Two-Sided Ledger of AI: What Most CSPs Miss Using AI to Cut Costs and Less on Growth & Loyalty

Yet, the monthly churn hasn't moved, still running between 0.9% and 1.2% per month, and ARPU growth has encountered headwinds. AI investment has hardly touched these areas, but the opportunity is there. While current investments gatekeep costly calls, they don’t move the needle nearly as much with the customer’s experience, which is critical to the retention game.

The Two Applications of AI and Where the Market Stands

AI deployment in telecommunications followed a natural path to quick value and ROI, with the first wave going to call centers, back-office processing and network operations. These are quantifiable environments where AI-driven efficiency delivers measurable output in the likes of shorter resolution times, lower cost-per-interaction, better employee or support agent productivity. In many deployments, AI-powered assist tools have meaningfully improved the agent’s ability to solve issues and then even upsell. Customers even experienced faster resolutions.

While the work was rational and the savings were measurable, a second, less-explored application of AI has been largely ignored and treated as a future horizon: using AI to affect the customer’s experience and what they’re looking for from a service provider. In a market fighting the commoditization label, value creation becomes critical and that comes in the form of better experience – the likes of proactive engagement and personalization of care and offers.

That's where the bigger opportunity lives, but most of the market isn't there yet.

The Ceiling on Cost Cutting-Led AI and the Case for Revenue Growth/Protection

However effective AI has been at automation and gatekeeping, efficiency argument has limits. Gartner projects that by 2030, the cost per generative AI resolution will rise to levels approaching, and in some scenarios surpassing, the cost of offshore human agents. As infrastructure costs increase and the most accessible automation targets have already been captured, the returns from cost-focused AI deployment compress.

Meanwhile, the revenue math on churn prevention tells a different story. Monthly churn in the 0.9–1.2 percent range is the accepted industry norm, but the financial impact of moving that needle is significant. For a major U.S. provider, one basis point of churn can represent over $120 million or more in revenue that could be retained.

The NPS relationship, however indirect to revenue, reinforces the point: more value delivered to customers drives higher NPS, which drives lower churn, which drives higher lifetime value. When customers experience a technical issue around their device, such as connectivity or its performance, NPS drops by 19-points. For reference, in the largest CSPs, one point alone across an entire subscriber base is worth tens of millions of dollars. So while optimizing care for cost is a powerful defensive strategy, optimizing care for loyalty is an even stronger revenue retention and growth strategy, and the gap between those outcomes is not marginal.

What the Higher-Value Application Actually Looks Like

The reframe is based on recognizing that AI's highest ROI for CSPs lies in reducing the number of negative customer experiences that lead to churn and creating more of the positive ones that drive loyalty and commercial outcomes. Addressing this starts on the mobile device and with the underlying signals and data it creates on a regular basis. 

When device health signals are combined with network performance data and customer profile information, the result is a complete picture of the customer's actual experience. That connected data set enables CSPs to be proactive with care: identifying and resolving issues before the customer knows they have one. It also enables more timely, relevant commercial outreach up-sell and cross-sell offers that reach customers when their context makes them genuinely receptive, rather than on seasonal timelines or demographic signals that lack the hyper-segmentation to bring value and induce conversion. And when device data and the device experience is part of the churn model, prediction accuracy improves considerably, because the full experience is accounted for.

When CSPs proactively engage customers using mobile holistic data (network, CRM and device) to guide the service interaction, engagement and conversion rates rise. When tested with TELUS’s retail brand Mobile Klinik in 2024 and 2025, a proactive device care-related strategy yielded higher results than a reactive one: 4x higher digital brand engagement and 3x more acceptance of commercial offers.

When analyzing customer care more broadly, a proactive strategy fed by live device data, also allows an AI agent to handle more care scenarios autonomously. Rather than aid an agent with solving an issue, when device data is available, technical mobile-related issues can be solved without an agent present and only when absolutely necessary.

CSPs that reframe AI from a cost lever to a strategic asset with a duality impact will build the kind of customer relationships that competitors cannot effectively lower their prices to beat while also maintaining the cost savings from operations they still need.