Vodafone Adopts CUJO AI’s QoO Metrics for Multi-Market Broadband Rollout

CUJO AI said in an announcement distributed by PR Newswire on October 6, 2026, that Vodafone has selected its Quality of Outcome (QoO) network-measurement technology for deployment across several of the operator’s European markets. The technology is already deployed in a live Vodafone broadband network, with further rollout underway across Europe.

Rollout Across Vodafone’s European Markets

According to the company, the deployment brings Quality of Outcome to millions of Vodafone gateways (customer-premises equipment) while supporting the operator’s Low Latency DOCSIS initiative. Vodafone has already tested QoO with Low Latency DOCSIS technology to assess how network improvements translate into better performance for applications such as gaming, video conferencing, and cloud services.

“Network quality should ultimately be measured by what customers can successfully do with their connection,” said Remko Vos, CEO of CUJO AI. Vos said the company helped develop Quality of Outcome to connect network engineering with the experiences that matter to users, and called Vodafone’s deployment across multiple markets an important step toward making outcome-based network measurement an operational reality at scale.

CUJO AI said Vodafone has been a significant supporter and promoter of the framework, contributing an operator perspective to its development, and described the operator as the first major operator to deploy Quality of Outcome at scale across live broadband networks.

Oscar Gallego, Director of Home Technology & Products at Vodafone, said delivering an excellent broadband experience requires looking beyond headline speeds to understand how the network performs for the applications customers use every day. He described Quality of Outcome as “a practical, application-focused way to evaluate and improve that experience,” and said the two companies are bringing the capability into live networks across Vodafone’s European footprint.

How Quality of Outcome Scores a Network

According to CUJO AI, the framework evaluates how well network conditions support the outcomes users expect from specific applications, such as responsive gaming, uninterrupted video calls, and reliable cloud services, moving the assessment of a connection beyond technical measures like speed or average latency. It converts measurements including latency, packet loss, and throughput into an application-specific score, allowing operators to see whether their networks provide the conditions an application needs to perform well and to identify where performance can be improved.

The IETF document that defines the framework specifies that QoO scores express measured network conditions as a percentage on a linear scale bounded by two application-specific thresholds: 0% at the Conditions at the Point of Unacceptable Performance (CPUP) and 100% at the Requirements for Optimal Performance (ROP). Latency requirements are expressed as sets of percentile-latency tuples, packet-loss requirements as a mean ratio, and throughput as a single global minimum. The overall score is the minimum of the latency, packet-loss, and throughput component scores, based on the rationale that the most degraded dimension is likely to determine an application’s perceived quality; if the required minimum throughput is not met, the score is forced to 0%.

The document states that QoO is conceptually based on the Quality Attenuation metric and that its reliance on latency distributions and packet-loss probabilities preserves spatial composability, allowing operators to combine measurements across network segments or decompose them for fault isolation and root-cause analysis. It also records known weaknesses and open questions: in volatile networks such as mobile cellular links, the capacity available to a device can change by an order of magnitude within seconds, and slow sampling rates can miss short-lived latency spikes and produce overly optimistic scores.

The Specification’s Standing at the IETF

CUJO AI said Quality of Outcome was co-created by its experts in collaboration with the wider networking community and advanced through the Internet Engineering Task Force’s IP Performance Measurement Working Group. The company characterizes the result as a new industry standard published by the IETF.

The IETF’s record for the document, draft-ietf-ippm-qoo-11, lists it as an active Internet-Draft of that working group, authored by Bjørn Ivar Teigen Monclair and Magnus Olden of CUJO AI, with Ike Kunze as editor. The draft replaces the earlier draft-olden-ippm-qoo and carries an intended RFC status of Informational. The record shows the document has been submitted to the IESG for publication and sits in the RFC Editor queue awaiting a first editor; the record was last updated on August 13, 2026, with the latest revision dated May 19, 2026.

The draft’s implementation-status section records qoo-c, an MIT-licensed open-source C library for calculating Quality of Outcome maintained by CUJO AI and described there as a complete implementation of the specification, alongside partial QoO support in goresponsiveness, an open-source network quality test written in Go. Internet-Drafts are working documents of the IETF and valid for a maximum of six months; the current revision of the QoO draft expires on November 20, 2026.