Circles, a digital telco operator, has integrated OpenAI and Codex APIs into its product infrastructure. The published results are significant: +22% in ARPU Average Revenue Per User -9% churn. Additionally, the efficiency of internal development has measurably improved. This is one of the first public case studies where generative AI is systematically applied to retention and personalization in the telecom sector.
However, the value of this case goes beyond the telecommunications sector. Therefore, marketing managers of Italian SMEs and mid-market companies—both B2B and retail—can draw concrete operational insights. In particular, the AI-driven personalization logic applied by Circles is transferable to any context where behavioral segmentation and one-to-one communication are strategic levers. We at SHM Studio let's analyze the chronology, mechanisms, and practical implications for the Italian market.
Finally, this case raises a relevant question for those managing digital marketing budgets: is AI applied to retention still a competitive advantage, or is it becoming a minimum requirement? The answer, as we will see, depends heavily on the speed of adoption.
Circles: chronology of an AI-native integration
Circles is a digital telco operator founded in 2014, with a presence in Asia, the Middle East, and Europe. The business model is based on an approach asset-lightno proprietary network infrastructure, but a software layer that orchestrates the customer experience. This architecture has made AI integration relatively faster compared to traditional operators.
During 2025, Circles initiated a structured partnership with OpenAI. The goal was twofold. On one hand, to personalize communications and offers in real time. On the other, to accelerate internal development through Codex, OpenAI's code generation model. The results were published directly on the official OpenAI website.
Therefore, we are dealing with a documented and verifiable case. These are not projections or internal estimates. These are KPIs measured on a real operational basis.
The numbers that define the case: ARPU, churn, and development speed
Three metrics emerge clearly from the case study. First of all, the’ARPU grew by 22%. This data indicates that existing customers spend more, likely thanks to personalized offers and contextual upselling. In a low-differentiation industry like telco, a similar increase is considered exceptional.
Afterward, the Churn decreased by 9%. Retention has historically been the most expensive problem for telecom operators. Acquiring a new customer costs on average five to seven times more than keeping an existing one, according to consolidated research by Bain & Company. A reduction in 9% therefore has a direct and measurable financial impact on the operating margin.
Finally,’development efficiency has improved thanks to the use of Codex. Technical teams have reduced the implementation time for new features. This aspect is often underestimated in marketing analyses, but it is relevant: greater development speed means faster testing and optimization cycles.
Architecture of personalization: how the AI layer works
The model adopted by Circles follows a logic that digital marketing professionals will recognize. The AI analyzes the customer's behavioral data — plan usage, support contact history, browsing patterns — and generates personalized recommendations in real time. Furthermore, the system adapts the tone and content of communications based on the individual profile.
This approach stands out from traditional segmentation for one specific reason. Classic segmentation groups customers into static clusters. Conversely, an AI-native system continuously updates each user's profile. As a result, the offer a customer receives today may be different from the one they would receive tomorrow, if their behavior has changed.
To delve deeper into the logic of language models applied to marketing, it is useful to consult the analysis of Harvard Business Review on generative AI and personalization. The conceptual framework applies directly to the Circles case.
Winners, losers, and those at risk of falling behind
Who wins in this scenario? Circles, clearly. But also OpenAI, which consolidates a high-profile case study in an industry—telecom—historically resistant to rapid innovation. Furthermore, this case reinforces OpenAI's narrative as an enterprise partner, not just a provider of consumer tools.
Who is at risk of losing? Traditional telco operators that have not yet started structured AI integration paths. However, the risk does not only concern the telecommunications sector. Any company that competes on retention and personalization — insurance, utilities, subscription-based retail, B2B SaaS — finds itself in a similar position.
Therefore, the real loser is not a specific company. It is the marketing manager who interprets this case as industry news, rather than as a strategic signal applicable to their own context.
Reading SHM Studio: what changes for Italian marketing
We of SHM Studio We are closely following the evolution of AI applied to marketing and retention. The Circles case confirms a trend that we also observe in the Italian market: the companies that achieve concrete results from AI are not necessarily the largest ones, but those that have integrated artificial intelligence into their operational processes, rather than just in their communication tools.
Specifically, three elements of the Circles case are transferable to Italian SMEs and mid-market companies. First, real-time personalization does not require proprietary infrastructure. Second, the impact on retention is measurable in a relatively short time. Third, AI integration also accelerates internal processes, not just customer-facing ones.
Therefore, the operational question for an Italian marketing manager is not
Related articles
Discover other articles that explore similar topics in depth, selected to give you a more complete and stimulating view. Each piece of content is carefully chosen to enrich your experience.
