Why Egypt's broadband speed figures disagree with each other

Broadband speed claims for Egypt vary widely depending on who is publishing them. The variation is mostly definitional rather than contradictory.

Three different measurements

Official figures are often averages of subscribed package speeds across the customer base. Crowdsourced indices such as Ookla's publish medians of tests actually run by users on their own equipment. Advertised speeds are contractual maximums. These three numbers can differ by a factor of two or more while all being accurate about what they measure.

Averages hide the tail

An average is pulled upward by a small number of very fast connections. A median describes the typical household better. When a figure moves sharply upward after a package upgrade programme, that reflects what people are subscribed to, not necessarily what their line delivers at 9pm.

Self-selection in crowdsourced data

Users who run speed tests are not a random sample. People test when something feels wrong, and they test on the devices they have, over Wi-Fi, in the room they happen to be in. That biases results downward relative to a wired measurement at the router.

What would settle it

Regulator-run measurement using fixed probes in sampled households, published with methodology, is the only approach that produces comparable numbers over time. Where such data is published for Egypt, Cairo Stream will report it in preference to either official averages or crowdsourced indices.

Practical guidance

For a household judging its own service, the useful test is wired, at the router, repeated at peak and off-peak hours. Anything else measures the home network as much as the connection.