Analytics Charter

Opensignal’s analysis methodology for our Network Experience insights has been developed independently over several years and examined by many industry stakeholders.

We never change our methodology to suit the needs of a particular country or operator, to allow us to offer consistent and independent metrics globally. Our methodology evolves as we make improvements and those changes are applied consistently worldwide.

Our analytics are designed to ensure that each research participant equally contributes to results.

Our analysis ensures that we only include analysis from devices that display normal behavior, and not the behavior of test and engineering devices, research projects, or other programmed routines.

We always use confidence intervals in our analytical conclusions. They are clearly displayed in our charts, unless including them lessens readability.

Confidence intervals represent the upper and lower bound interval that a value will fall between. Using confidence intervals is standard scientific practice when reporting on sampled analysis.

If a difference between operators for a given metric is not statistically significant we declare it a draw and report each operator as a 'joint winner'.

Where confidence intervals are large we do not report the analysis at all to ensure that we only publish results in which we are confident.

To ensure that our analysis truly represents the experience of an operator’s own customers we classify measurements according to the customer facing brand they were generated from. This means MVNOs are treated separately from MNOs and consumer facing broadband providers are captured, not the wholesale networks on which the traffic is carried.

Measurements generated by national roamers are attributed to the user’s home operator as that is part of their network experience. International roaming traffic is excluded from publicly reported values.

We normally only report on standard time intervals and the most recent analysis; we do not “cherry-pick” an arbitrary time period in a report to support a predetermined conclusion. Where a different time window is being used this will always be clearly identified.