CUJO AI Network Intelligence
CUJO AI Network Intelligence is a service for broadband and mobile operators that analyzes traffic application by application, scores the Quality of Outcome actually delivered, and runs automated root-cause analysis instead of stopping at a speed test.
What is CUJO AI Network Intelligence?
Network Intelligence is one of the products in the CUJO AI catalog, and it is sold to broadband and mobile operators rather than to subscribers. It detects and analyzes traffic in order to recognize Internet use categories or individual applications, and evaluates the Quality of Outcome (QoO) the operator is actually delivering. The premise CUJO AI states plainly is that customers care about applications and experiences, not about networks or latency — so a measurement that stops at throughput measures the wrong thing.
QoO is presented as an open framework supported by the IETF and built on a Quality Attenuation metric, which turns technical network measurements into quality scores tied to the application concerned. CUJO AI sets it against the two familiar approaches: QoS, whose dry mathematical metrics correlate poorly with what a person experiences, and QoE surveys of the NPS kind, which it calls subjective and unreliable. What QoO is meant to produce instead is a diagnosis the customer can understand.
On top of that measurement the product runs an automated root-cause analysis and produces proof-of-enhancement metrics. Weightings are applied across more than a dozen factors, from the access network through to the core and the Wi-Fi segments, and real-time metrics are combined with application intelligence to quantify how much a given action would improve the outcome — before that action is taken. The result is a real-time, granular view of how subscribers use the network; video conferencing is the worked example on the product page, illustrated by a Sankey diagram that breaks an imperfect session down factor by factor.
CUJO AI groups the beneficiaries into three families: user-facing uses, network operations, and sales and marketing. The QoO technology came in with Domos, described as a pioneer in application outcome analytics, whose acquisition was announced on 9 April 2025 and whose framework is combined with CUJO AI's device intelligence.
CUJO AI says the product sits on a single cloud-native platform that also carries its cybersecurity, device intelligence and personalized experiences services, and that one agent on the customer premises equipment serves all of them; a containerized application for RDK-B equipment was demonstrated in June 2026, deployable with a single command and independent of gateway firmware and of OEM or operator update cycles. The publisher, Cujo LLC, is based in California and reports offices in eight countries.
What it does
- Measure the Quality of Outcome delivered application by application, instead of throughput alone
- Run an automated root-cause analysis on a degraded session
- Quantify the expected gain of an intervention before carrying it out, through proof-of-enhancement metrics
- Pinpoint which network segment — access, core or Wi-Fi — is weighing on an application's performance
- Give subscribers quality metrics they can understand, together with self-troubleshooting recommendations
- Equip customer service teams with real-time network context
- Prioritize infrastructure investment where incidents are most likely
When to use CUJO AI Network Intelligence / When not to
A quick filter to help you decide if CUJO AI Network Intelligence is the right fit.
When to use CUJO AI Network Intelligence
- Broadband and mobile operators that want to measure perceived quality application by application rather than raw throughput
- Network operations teams and NOC analysts looking for automated root-cause analysis and a segment-by-segment view of the network
- Telco customer care and technical support teams that need real-time network context to shorten handling times and avoid truck rolls
- Network planning and engineering teams that have to arbitrate infrastructure investments
- Operator marketing and sales teams segmenting subscribers and looking for upsell opportunities in usage data
When not to use CUJO AI Network Intelligence
- Individuals and households: the product is sold to the operator, never to the subscriber
- SMBs and non-telecom companies wanting to monitor their own internal network
- Teams that expect to test or buy online: there is no public price, no trial and no self-service sign-up
- Developers looking for a publicly documented API: none is accessible
- Organizations that require data to be hosted in a named country: the site commits only to data centers in the United States and the EU, with no list of countries
How to use CUJO AI Network Intelligence
A typical end-to-end flow, from setup to results.
- Start from the Network Intelligence product page, which is the only public description of the service
- Accept the commercial reality first: there is no online sign-up, no self-service trial and no published price — this is a sale to operators
- Use either call-to-action button on the product page: both lead to the CUJO AI contact page, and sales@cujo.com is the published commercial address
- Expect the technical detail to come out of that conversation: no public integration guide, API page or documentation is accessible, developer.cujo.com being locked behind Microsoft Entra ID authentication
- Scope the deployment on your CPE fleet: CUJO AI says a single agent on the customer premises equipment carries all of its services and supports every major CPE architecture
- Consider the containerized RDK-B variant CUJO AI demonstrated in June 2026, which it describes as a one-command deployment with no dependency on gateway firmware
- Plan how the metrics will reach your teams and your subscribers: there is no CUJO AI mobile app or browser extension for this product, and operators surface the data in their own applications and portals
Pros & Cons
Pros
- The measurement framework is not proprietary: Quality of Outcome is backed by a standards body, the IETF
- It goes beyond the speed test by measuring the outcome for the application, which is what the end customer actually notices
- Proof of enhancement quantifies what an action would gain before that action is taken
- A single CPE agent carries the whole set of CUJO AI services, which limits vendor complexity
- The containerized RDK-B deployment is claimed to be independent of firmware release cycles
- The security posture is documented publicly: SOC 2 Type II, ISO 27001, encryption at rest, TLS 1.3, annual penetration testing and a bug bounty
- Sub-processors are named in the privacy policy: Amazon Web Services, Microsoft, Atlassian, Slack and WPEngine
Cons
- No published price at all: no pricing page, no tier, no range — everything goes through the sales team
- No trial and no free plan, and no self-service sign-up
- No technical or API documentation accessible publicly: developer.cujo.com exists but is locked behind Microsoft Entra ID authentication
- The three consumer percentages on the product page (86%, 96% and 93%) are published with no source, no sample size and no methodology
- No demonstration page: the only entry point is the sales contact
- The terms of service on the site date from 5 August 2023 and describe the historical consumer product — hardware box, mobile app, subscription — rather than the current operator offering
- No data-location commitment finer than data centers in the United States and the EU, no DPA published or mentioned, and a privacy policy that says nothing about whether customer data feeds model training
Pricing & Plans
No price is published for Network Intelligence. A full inventory of the site's published pages and of its six sitemaps contains no pricing page, and no amount, currency or billing unit is stated anywhere for this product. No free plan and no trial are announced either. The commercial model is direct sales to operators: both calls to action on the product page lead to the CUJO AI contact page, and terms are therefore negotiated with the vendor.
Data, GDPR & hosting
A consolidated view of how CUJO AI Network Intelligence handles your data.
GDPR overview
The privacy policy in force since 25 May 2026 covers CUJO LLC and its subsidiaries. A dedicated section for EEA and UK residents lists eight rights — information, access, rectification, erasure, restriction, portability, objection, and rights over automated decision-making and profiling — plus the right to complain to a supervisory authority, with contacts given through the EDPB site and the ICO for the UK. CUJO has certified to the U.S. Department of Commerce under the EU-U.S. Data Privacy Framework and its UK Extension, and declares itself subject to FTC investigation and enforcement. It commits to cooperating with the European authorities' panel and the ICO on unresolved complaints, names the AAA-ICDR as a free recourse body, and offers binding arbitration under DPF Annex I. A Data Privacy Officer answers at privacy@cujo.com. No Article 27 representative is designated.
Who owns the data?
The privacy policy separates the Customer, the organization buying the services, from the Consumer, the end subscriber whose devices are covered. For Consumers' personal data, CUJO LLC acts as a data processor and states that it is not responsible for its Customers' privacy practices: to learn how their data is used, a subscriber must ask their operator. The data described includes contact details, network and device identifiers (IPv4/IPv6, MAC addresses, IoT device identifiers, platform-generated CUJO Agent IDs, advertising identifiers), Internet activity information and support call recordings. The policy, effective 25 May 2026, states that no sensitive data — racial or ethnic origin, medical, financial or wellbeing data — is collected.
Reuse rights
The privacy policy effective 25 May 2026 states that CUJO LLC processes personal data to deliver its Services and Digital Resources; for internal purposes such as accounting, audit, IT security, process improvement and research and development of new services; for legal, safety and security purposes; for marketing; and to handle job applications. The legal bases claimed are legitimate interest, contract, legal obligation, consent and, for the security strand, public interest; for Consumers' network and device identifiers and Internet activity, the stated basis is performance of the contract or pre-contractual steps under Article 6(1)(b) GDPR. Objecting to marketing is explicitly offered. Data may be disclosed to authorities on lawful request, including for national security requirements, and may be de-identified, anonymized or aggregated and then shared for any legally permitted purpose. Two silences deserve to be noted rather than interpreted. Training AI models does not appear among the declared purposes, and the policy neither authorizes nor rules out the use of customer data for that. And nothing is published on what an operator may in turn do with the data it obtains through the service: CUJO LLC only says that it is not responsible for its Customers' practices, which places that question in a contract the site does not disclose.
Data retention & training
Hosting summary
The privacy policy states that the virtual servers powering the company's IT infrastructure sit in vendor-controlled secure data centers in the United States and the EU, and that personal data may be stored in any one of them. No EU country is named, no finer region is specified, and no choice of region is offered to the customer — a latitude the document states openly. Section 10 of the policy names the sub-processors involved: Amazon Web Services, Inc for the cloud compute and storage platform behind the CUJO AI services; Microsoft Corporation for Azure access management, file hosting, SharePoint and O365 applications; Atlassian Pty Ltd for the engineering wiki and ticket tracking; Slack Technologies LLC for internal and Customer communication; and WPEngine, Inc for the WordPress CMS behind cujo.com. Transfers from the EU and the UK to the United States rely on certification under the EU-U.S. Data Privacy Framework and its UK Extension. Technically, the company describes encryption at rest with keys specific to each environment — service-managed or customer-managed — TLS 1.3 in transit, and key material held in hardware security modules. Policy effective 25 May 2026.
Things to keep in mind
Risks and trade-offs to weigh before adopting CUJO AI Network Intelligence.
- The headline figures CUJO AI puts forward — 63 million homes, 3.8 billion devices, 72,000 device models — describe the footprint of the CUJO AI platform across its operator customers; they measure neither Network Intelligence nor its performance
- The three consumer percentages on the product page (86%, 96% and 93%) carry no source, no date, no methodology and no sample size
- The terms of service published online date from 5 August 2023 and describe a consumer product — hardware box, application, subscription — that is not this offering: current commercial conditions cannot be inferred from them
- The terms set a minimum age of 18 while the privacy policy states that the sites and services are not directed to anyone under 16: the site contradicts itself
- The RDK, prpl, IETF, Camara and Broadband Forum logos are presented as partnerships and memberships; only the RDK-B integration is documented by a press release
- The publisher does not publish a verifiable incorporation date, and the earliest web archive capture of the domain belongs to an unrelated earlier occupant of the name rather than to CUJO AI
- No representative under Article 27 of the GDPR is designated, even though the product handles data on European subscribers
Setup & Integrations
Technical difficulty
High, and confined to operators: this is an integration project, not an installation. An agent has to be rolled out across a CPE fleet and the platform connected to operations and customer systems. CUJO AI claims simple deployment and low vendor complexity — one agent for all its services, support for every major CPE architecture, and a containerized RDK-B build demonstrated in June 2026 that deploys with one command. Against that, no integration documentation is public (developer.cujo.com sits behind Microsoft Entra ID), and nothing is published on timelines, effort or network prerequisites. Evaluation necessarily starts with a commercial discussion.
Integrations
Behind CUJO AI Network Intelligence
Fundraising
Social
Resources
All the official URLs gathered for verification and reference.
Frequently asked questions
What is CUJO AI Network Intelligence?
How does Quality of Outcome differ from QoE and QoS?
Who is the product sold to?
How much does it cost?
Is there a free trial or a free plan?
Is there an API?
How is the product deployed at an operator?
Where is the data hosted, and are the sub-processors disclosed?
Which certifications does the publisher claim?
How do you get in touch?
Should you pick CUJO AI Network Intelligence?
CUJO AI Network Intelligence is a specialist product with a clear audience: broadband and mobile operators, not general users and not businesses looking to monitor their own network. What sets it apart is the measurement itself. Instead of another throughput dashboard, it scores the Quality of Outcome delivered to each application using a framework supported by the IETF, then adds proof-of-enhancement metrics that put a number on what a given fix would actually gain before anyone commits to it. That combination — an open measurement framework plus a quantified reason to act — is the strongest argument the product makes.
The technology arrived with the acquisition of Domos, announced on 9 April 2025, and is combined with CUJO AI's device intelligence. CUJO AI says the product rides on a single cloud-native platform and a CPE agent shared with its other services, which it presents as a way to limit vendor complexity for the operator. Its published security and compliance posture is detailed and easy to check: SOC 2 Type II, ISO 27001, EU-U.S. Data Privacy Framework certification and named sub-processors.
The reservations are all about what is not published. There is no price, no accessible technical documentation, no demo and no trial, so evaluating the product means opening a sales conversation first. The consumer percentages on the product page carry no source or methodology, and the terms of service on the site describe a different, consumer-facing offering altogether. None of that makes the product weak; it makes it impossible to assess from the outside. If you run a network and the case for measuring outcomes rather than speeds resonates, it is worth the call — with a list of questions about pricing, integration effort and data handling ready.
- Choosing a selection results in a full page refresh.
- Opens in a new window.