Quality on Demand and network slicing APIs: The differentiated connectivity opportunity

Download Listen

As operators look for new ways to monetise network capabilities, Quality on Demand (QoD) and Network Slicing APIs stand out as two of the most important emerging approaches to differentiated connectivity. The challenge now is turning trials and early deployments into scalable commercial offers.

Network APIs are an emerging area of telecom innovation that could unlock significant value by changing how enterprises and developers interact with telecom services and capabilities. Particularly promising are APIs that enable differentiated connectivity, allowing applications to programmatically request specific network performance characteristics such as enhanced bandwidth, lower latency or prioritised traffic. This makes connectivity more flexible and easier to integrate directly into digital services.

STL Partners estimates that the global network API monetisation market could reach USD 31.5 billion by 2030. Within this broader opportunity, network configuration APIs, such as Quality on Demand (QoD) and network slicing APIs, represent a particularly strategic and high-value growth area as they unlock a route for operators to monetise differentiated connectivity. In our forecast, QoD alone represents the third largest revenue opportunity by API by 2030.

For operators, these APIs offer an opportunity to transform the network from a largely commoditised best-efforts connectivity utility to a higher-performance, customisable and monetisable platform. In doing so, operators can create a productisable layer between standard public connectivity and more bespoke solutions such as private networks, enabling differentiated connectivity to be delivered at scale.

Developers benefit by gaining more direct control over network performance. Applications can dynamically request capabilities such as lower latency, higher bandwidth or prioritised traffic, enabling new types of services that depend on predictable connectivity. For example, a cloud gaming platform could request lower latency for a user during a game session to ensure smoother gameplay.

Enterprises, meanwhile, can access enhanced connectivity without the complexity of bespoke network agreements. Instead of negotiating custom SLAs and long procurement processes, enterprises can request improved network performance on demand. For instance, a broadcaster covering a live event could temporarily request higher bandwidth and prioritised connectivity to ensure a reliable video uplink during the broadcast.

Enabling differentiated connectivity: three key network APIs

Differentiated connectivity is not delivered through a single technical mechanism and is a broader concept than network APIs alone. It refers to connectivity services that offer performance beyond best-effort public mobile connectivity, whether through private networks, managed services or capabilities exposed through APIs. In this article, we focus on three API-related approaches that are particularly relevant to differentiated connectivity in public networks: QoS provisioning, Quality on Demand (QoD) and network slicing. These sit on a spectrum in terms of how dynamic they are, how strongly performance can be guaranteed, and how much network resource isolation is required.

At one end of the spectrum is quality of service (QoS) provisioning. This allows developers to assign predefined QoS profiles to a device, application or traffic flow, so that traffic associated with a given service is prioritised according to its assigned level. In practice, this is a relatively static and long-lived policy. While this approach can support differentiated treatment of traffic, it typically lacks the ability to dynamically adapt to the needs of individual sessions or applications.

At the other end lies dynamic network slicing. This builds on QoS principles but goes further by creating logically isolated slices of the public network with guaranteed performance characteristics, such as bandwidth, latency and reliability. Unlike static QoS policies, network slices can be instantiated and scaled dynamically based on demand. Emerging network slice booking APIs are enabling enterprises to reserve, configure and manage these resources in advance, specifying parameters such as geographic coverage, duration and performance requirements.

Between these two APIs, sits Quality on Demand. QoD operates as an API layer on top of the public network, enabling applications to request temporary improvements in network performance when needed. Through traffic prioritisation and network orchestration, a QoD API can dynamically adjust characteristics such as latency or bandwidth for specific sessions or traffic flows. However, unlike slicing, QoD does not create a logically isolated network environment. Instead, it provides performance enhancements within the shared network, meaning that improvements still depend on overall public network conditions and available capacity. This makes QoD more flexible and easier to deploy, but also means it offers weaker guarantees than full network slicing.

QoD has seen the most momentum so far, but evidence of scale remains concentrated in China

Among these APIs, QoD has seen the greatest level of activity so far, particularly in the form of pilots, trials and early commercial deployments. According to the GSMA Open Gateway tracker, seven operators have deployed QoD APIs globally. Operators including Deutsche Telekom and Telefónica, including Vivo in Brazil, have been among the most active in developing and testing these capabilities. However, most initiatives remain at an early stage and far from broad commercial scale.

That said, early signs of larger-scale commercialisation are starting to appear in China. China Telecom and China Unicom have both launched GSMA-certified Open Gateway deployments and are already seeing initial traction. China Unicom, for example, states that its Open Gateway platform now reaches around 70% of the country’s connected vehicle market. The platform is reported to have handled over 8.3 billion automotive API calls, with annual revenue for these services reported to generate over $250 million as of Q3 2025. The operator is leveraging APIs like QoD and dedicated network access and profile APIs to support private 5G deployments, delivering tangible business impact across sectors such as automotive and industrial services.

QoD is particularly suited to use cases that require flexible, session-based performance improvements rather than permanent guaranteed connectivity. These include applications that need temporary improvements in latency or bandwidth, while still balancing overall network resource use and cost. QoD is also well suited to mobility scenarios, where application performance must be maintained as devices move across cell sites, and to wide-area deployments where guaranteeing high performance everywhere would be difficult or uneconomic. Several operators are currently exploring QoD through targeted trials, such as:

1. In Germany, Deutsche Telekom, O2 Telefónica and Vodafone jointly tested QoD APIs with Siemens Energy and Microsoft for remote industrial maintenance. Field technicians used Microsoft HoloLens 2 and the Remote Assist application to connect with remote experts while repairing equipment. QoD enabled the application to request higher network performance when the AR session begins, ensuring stable video streaming and data exchange even in challenging connectivity environments.

2. Telefónica, together with VR tech company YBVR and AWS Wavelength, demonstrated QoD for immersive sports viewing during the Copa del Rey ACB basketball tournament. Using VR headsets or mobile devices, fans could watch matches through 180° and 360° camera views. QoD dynamically prioritised network traffic to support high-resolution immersive streaming and maintain smooth video delivery in crowded stadium environments.

3. Elisa, working with Nokia and Elmo Cars, demonstrated QoD for remote vehicle operation (also showcased at MWC 2026). In this case, a driver controlled a vehicle remotely using multiple live video feeds and control signals. QoD prioritised the network connection, even during periods of high network traffic, to maintain low latency and stable video transmission, both of which are critical for safe tele-driving.

What is holding differentiated connectivity APIs back?

Despite significant industry attention since around 2023, differentiated connectivity APIs remain at an early stage of development and deployment. Technical progress is being made through initiatives such as GSMA Open Gateway and CAMARA, but the more persistent barriers are commercial. In particular, operators face two interrelated challenges:

1. Justifying large capex investments in 5G standalone (5G SA). Differentiated connectivity services such as QoD and network slicing depend on a more programmable network architecture, typically based on 5G standalone. Yet deployment remains limited. According to the latest release of our Telco Cloud Tracker, as of January 2026 only 22% of operators that have launched 5G had deployed either converged or fully standalone 5G. The hesitation is partly technical, but it also reflects concerns over return on investment. Even where 5G SA is in place, delivering differentiated connectivity at scale requires additional capabilities such as orchestration, automation across the network, service assurance, and API exposure platforms.

2. Identifying the right, scalable commercial model. Even where a strong business case exists, converting the value of differentiated connectivity into a scalable commercial offering is not straightforward. The commercial model is still evolving, as we explore in our upcoming report on Monetising Network Performance: Commercial models from QoD to slicing. It remains unclear who should pay for these capabilities, whether developers, enterprises, or end users, and on how they should be priced. QoD, for example, could be charged per API call, session, or traffic flow, each of which raises different challenges. At the same time, the value of differentiated connectivity varies widely depending on both the use case and network conditions. For instance, guaranteed low-latency connection may be critical for industrial automation or remote control of machinery, but far less valuable for video streaming applications for consumers. This wide variation in both network cost and customer value makes it difficult to design simple, scalable pricing models for differentiated connectivity.

The road ahead for differentiated connectivity

Rather than focusing on QoD or slicing as separate technical features, operators should concentrate on the customer outcome: delivering the right level of connectivity, on demand, with as little complexity as possible. Most customers are unlikely to care about the underlying network technology. What matters is whether their application receives the level of performance it needs, when it needs it. For this reason, it may be more effective for operators to position these capabilities under a single concept, such as “differentiated connectivity,” rather than marketing multiple technical solutions.

The more immediate challenge for operators is turning promising trials into repeatable commercial offers. One route is to build on successful pilots in specific use cases, such as remote support, immersive media and tele-operated driving, and translate these into propositions that can be sold more widely to similar customers. Another is to prioritise use cases where improved network performance clearly adds value, but where the service is not fully mission-critical (such as cloud gaming or enterprise video calling), offering a more practical route to building confidence, refining the offer and developing viable commercial frameworks before scaling further.