Telecommunications: An Industry at a Crossroads
In a few days I will go to the Mobile World Congress in Barcelona – an event that I have participated in many times during my 20 years in telecommunications and technology. The themes evolve every year: Smartphones, 4G, 5G, 6G, Cloud, IoT, Edge, AI. Technology changes rapidly, 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 text, access healthcare, run a business, or send a simple message, we simply expect it to work. Behind that simplicity lies complexity: a global web of connectivity. When disasters strike – floods, fires, earthquakes – one of the first questions people ask is: Is the network on? Can we reach each other?
Connection is not just a convenience. This is assurance. This is security. It enables societies to function and people to flourish. And yet, the industry itself stands at a turning point.
In conversations with telecom leaders around the world, I consistently hear the same challenges: revenues are under pressure, costs remain high, and personalization still seems out of reach. Churn remains high, growth is slow, cross-sells and upsells are limited, and margins continue to tighten.
Strategy is not the problem. Is execution.
Operators already have the customers, networks and data, but many operators struggle to translate this 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.
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.
leaders are now recognizing this Data strategy is business strategy. Improving financial performance requires becoming a data-driven organization, where decisions about marketing, network investment, customer service and fraud prevention are driven by reliable, real-time insights. Not incremental improvements, but a different operating model.
This change is not about adding more dashboards. It’s about changing the way a company is run day-to-day. When controlled, accurate data flows continuously, AI continuously learns, and telecommunications itself becomes intelligent. This is the moment when telecommunications moves from reacting to events to anticipating them.
Transforming data intelligence into measurable results
The telecom companies we see driving sustainable business growth are those whose business strategy is based on a data strategy that allows them to move from simply selling a product, to anticipating and solving in-the-moment needs. Some prominent examples include:
- Making Upsell and Next-Best-Offer (NBO) Decisions: By bringing prospect and customer data onto a unified platform, operators can now determine the next best action in real-time on each customer interaction. AI models analyze usage behavior, service eligibility, network performance, outage history and competitive signals to recommend the most relevant offer, whether it’s a speed upgrade, backup connectivity or a streaming bundle. These insights are then delivered directly to customer-facing channels – digital assistants, call-centre agents and retail staff – as well as AI-generated guidance tailored to the individual customer and context. The impact has been significant: Attach rates improved by ~250% and ARPU increased by approximately $8 per customer per month.
- Reduction in churn: Operators are also using conversation 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 detect the root cause of dissatisfaction, not just symptoms. Performance dashboards then provide visibility at both the individual and team levels. the result is Reduction of about 5 percentage points per monthWith improved agent effectiveness and customer satisfaction.
- Proactive Fraud Prevention: Fraud is traditionally dealt with after the damage has been done; Investigations were carried out manually, often slowly, and at significant cost. But with integrated data and AI, operators are now shifting from reactive response to proactive security. Modern AI can now identify anomalies in real-time, extract fraud signals from unstructured interactions like call and chat transcripts, and simulate future fraud patterns to stay ahead of bad actors. Operators can monitor millions of customers simultaneously, significantly improving identification accuracy while reducing false positives. Checks that once required manual effort are now automated, computation costs have dropped dramatically, and new security models can be deployed in hours rather than days. Most importantly, fraud prevention shifts from post-incident response to ongoing proactive prevention, Reducing fraud attempts by up to 80% and saving millions of dollars annually.
Opinion: What it means for telecom leaders
If I were a leader in telecommunications again, I would base my thinking on a few key principles:
- A business strategy can no longer be executed without a data and AI strategy. Across the industry, data and AI assets have been developed piecemeal, resulting in proprietary lock-ins, arcane security policies, duplicated efforts and, most importantly, slow decision-making; A significant problem in an industry where speed of decision making is everything. Customers expect instant personalized service. The network must heal itself in real time. Fraud must be stopped before it occurs. Field technicians get the answers they need while standing in the customer’s home. None of this is possible when data can’t move freely and securely across an organization.
- Data preparation is now a core operational capability, not a technical project. Leaders who are succeeding are building integrated, governed data foundations: a single source of truth across network, customer, operational and partner data. Governance is built into all data and AI assets, enabling teams to trust and securely use information. Instead of locking data inside applications, they make it usable across the entire business at speed and scale.
- To truly win with AI, operators must ensure that the 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, enable teams to work directly with data, and embed AI into everyday workflows. AI should become part of how work is done, not a separate initiative owned by a specialist team.
- The goal should be to become an AI-enabled enterprise. In this model, data intelligence is embedded everywhere:
- Anticipate and resolve problems before network customers notice them
- Customer channels are optimized and personalized in real time
- Operations are constantly optimized
- Investments are guided by actual usage patterns
- Savings are reinvested in 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 this is where the next era of telecommunication development will come.
Our team would be delighted to join you at MWC 2026 to share real customer examples, introduce you to telecom operators who are already transforming their businesses, and demonstrate how these results are being delivered in practice with Databricks.
If you plan to be onsite, book a meeting Or come and meet us in our executive meeting space in Hall 3, Stand 3A53PEX, our demo space at the Amdocs booth in Hall 3, Stand 3G10, or contact us directly via email or LinkedIn. You can also find us on stage throughout the week to hear from our data and AI experts and trusted partners in sessions
- Panel hosted by MWC: The Age of Intelligence: AI, IoT, and 5G, Ft. Nevash Puppies, Monday, March 2, 4:30-5:30p CET, Marconi Stage, Hall 6
- Panel hosted by Wipro: The Revelation of Agentic Enterprises: Hype vs. Reality ft. Newash Puppies, Tuesday, March 3, 1-1:45p, Hall 2, Stand B30
- Panel hosted by Tech Mahindra: Zero-Touch CX: When the Network Resolves Issues Before a User Call, ft. Nevash Pillay and Mark Austin, Vice President of Data Science, AT&T, Tuesday, March 3, 2:15-2:45p CET, Hall 2, Stand 2D46
- Panel hosted by Microsoft: Integrate Data and AI with Azure Databricks, ft. Dale Williamson, EMEA CTO, Wednesday, March 4, 12:40-1p CET
Apart from the event, we are also part of the TM Forum Data & AI Board, which is helping telecom operators shape the blueprint for successful adoption of data and AI at large scale.
Do you want to learn more about what Databricks can do for communications service providers? Download our eBook today.
See you at Mobile World Congress!
