently, 170 business users and about 800 contact center agents use these reports,” explains Steven. “Most dashboards are updated every 10 seconds now, which is six times faster than before.”
“Most dashboards are updated every 10 seconds now, which is six times faster than before.”
Steven Easton, BI Channels Specialist, One NZ
Equal data access: a catalyst for customer centricity
Now, if a customer waits for 30 seconds, the operations team can immediately allocate an agent to that call. “Our previous target to answer an incoming call was 45 seconds. Now we're at 27 seconds, a significant improvement from our customers’ perspective,” shares Strathan. This granularity and real-time nature of data have equipped operational teams to react swiftly to customer needs, centralizing customer experience as a key performance indicator for the entire organization.
However, the benefits aren't confined to queue improvements. Before, only specialized teams at One NZ had this kind of access. Fabric has equipped everyone in the organization to leverage data. “Previously, you needed to be a data engineer or scientist to access and understand customer information. Now we're making it user-friendly, so anyone can easily make data-driven decisions," Strathan says.
Using the new analytics capabilities, One NZ has transformed its customer data into a valuable product, implementing proactive customer engagement strategies. As Strathan points out, “This data is now consumed by our marketing and product teams, who incorporate it into their machine learning models. This way, we can identify customer behavior patterns, such as likelihood to churn or preference for certain products."
Expanding the data magic
Encouraged by its initial successes, One NZ has charted a strategic roadmap for leveraging Microsoft Fabric further. “We've just finished building out a large language model that identifies customer intents to add more personalized insights,” Strathan confirms. Over time, One NZ aims to empower its customers with even more self-service capabilities. "Last year, we had 16 million customer contacts. Of those, 27% required an agent to resolve, amounting to 95 years' worth of agent talk time," explains Strathan. "Our goal is to minimize this and reduce the effort for agents and customers."
Looking ahead, One NZ aims to roll out a Data Activator feature to automate responses based on the collected data. “This will be especially powerful when combined with historical data, enabling us to create a comprehensive solution for customer engagement,” adds Strathan.
With these future expansions, One NZ looks beyond collecting data to empowering its entire organization to act on it. "Democratizing data is crucial for us," asserts Chris. "It allows us to apply machine learning models to predict issues or upsell opportunities. And we can do all this without intruding, benefiting both the customer and the organization."
“Previously, you needed to be a data engineer or scientist to access and understand customer information. Now we're making it user-friendly, so anyone can easily make data-driven decisions.”
Strathan Campbell, Channel Environment Technology Lead, One NZ
Follow Microsoft