Why does QoE differ so much from Subjective User Quality – SUQ?

The quality of a network proves its potential in the interaction with the ‘real’ user.  For users, the network that they are working in, represents a means to perform a task or solve a problem. Any kind of network quality is perceived in such a functional context, which includes rational and emotional elements. ‘Real’ quality, quality in a real-world-context, may differ largely from quality that is technically measurable (like in QoE), due to the situational influence on user perception. To carriers, this issue is know as the ‘red-green’ problem. All technical network indicators are reported to be ‘green’ – working at best performance – but the user perception of quality is ‘red’, negative.

Synthetic monitoring, generating an QoE that indicates how users ‘feel’ the quality of a network,  is widely used  as a means to improve network quality in a usage scenario. Artificial monitoring is based on simulating transactions in a specific networks environment. This solution periodically sends a request across the network and measures the antiphon time of the network across the whole organisation.


As a quality measure, QoE represents a true improvement over purely technical standards. However it does not solve the problem of remoteness to the user perceived world. This is where  Metrinomics’ SUQ (Subjective  User Quality) comes in to close the gap.

Subjective User Quality consequently takes the real user view to measure the subjective/perceived quality of services, analyses the success drivers and generates action and communication triggers in real time. To the contrary of a synthetic indicator, SUQ is able to consider success criteria as part of the reference in an ongoing process.

SUQ  elements:

  1. quality-success logic
    connect information, use adaptive math to explain customer loyalty and business, success
  2. data collector
    implement feedback sources to gather subjective quality measures along with technical indicators
  3. data integrator
    shape input data in order to become part of one consistent model landscape and connect output data to further applications
  4. predictive action motor
    anticipate user requirements and generate action chains, nutured by user demand to adapt network performance to the actual level of necessities
  5. real time dashboarding
    provide live data, customer feedback, action triggers and to-do lists to individual professionals