The good thing is there is a template to follow for a successful network merger in the UK, although some technical differences in mobile networks now compared to 2012 may present a few new challenges along the way, but I don't think we will see something wildly off what we had previously with EE from a user point of view.
Likely a new brand will launch (not a marketing expert so not going to guess what this will be called, how it will be priced or pitched). But from a technical point of view the combined organisation will probably issue SIMs homed to Three PLMN 23420), while those SIMs can join VF 23415.
Just like when Everything Everywhere was launched, for a while customers will probably be able to stay on Three or VF, and take out contracts with either, but I suspect they will be strongly encouraged to move to the new brand using some pricing strategy or network restrictions.
For the short term, this will provide users on the new brand with a ridiculous amount of capacity and coverage and will likely be the main selling point initially, but don't get too excited...
Overtime ex-VFs capacity will be shifted to Threes RAN (23420), while ex-Three sites will gradually be connected to VFs RedStream core and take advantage of SA in the process.
On the RAN side, the first change (possibly pretty soon into the merger) I expect what we will see is ex-VFs small allocation of B3 shut down on ex-VF sites, and reallocated to all ex-Three sites (PoWs, rooftops, rural towers, everything)…giving Threes 15MHz 4G spectrum an immediate boost to 20MHz straight away, alleviating congestion in areas where Three never got around to deploying 5G. The second thing we may see see is Threes RAN 23420 broadcasting ex-VFs band7 in dense urban areas like some parts of London where Three had not deployed 5G yet, but i'm less sure on this one. All this can be done quite early without shutting down sites or building new sites, and using existing 4G equipment on old sites from either operator.
Then we get to the urban areas where both (or either network) have invested in recent years, and things get a bit more interesting, as we will start to see sites shut down and/or relocated. Urban areas outside the 12 or so bigger cities, like ones mostly made up of unilateral poles from both operators currently, we will not be observing much, or any, coverage or capacity boost from the merger (depends on area) long term; as both networks tend to provide pretty decent indoor coverage as of now. Often the unilateral poles from VF and Three are metres apart, or just a bit further along the same stretch or road; so it would be uneconomical to keep two sites next to each other (2x backhaul costs, 2x power costs etc).
Overall, which sites remain and go might depend on the city, I can say from looking around the home counties, it seems more likely the VF unwind poles will be going, as it seems like the unwind project ground to a halt 3 years ago, whilst Three have still been busy building and activating phase8 +9 poles through 2024. With a little bit of capacity shifting again, many uni poles will be removed and relocated. But this is good news for the smaller urban areas that never got either. Expect a case-by-case for each site, but I suspect will mostly made up from Three unilateral poles as there are simply way more of these across the country than VF unwind poles (most of Threes urban coverage is made up form new capability, whereas VFs is mostly made up from older sites shared with o2). There may be some fringe cases where either network right now has a gap to fill so it’s obvious these sites are staying for now.
Rural areas gets a bit uncertain, as this will entirely depend on how the combined organisation plans to work with CTIL and MBNL going forward - Either continue working with both (VF have some commitments), or build out new physical infra. But I can see many reasons for this to lean on Threes RAN yet again, whilst adding ex-VF capacity to it, one reason is Three have a clear strategy with b28 in rural areas (from a variety of sites, including TV masts), and for those lucky enough to experience this have had good things to say, often even exceeding VFs band8 only rural coverage. So we will probably continue to see 23420 expand b28 rural availability using a cash injection from the merger, whilst adding ex-VF capacity to it where technically possible (by this we may even see ex-Three 23420 on CTIL sites, and we may actually see VFs b32 deployed beyond Newbury and on some MBNL sites!).
Overall, long term, from a technical point of view, it's going to be a mix across the country depending on where Three and VF have been investing each in the past few years, and where there is good overlap, we will see many sites taken offline and relocated over time. We will likely see Three primarily continue their RAN strategy, 140MHz 5G urban + b28 rural, with a bit of help from VF adding to this capacity* and using a few really well located VF sites in some areas. Whilst we will likely see VF play a bigger part in the network core. *Some spectrum might need to be handed back to ofcom as part of the merger.
As far as I am aware no technical details have been proposed as part of the merger yet so nothing is confirmed and is just a prediction based on my observations of the Three and VF networks over the past 4 years.