November 25, 2021
Telecoms.com periodically invites expert third parties to share their views on the industry’s most pressing issues. In this piece Suresh Chintada, CTO at Subex, explains why enterprise AI and augmented analytics are important technological trends for telecoms.
Powering all manner of person-to-person, machine-to-machine, and machine-to-person communications, the telecommunication industry is undergoing deep shifts. Competition is stiff; trends like deregulation and consolidation makes it vital for operators to differentiate themselves, cost effectively.
Similar to other industries, artificial intelligence (AI) has been making waves in telecom too. The promise? Rapid differentiation through cost optimisation, service agility, and new business models. These are an expected given for any telecom enterprise aiming to succeed in today’s digital, competitive, and disruptive marketplace.
Gartner predicts that by 2024, 75 per cent of enterprises will shift from piloting to operationalising AI. This amounts to a five-fold increase in streaming data and analytics infrastructures.
There are plenty of success stories of AI within telecom, already. Those who have piloted AI implementations testify to its ability to improve decision-making, profitability, and operational efficiency. It is not hard to imagine a future with AI playing a key role in functions such as proactive customer service, timely fraud management, holistic business assurance, optimised network quality, better asset management, and 360-degree partner management, to name a few.
Despite its potential, though, operators remain cautious. The entire telecom industry plays on managing margins and so operators are wary of the high Capex costs of adopting AI. Ironically, the ability to respond quickly to market changes will propel profitability – and this ability is exactly what AI and augmented analytics provide.
Previously, companies had the advantage of business intelligence tools. But in a post pandemic world, responsiveness and resilience are key traits that separate the leaders from the laggards. Achieving this begins with democratising access to data and leveraging AI in a cross-functional manner.
Build to scale – enterprise AI for telecom
Innovation in operations as well as business models is limitless within the telecom industry. A study by McKinsey Global Institute finds that the high-tech and telecom industries are forerunners when it comes to implementing AI solutions. Many adopters are also reporting revenue increases thanks to AI.
In telecom, the buzz is tangible: operators are piloting machine learning models to test efficacy and measure value. The measurable success levers are capex minimisation, revenue maximisation, and network optimisation. Some examples include:
Customer experience – AI can drive customer experience analytics, provide personalised recommendations, and improve campaign management.
Network – It can spot anomalies based on usage patterns, allowing communication service providers to remedy disruption through timely maintenance or asset re-harvesting.
Revenue – AI models can help operators tap into lucrative business opportunities that leverage existing investments.
Looking ahead, telecom operators must shift their focus from simply implementing AI to scaling it across the enterprise. AI drives a fundamental internal shift, enabling telcos to stave off capex-heavy investments while delivering ongoing value that translates to direct, tangible benefits. Being all-pervasive, it can easily extend its capabilities across telecom enterprises delivering supreme efficiency, smarter insights, continuous improvements, and new business opportunities.
Barriers to enterprise AI
Holistic AI deployment, while certainly an enterprise goal, is typically hamstrung by operational constraints, rigid mindsets, and manual processes.
AI is inherently complex making the right technological expertise imperative for successful deployment. For example, AI encompasses numerous machine learning, deep learning, and computer vision models. Business users would be clueless as to which model to choose for their specific business problem.
Being esoteric, AI models also call for niche skillsets. The seemingly simple task of data preparation mandates knowledge of activities like exploratory data analysis, data transformation, missing value treatment, normalisation, encoding, etc. Thus, dearth of talent is an ever-present challenge. Availability of data scientists and data engineers is sparse, influenced heavily by cost and expertise.
Altering human mindset and organisational culture is a prerogative. Many companies still depend on manual data science processes, which affect productivity. A 2020 report by Anaconda, data science leader and distributor of Python and R programming languages, found that nearly 45 per cent of a data scientist’s time is spent simply on getting the data ready for models and visualisations.
Building user trust is another primary concern. The granular aspects of black box models usually remain unknown to the actual users of AI. Inherent model complexity means that a telco’s business teams do not understand the logic or how the algorithm arrives at the final result. In processes that call for decision-making, this knowledge is paramount.
Transparency supports enterprise AI
Today, companies are increasingly demanding that AI models be transparent, explainable, and accountable. According to Forbes, explainable AI is about understanding how a model comes up with certain results. It is also about understanding how decisions are made by models and how models correct their own errors. Without some manner of ‘explainability’, the propensity for change is rigid and adoption idles.
Models must be accurate if they are to gain user trust. Hence, model bias is an issue. AI-based lending models providing recommendations that are skewed towards customers from a particular region, gender, or race is an example of model bias. Fixing these issues involves lengthy experimentation. Parallelly, nearly 50 per cent of initial experiments fail, calling for consistent readjustments to the model.
The trick here is to fail fast and move iteratively – and rapidly – to the next prototype. But to experiment and fail fast, organisations should also be able to accelerate how they choose, build, deploy, and test models.
Without a single, comprehensive, and proven platform to perform the above activities, telecom enterprises rely on disparate systems even when implementing AI, which pose integration challenges, compromise the user experience, and limit the potential of enterprise AI.
Augmented analytics – democratise AI and empower citizen data scientists
Self-serve augmented analytics platforms help telecom enterprises democratise AI in a simple, user-friendly, and automated manner. They help operators experiment fast using a no-code set-up that offers self-service capabilities in a single, one-stop solution. Some of the primary ways in which augmented analytics support enterprise AI are as follows:
Experiment iteratively for accurate models – There are several tools embedded within augmented analytics platforms that take away some of the burdensome tasks of data scientists. For example, Auto-ML (machine learning) supports ‘last mile optimisation’ and is based on actual findings allowing models to course-correct and deliver results with increasing accuracy. Here too, the process is iterative and automated. The Auto-CASH (Combined Algorithm Selection and Hyperparameter Optimisation) modules help select the best model and the best set of hyperparameters to optimise the chosen evaluation metrics (accuracy, precision, lift, etc.). These tools greatly increase the productivity of users.
Accelerate data lifecycles through automation – Augmented analytics provides a governing framework with workflows to manage data effectively. Tasks such as data preparation, tuning the hyperparameters, selecting the best-fit model, deploying it into production, and monitoring its performance are automated end-to-end.
Expose model logic through explainable AI – Augmented analytics platforms support interpretability of black box models. Users can easily understand the logic behind predictions across global, regional, and local explainability levels. Machine learning de-biasing is an added feature to combat model bias. It assists business users in trusting the model’s outcomes, especially for outputs related to fraud, customer service, business assurance, revenue leakage, and so on. A robust augmented analytics platform also helps monitor key evaluation and performance metrics like precision, recall, feature drift, model drift, etc.
Benefits of enterprise AI
A topmost benefit is that augmented analytics platforms empower business users to become citizen data scientists. For one, the intelligent automation of data management improves efficiency and productivity of existing data scientists. Secondly, it helps users easily leverage AI to solve business problems without having to depend on exhaustive training and domain knowledge.
While these provide a massive boost to the quality of work, there are also clearly quantifiable benefits.
Companies that have adopted augmented analytics platforms report a 50 per cent increase in analytics efficiency and decision-making confidence. Automated feature synthesis helps data scientists roll out more accurate models, iteratively, quickly, and without user bias. Insights are available quickly as data processes run up to 100 times faster. Crisp visualisation of these insights and patterns is a bonus. Finally, conversational analytics make it even easier for business users to consume these insights.
Moving to the bottom line, augmented analytics amplifies revenue in two ways:
Revenue through efficiency gains – Telecom operators can expect operational profitability to increase by 23 per cent. Employees become more productive, feeling assisted in their work; retention increases by 31 per cent as does the newfound scope for value-adding tasks. This includes nurturing citizen data scientists who can then build AI models for other functions, amplifying value – and return on investment – across the enterprise. Some companies report increasing their citizen data scientist pool by nearly five times thanks to augmented analytics platforms.
Revenue from customer delight – On the front-end, customers enjoying increased personalisation, faster issue resolution, and network quality (among others) report greater satisfaction. Some adopters report 35 per cent year-on-year increase in customer acquisition.
Conclusion: Accelerate data-to-decisions
Expectations from AI are soaring. Augmented analytics is the crucial differentiator that will separate those who win big through AI investments and those who lag. Augmented analytics platforms can help players accelerate the data-to-decision lifecycle giving them a sharper edge. In telecom specifically, they balance the investment-vs-benefit debate by optimising existing processes and unearthing new business opportunities, thereby minimising costs and maximising revenue.
Suresh is the CTO of Subex, and brings with him a wide ranging leadership, managerial and technical experience of over 27 years. Prior to Subex, he has worked with companies like Motorola, ARRIS and CommScope, where he built and scaled large global software engineering, professional services and technical support services operations, serving Industry verticals like cable, telecom, mobile and wireless networking. Suresh holds a Bachelor’s and Master’s Degree in Electronics & Communications Engineering from Osmania University and Post Graduate Diploma in software enterprise management, from IIM, Bangalore.
Read more about:Discussion
About the Author(s)
You May Also Like