VodafoneThree now expects £1 billion ($1.3 billion) in annual cost savings by fiscal 2032, up from earlier guidance of £700 million ($926 million) by 2030 — and it got there largely by abandoning open RAN.

The decision was not ideological. It was regulatory arithmetic. The UK's Competition and Markets Authority (CMA), when it approved the Vodafone-Three merger at the end of 2024, attached strict deadlines on site numbers and spectrum milestones. Open RAN's roadmap simply could not keep pace.

"The CMA have imposed very strict deadlines in terms of number of sites and in terms of spectrum solutions on those sites, by certain clear deadlines," Andrea Dona, VodafoneThree's chief network officer, told equity analysts at last week's investor update. "Unfortunately, the current-technology open RAN roadmap did not enable us to meet those deadlines. That's why we took the difficult decision to abandon open RAN for now in the UK network."

What did the operator give up?

Open RAN's standardized interfaces would have let VodafoneThree mix and match equipment from different suppliers at the same site. Pre-merger Vodafone had planned exactly that at about 2,500 sites served by Huawei, which a government order forces out of UK networks by the end of 2027. The plan was Samsung radios plus general-purpose IT kit from Dell and Intel.

Instead, the operator signed a £2 billion ($2.7 billion) contract with two incumbents. "We doubled down on two main suppliers with a £2 billion contract, Nokia and Ericsson," said Dona. "They developed specific radios that meet our spectrum holding. They weren't available — they're developing that specifically for us, something that unfortunately the open RAN vendors could not meet."

That is a poor advert for both the technology and Samsung, the vendor Vodafone Group still champions in Germany and elsewhere in Europe. Open RAN was meant to widen choice and cut costs. In the UK, at least, it did neither fast enough.

How far ahead is the build?

VodafoneThree claims it is running ahead of the CMA targets, though Dona declined to give site numbers, calling them "commercially sensitive." The concrete figure he did offer: the C-band network, using frequencies around 3.5GHz, now reaches 50 million people.

The operator's spectrum position reinforces that lead:

• 437MHz of total mobile spectrum, versus BT's 360MHz and Virgin Media O2's 326MHz
• About 200MHz held in C-band alone, more than either rival
• A combined 37,000 sites at merger completion, against BT's 20,000 and VMO2's 18,700

The footprint is shrinking by design. Iain Milligan, the operator's director of network development and infrastructure, said in April that rationalization would cut the count to between 26,000 and 28,000. Investor slides now show it settling at the lower end. Even so, the company expects 5G standalone — the version with a new core — to cover 99% of the population by 2030, up from 60% in September.

Where do the savings come from?

Network pruning is only part of it. By the end of 2028, VodafoneThree plans to decommission all of Three's legacy IT systems. On the commercial side, it will shut more than 200 overlapping retail stores and cut pre-merger costs there by 30%.

Staff have already felt it. Vodafone's latest annual report puts the average UK workforce at 12,416 for the fiscal year ending in March — 1,716 people, or 12%, fewer than Vodafone and Three employed together the previous year. CEO Max Taylor hinted the cuts could go deeper, citing AI-assisted code development tests that showed productivity gains "up to 40%" against the 10% assumed in plans to cover inflation. If that gap closes, headcount may follow it down.

Not everyone bought the story. Andrew Lee of Goldman Sachs noted that, stripping out synergies, the forecast implied EBITDA growth of just 4% in the UK's three-player market. "Are you doing enough on cost efficiencies?" he asked executives.

What about 6G and the upper 6GHz band?

Next year should be VodafoneThree's peak capex year, at roughly £1.4 billion ($1.9 billion). Equipment normally lasts seven to ten years, and a fresh 6G overhaul in the early 2030s would sit awkwardly with the company's pledge to push returns above its cost of capital by 2032. A software upgrade to 6G in today's 5G bands looks likelier, with 6GHz reserved for urban hotspots.

The band's fate sits with Ofcom, which is examining the upper 6GHz range. Dona was blunt: "I can categorically say that is valuable spectrum required for mobile and should be ringfenced for mobile." VodafoneThree plans to grow network capacity two and a half times between now and 2034, and it wants that spectrum in hand when the traffic arrives.