How many G.fast pods will Openreach need?
Openreach is gearing up for its push to get G.fast ultrafast broadband to some ten million premises by 2020 and a further 2 million with access to native FTTP all funded by their own money. There may be more FTTP and maybe some more G.fast delivered by the BDUK projects but that will very much depend on the goals set by projects and who gets the contracts for extension work beyond the existing Phase II contracts.
We have crunched the numbers and it looks possible that Openreach can deliver ten million premises that will count as premises passed with access to a minimum speed of 100 Mbps via G.fast by providing a pod on just 27,913 cabinets. This is based on analysis of postcodes within rough line length 400m of cabinets that are already providing VDSL2 and while the number of cabinets looks amazingly low the spread of lines across cabinets is not uniform in the Openreach network, in fact the 27,913 cabinets have an average of 449 lines connected with the distance rule dropping the number of premises from 12.5 million to 10 million. The 100 Mbps is based on our usual conservative approach and could be refined once the G.fast trials scale up.
It is possible that Openreach may prefer to pick cabinets another way, and that is to minimise the number who cannot get ultrafast speeds, and thus by cherry picking the cabinets with the highest proportion of premises within the 400m range you get a much higher number of cabinets at 38,197 but the number of not getting ultrafast drops from 20.1% to 10.5%.
Of course one does not know the precise way in which Openreach will determine the cabinets to roll-out G.fast to, but the insight of showing its possible with between 27,913 and 38,197 cabinets gives an important insight into the scale of the work involved.
One of the sticking points will be that with 96 port capacity per pod adding the pods to large cabinets will mean while many will count as passed if demand for ultrafast connectivity is somehow massively larger than demand for superfast speeds (i.e. highly unlikely looking at years of data on product choice from consumers) not every passed will be able to order. We did ask Openreach about the possibility of adding a second pod to a cabinet and the answer was potentially yes that could be done.
Once the next wave of processing is over we will publish a map showing the postcodes that fitted within the 27,913 model, of course appearing on the map does not in anyway guarantee G.fast arriving but everyone loves a map of the UK.
It may be that the changes to the G.fast standard that allowed for the longer reach have been instrumental in the change from the ambition of just 10 million ultrafast premises passed by Openreach to it being 10 million G.fast PLUS 2 million native FTTP. In other words since the ten million can be reached with less work and in a faster timescale combined with changes to working practices in the FTTP roll-out a more FTTP rich future is in prospect.
At the end of the day, the roll-out should take Openreach from just 1% ultrafast availability today to higher than 41% - how much higher depends on what is delivered for some BDUK Phase II and extension projects. This means that Virgin Media will be the dominant ultrafast broadband provider in the UK in 2020 with a footprint covering around 59% of UK premises.