Databricks at MWC 2026

Telecom: An industry at a crossroads

In a few days I’ll be heading to Mobile World Congress in Barcelona — an event I’ve attended many times across my 20 years in telecom and technology. Every year the themes evolve: smartphones, 4G, 5G, 6G, cloud, IOT, edge, AI. The technology changes quickly, but what hasn’t changed is the role this industry plays in our lives.

When we call a loved one, join a video meeting, navigate to a destination, stream a lesson, access healthcare, run a business, or send a simple message, we simply expect it to work. Behind that simplicity sits complexity: a global fabric of connectivity. When disasters happen — floods, fires, earthquakes — one of the first questions people ask is: Is the network up? Can we reach each other?

Connection is not just a convenience. It is reassurance. It is safety. It enables societies to function and people to thrive. And yet, the industry itself stands at a turning point.

Across conversations with telecom leaders around the world, I consistently hear the same challenge: revenues are under pressure, costs remain high, and personalization still feels out of reach. Churn remains elevated, growth is slow, cross-sell and upsell are limited, and margins continue to tighten.

The strategy is not the problem. Execution is.

Operators already have the customers, the network and the data, but many struggle to translate that into measurable growth, retention and operational efficiency. Decisions about where to invest in the network, which customers to prioritize, or how to optimize field operations often cannot be made quickly or confidently enough.

Why enterprise intelligence requires data intelligence

At the root of these challenges is a common issue: data is fragmented, slow to access, and difficult for teams to trust.

What leaders are now recognizing is that data strategy is business strategy. Improving financial performance requires becoming a data-driven organization, where decisions across marketing, network investment, customer care and fraud prevention are powered by reliable, real-time insight. Not incremental improvements, but a different operating model.

The shift is not about adding more dashboards. It is about changing how the company runs day-to-day. When governed, accurate data flows continuously, AI learns continuously, and the telecom itself becomes intelligent. This is the moment telecom moves from reacting to event to anticipating them.

Turning data intelligence into measurable telecom outcomes

The telecoms we see driving sustainable business growth are those whose business strategy is underpinned by data strategy, allowing them to move from simply selling products, to anticipating and solving needs in the moment. A few key examples include:

• Upsell and Next-Best-Offer (NBO) Decisioning: By bringing prospect and customer data onto a unified platform, operators can now determine the next best action in real time at every customer interaction. AI models analyze usage behavior, service eligibility, network performance, outage history and competitive signals to recommend the most relevant offer, whether that is a speed upgrade, backup connectivity or a streaming bundle. These insights are then delivered directly into customer-facing channels — digital assistants, call-center agents and retail staff — along with AI-generated guidance tailored to the individual customer and context. The impact has been significant: attach rates have improved by ~250% and ARPU has increased by approximately $8 per customer per month.
• Churn Reduction: Operators are also using conversational data — call transcripts and chat interactions — to better understand customer intent and risk. AI models trained on these interactions recommend retention actions during live conversations and help agents address the root cause of dissatisfaction, not just the symptom. Performance dashboards then provide visibility at both the individual and team level. The result is churn reductions of approximately 5 percentage points per month, alongside improved agent effectiveness and customer satisfaction.
• Proactive Fraud Prevention: Fraud has traditionally been handled after the damage is done; investigated manually, often slowly, and at significant cost. But with unified data and AI, operators are now shifting from reactive response to proactive protection. Modern AI can now identify anomalies in real time, extract fraud signals from unstructured interactions such as call and chat transcripts, and simulate future fraud patterns to stay ahead of bad actors. Operators can monitor hundreds of millions of subscribers simultaneously, significantly improving detection accuracy while reducing false positives. Investigations that once required manual effort are now automated, compute costs drop dramatically, and new protection models can be deployed in hours instead of days. Most importantly, fraud prevention shifts from post-incident response to continuous proactive defense, reducing fraud attempts by up to 80% and saving tens of millions of dollars annually.

What this means for telecom leaders

If I were leading in telecom again, I would anchor my thinking in a few key principles:

1. A business strategy without a data and AI strategy is no longer executable. Across the industry, the data and AI estate has evolved in fragments, resulting proprietary lock-in, siloed security policies, duplicated effort and, most importantly, slow decision-making; a significant problem in an industry where speed of decision-making is everything. Customers expect personalized service instantly. Networks must self-heal in real time. Fraud must be stopped before it occurs. Field technicians need answers while standing in a customer’s home. When data cannot move freely and safely across the organization, none of these are possible.
2. Data readiness is now a core operational capability, not a technical project. The leaders who are succeeding are building unified, governed data foundations: a single source of truth across network, customer, operational and partner data. Governance is embedded across all data and AI assets, enabling teams to trust and safely use information. Instead of locking data inside applications, they make it usable across the business, with speed and at scale.
3. To truly win with AI, operators must ensure culture keeps pace with evolving technology. With the rise of conversational intelligence and context-aware tools like AI/BI Genie, engineers, marketers, care agents and operations teams alike can now interact with data using natural language. But to use these tools effectively, enterprises must invest in training, enabling teams to work with data directly, and embedding AI into everyday workflows. AI must become part of how work gets done, not a separate initiative owned by a specialist team.
4. The goal should be to become an AI-enabled enterprise. In this model, data intelligence is embedded everywhere:
   • networks predict and resolve issues before customers notice
   • customer channels adapt and personalize in real time
   • operations optimize continuously
   • investments are guided by real usage patterns
   • savings are reinvested into innovation and growth

Ultimately, telecom operators will not be measured by coverage or speed alone. They will be measured by how intelligently they use their data to serve customers and industries. And that is where the next era of telecom growth will come from.

Our team would be delighted to connect at MWC 2026 to share real customer examples, introduce you to telecom operators who are already transforming their businesses, and demonstrate how these outcomes are being delivered in practice with Databricks.

If you plan to be onsite, book a meeting or come and see us at our executive meeting space in Hall 3, Stand 3A53PEx, our demo space at the Amdocs booth in Hall 3, Stand 3G10, or reach out directly via email or LinkedIn. You can also find us onstage throughout the week to hear from our data and AI experts and trusted partners at sessions like

• MWC-hosted panel: The Edge of Intelligence: AI, IoT, & 5G, ft. Nevash Pillay, Monday, March 2, 4:30-5:30p CET, Marconi Stage, Hall 6
• Wipro-hosted panel: Demystifying Agentic Enterprises: Hype Vs Reality ft. Nevash Pillay, Tues, March 3, 1-1:45p, Hall 2, Stand B30
• Tech Mahindra-hosted panel: Zero-Touch CX: When the Network Solves Issues Before the User Calls, ft. Nevash Pillay and Mark Austin, VP of Data Science, AT&T, Tuesday, March 3, 2:15-2:45p CET, Hall 2, Stand 2D46
• Microsoft-hosted panel: Unify Data & AI with Azure Databricks, ft. Dael Williamson, EMEA CTO, Wednesday, March 4, 12:40-1p CET

Beyond the event, we are also part of the TM Forum Data & AI Board, helping shape the blueprint for how telecom operators can successfully adopt data and AI at scale.

Want to learn more about what Databricks can do for Communications Service Providers? Download our ebook today.

See you at Mobile World Congress!