Machine Learning and Artificial Intelligence (AI) will change virtually every industry. Smart tools are penetrating the hardware and software used by organizations around the world, helping them to make better decisions and automate processes.
The intensely competitive telecoms sector is no exception. Operators manage large estates of network infrastructure and service teams; any efficiencies gained can have a huge impact on the bottom line.
These digital transformation megatrends can help operators run better, build smarter networks that keep customers happy, reduce costs and open up new market opportunities.
Today’s 24/7 connected world means customers expect uninterrupted service. Outages can have a serious impact on brand reputation.
Machine learning algorithms could be trained using historical and real time data to predict failures, such as power outages or transmission issues, before they happen. Symptoms such as unexpected packet loss could lead the model to issue an alert to maintenance teams who can sort the problem before customers notice anything at all.
Smart networks can also improve the customer experience. Self-optimising networks (SON) analyze call, text and data quality, and make remote adjustments using equipment such as tilting antennas. Meanwhile, data modelling can sense potential demand surges, such as a major sporting event, and recommend deploying additional network capacity.
AI can analyze customer relationships to spot early signs of discontent and take pre-emptive action to stop attrition or to launch highly personalized retention campaigns.
AI-powered chatbots are a simple, cost-effective way of handling common queries. They can also redirect more complex issues to a call center.
While many operators have outsourced their contact centers, some have repatriated these services to bring agents closer to customers. Potential increased costs can be offset by services that analyze data, such as customer history, credit scores and social media behaviour, to match people to the most-suited agent, boosting satisfaction and reducing call times.
Quality of data
Machine learning models are only as good as the data that trains them. Operators have treasure troves of customer and network information to draw on, but using traditional ‘network-centric’ metrics such as sites, capacity and other internal network KPIs have their limitations. If the goal is to improve customer experience, then the machine learning model should be trained with a ‘customer-centric’ perspective. The only way to do that is to base it on actual customer experience data.
Network performance data may suggest, for example, that coverage in an area is optimal. However, it can’t see how walls, buildings and other users in the same location impact experience, so it’s missing the true customer-centric picture. That’s where analyzing billions of smartphone measurements collected from actual users, as OpenSignal does, can help close the data gap for operators and give them an edge in building AI that maximally impacts customer experience.
The real competitive advantage for operators using AI comes not from creating a better machine learning algorithm, but in using the most relevant and extensive data to train the AI engine. Operators will never be Google DeepMind, but they do have access to unique and comprehensive datasets. Those who focus on unlocking the value in that data they have access to, and on closing customer-centric data gaps will be the real winners in the age of AI.