Observations on the First Quarter Report and Conference Call
1) Clear areas of growth in Q1 despite modest total growth (4% at constant exchange rates)
- Optical Networks: 20% organic growth
- AI & Cloud: +49%
- Critical Enterprise and Defense Solutions: +19%
- Technology Licensing: +10%
2) Q1 orders on an upward trend
- €1 billion in AI & Cloud orders vs. €350 million in revenue (book-to-bill ratio ≈ 3x), indicating strong backlog building. This can be compared to the full-year 2025 AI & Cloud order estimate of €2.4 billion (averaging €600 million per quarter).
- Group book-to-bill ratio is >1, and clearly over 1 in the Network Infrastructure business.
- Hotard stated that lead times from order to delivery for optical products are typically 12–18 months (somewhat shorter in IP networks). This means a significant portion of recent orders is yet to be delivered, supporting revenue visibility extending partly into 2027.
3) Guidance and market expectations raised significantly
- NI growth: 6–8% → 12–14%
- Optical Networks + IP: 10–12% → 18–20%
- Hyperscaler investments (2026): $540bn → $700bn
- AI & Cloud market compound annual growth rate (CAGR 2025–2028): 16% → 27%
4) Additional investments in the San José factory
- Demand is growing so substantially that Nokia is increasing its investments beyond the planned levels.
- Justin Hotard: “I think one thing we may need to clarify if we haven’t clarified it before. Regarding Fab 2, when we shared that in November, we talked about Fab 2 being able to meet the requirements of the guidance we provided, and that there is additional capacity beyond that. We aren’t making any announcements about additional production capacity at this time, of course, but I would think of it this way: in the previous guidance, there was extra capacity and the ability to build. I would assume if you weave the conversation together, I’ll do it for you. We are making additional investments. That probably means some of what we’re doing there is investing in the deployment of Fab 2 at scale.”
5) San José InP factory progressing on schedule
- Production will begin in 2026, with more significant impacts visible after 2027. This is crucial for capacity and cost-efficiency. The new factory represents up to 20 times the capacity for InP (Indium Phosphide) components compared to the current Sunnyvale plant. It is also worth noting that the 2026 guidance depends very little on the new factory, as Hotard stated: “In reality, the second factory is only a fraction of the 2026 plan. It is much more relevant in the longer term.”
6) Infinera integration is ahead of schedule
- Both the realization of synergies and operational integration appear to be progressing smoothly.
7) IP Networks as a second growth engine going forward
- Design wins are expected to convert into orders starting from Q2, which could expand growth beyond optical networks. Hotard: “I would say our 18–20 percent optimism currently applies to both sides.” In other words, both optical and IP networks are expected to drive the combined 18–20 percent growth for those units.
8) Growth is mainly volume-driven, not price-driven
- Margin growth is likely to come later through economies of scale and lower unit costs (factory ramp-up), rather than through short-term pricing.
9) Wireless Networks (Radio Networks) remain a profitability challenge
- Revenue is stagnating and transparency is limited due to the reporting structure. Given the high margins of patent licensing (Technology Standards), its inclusion in the Mobile Infrastructure segment likely masks the weaker underlying profitability of the wireless unit. Light Reading has speculated that the Core Software and Radio Networks units combined could currently be loss-making.
SUMMARY
Q1 demonstrated the strength of the demand-driven optical cycle and increasing visibility of the order backlog. Current profitability is quite low, but visibility into 2027 and beyond has improved, at which point capacity (San José factory) and scale benefits are expected to turn results into more significant growth in optical and IP networks. The zero-growth wireless networks business is still far from acceptable profitability. Cost savings and the AI-RAN collaboration with Nvidia are intended to change this.
Regarding AI and Cloud growth, the following should be emphasized:
“We now expect our target market for AI and Cloud customers to grow by 27% annually (2025–2028), compared to the 16% we estimated in November.”
An annual growth rate of 27% means the AI & Cloud market will double in three years. Such rapid growth increases the likelihood that demand will exceed supply in at least parts of the market. In such a situation, customers cannot rely solely on their current suppliers but must expand their supplier base. Here, vertical integration (InP, in-house production capacity) transforms from a mere cost advantage into a strategic competitive advantage. This improves Nokia’s chances of entering customer accounts where it previously would not have had a realistic entry point.
Of course, even NI’s sales do not correlate one-to-one with the 27 percent growth for two reasons: 1) Fixed Networks do not participate in this (but Optical and IP networks do) and 2) the current customer base includes a lot of telecom sales that do not fall under AI & Cloud sales. For this reason, according to Hotard, NI’s target market grows by 14% per year while part of the market is stagnant and another part is red hot. Gradually, as the share of AI & Cloud sales increases, NI’s average target market growth rate will creep closer to the 27 percent level. As NI’s growth profile strengthens, its margins may also strengthen in a situation of strong demand.
At the beginning of the year, Nokia’s investment story was largely based on AI demand and related growth assumptions. After Q1, this is supported by measurable evidence: clear backlog growth, raised guidance, and a significantly strengthened outlook for target market growth.
!
Bottom line Realistic range 2028:
Simplified:
