The theoretical split is 32/1 but in practice it’s a maximum 30/1 and given the average current take up on OR FTTP is around 30% of the addresses covered , any individual is probably ‘sharing’ the available bandwidth with 9 or 10 others , obviously there will be variations, on a newsite that has no other network competitors it may well be 30 users , but conversely on a retro build area with competitors like Virgin and possibly an Alternative Network, someone may have an entire PON to themselves.
It’s not a useful way to consider bandwidth allocation anyway ….its not simply a case of 2.4Gb divided by 30 users therefore everyone gets around 80Mb each ….( or 1.2Gb divided by 30 for upload ) , usage is not linear, PONs have dynamic bandwidth allocation , higher speed profiles will alway be better off in absolute terms , if conditions were such that a 20% reduction was applied obviously 80% of 110Mb is more than 80% of 50Mb .
If congestion is suspected ( if someone was unlucky to be on a PON with couple of bandwidth hogs ) try a speed test at a time that would be expected to be ‘quiet’ , early morning ( 1am - 9am ) or late evening ( after 9pm ) if whatever issue exists during these times it’s probably not going to be congestion.
I recall there was a similar problem , if memory serves correctly, the issue turned out to be a problem on the BTw ‘backhaul’ , so many users were affected but only the one complainant , so the focus incorrectly but understandably was not on common parts of the network but on items specific to the one user that was complaining , the solution as it turns out was pretty simple , the interconnection connectors on the optical leads were cleaned ( dirty optics can impact performance severely )
The resolution supposedly also was to include a change in process/procedure so in cases or poor speed , these interconnection points to be considered as possible problems, even if it’s just a solitary user that was complaining,
TBH it is also the case an individual user won’t notice the difference between a 60Mb upload and a 110Mb upload in normal day to day use ( which is why in that case only one user complained, no one else even noticed ) only that particular user , that presumably ran speed tests continually, for no other reason than to ‘check’ they were indeed getting what they were paying for , was the only person that even noticed a problem, or was motivated enough to report it .
