Where to next for the augmented towerco business model?

As well as sheer scale in the European towerco landscape, the Cellnex name is synonymous with innovation in the towerco space. More akin to a netco in its DNA, Cellnex’s finger is firmly on the pulse of emerging trends in the communications industry, embedding itself in an innovation network that continues to push the boundaries of what the industry can achieve. In this interview, TowerXchange spoke to Cellnex’s Chief Commercial Officer, Oscar Pallarols to understand how new service lines are featuring today and what opportunities will arise in the coming years. 

 

 

In our past financial year, we generated €2.5bn in revenues, with €2.2bn (88%) coming from telecoms infrastructure services (TIS). Whist the vast majority of this TIS revenue comes from co-locating equipment on towers, other passive infrastructure services are also grouped under this umbrella – areas such as Fibre-to-the-Tower (FTTT), small cells, DAS and small datacentres.

 

Of the €2.5bn FY21 revenues, €200mn came from our broadcast business in Spain where we manage a single broadcast network (with no plans to expand our broadcast business to other markets), and €100mn revenue was generated from network services such as the provision of tetra networks.

 

Whilst we have diversified and expanded our portfolio of services, revenues outside our traditional telecoms infrastructure services are still marginal.

 

 

 

Fibre-to-the-tower (FTTT) is the fastest growing area, driven by the rollout of 5G. 5G promises low latency but in order to deliver this, it is important for the backhaul not to present a bottleneck. Connecting sites with fibre avoids this bottleneck arising. Currently about 25-30% of Europe’s sites are fibre connected and this is expected to reach 70-75% over the next 2-4 years as 5G is rolled out (the remaining 25-30% of sites are unlikely to require a fibre connection due to low population density, in rural areas for example).

 

There has also been growth in demand and revenues from indoor coverage solutions, with this being more marked in some countries than others. When speaking about indoor coverage, there are generally three key stages to rollout in a given country. The first stage covers the top venues, major stadiums and concert venues, areas of high digital density with links to content and entertainment. There are generally around 10-20 of such venues in each country, and for the most part, these have all been covered across Europe. Such solutions have been deployed both by neutral hosts and by MNOs that have been willing to open up their single DAS network to multiple operators. 

 

The second stage covers public buildings with high footfall, buildings such as hospitals, transport halls and shopping malls. There are a number of verticals and there is also some variation in who pays for such installations, sometimes it is the MNOs, sometimes it is the venue owners. The rollout of such systems is still ongoing in most European countries, with some markets more advanced than others. In terms of Cellnex’ footprint, Italy and the UK are both quite mature, whereas countries such as France and the Netherlands are in the much earlier stages development.

 

The third stage of in-building rollout covers hundreds of thousands of premises, for example regular business offices. Such locations generally require cheaper, less complex solutions but there is often less willingness amongst the building owners or MNOs to pay for the coverage. The UK has seen great progress in bringing coverage to such buildings thanks to the introduction of the Joint Operator Technical Specifications (JOTS) which enables neutral hosts to deploy a shared solution into which the four MNOs can plug. In addition to the UK, across Cellnex’ footprint both Spain and Italy are also relatively advanced, with more work to be done in other markets.

 

Outside of FTTT and in-building coverage, a lot of growth had been forecast in outdoor small cells, but this is yet to materialise. Thanks to the dense population, street furniture concessions being awarded by boroughs and ways in which the MNOs have been combining macro and small cells to use spectrum efficiently, London has seen more small cells rollout than anywhere else in Europe. We expect other packets of outdoor small cells rollout to pick up, but not to the levels initially forecast.

 

The reason for the dampened rollout is linked to the limited ARPU increase observed from 5G rollout. With a low uplift in ARPU, MNOs aim to first squeeze as much as they can out of the macro network before investing in small cells. COVID has also had a negative impact on outdoor small cell rollout with reduced footfall in urban centres.

 

 

 

Cellnex, by nature is an industrial group. What this means is that we are not just focussed on financials, we are extremely active in listening to and responding to what is going on in the market.

 

When it comes to developing new service lines, we very much follow traditional innovation methodologies. It is, however, the execution, the decision criteria, the management of risk, our partners and teams and our business approach which are much more crucial and where we really pay attention and try to excel. 

 

We have an innovation and product strategy team at group level, but there is very much a bidirectional flow of information from our country operations to and from group level. We export and import knowledge between operations which keeps the process lean and flexible – everything a country is exploring can be scaled up at group level.

 

About 80% of our collective brain activity is focussed on an agreed arena, areas which align with market trends, our vision and new technologies – with a committee deciding upon this scope. When it comes to market trends, as Europe has not been a leader in the telecommunications space for some years, we look to countries such as the US, Japan, Korea and China to see what is happening and what could be translated into the European market. When it comes to our vision, Cellnex does not see itself as a towerco, rather something closer to a netco – this expands our areas of focus that we can cover. Plus, with technologies, we keep a close eye on what is being developed and is in the pipeline globally. 

 

Once we have set this 80% arena in which we want to hunt, we put together teams to start to understand technologies and the key players in the space. We put together a non-commercial project, with an open ecosystem involving a customer and skilled or specialist partners to validate the idea, usually setting the project within a given context.  For example, we work together with a 5km racing circuit in Barcelona (Parcmotor Castellolí) equipped with a 5G network and edge computing to explore V2X opportunities. This testing lab is being used by the automotive industry, much reputed brands are using the circuit to get data from connected vehicles using Cellnex technologies.  Another example is a lab we have set up in Barcelona to test OpenRAN with an MNO in a city context. 

 

Once we have validated the idea from a technology and service perspective, we then add the business component on top. Initially we forget about the commercial element but following the initial stage we start to look at things such as how is the service designed, how we would invoice for the service (e.g. a flat rate or by usage), what is the operational model, what are the legal restrictions, etcetera. 

 

Following this we then look for our first customer and oftentimes the customer will be the same company with whom we did the pilot. After this the goal would be to scale to other customers and countries. Whilst a key KPI is to be approved by one company, in the innovation space many projects do not fit this, and for every successful project, there is a long tail or projects which have not made it that far.  Reasons for projects not progressing forward include the technology not working well, the scope falling too far outside our strategy, the idea not stacking up commercially, and a lack of demand for the service. With innovation, such “failure” is an accepted part of the process. 

 

It is also important to note that when it comes to innovation, Cellnex are just a piece of an innovation network. We are able to do only a very few things in isolation, and thus it is important to have strong links with academic institutions, government, vendors, MNOs, start-ups and other stakeholders. We take a cooperative, open approach to innovation. 

 

  

 

Being pragmatic, there are two different things to look at. On the one hand you have the reality of what is happening today whereby network providers are co-locating equipment on towers. This includes not only the 3GPP commercial networks, so the MNOs who are our biggest customers, but also a new set of connectivity providers. Amongst this new set of connectivity providers, you have fixed wireless access (FWA) players (from whom we are seeing a lot of demand at present), IoT players (who whilst requiring much smaller configurations and thus paying smaller lease fees, are becoming more numerous especially in dense urban areas), plus dedicated networks such as air traffic control and maritime services (for whom a pan-European portfolio that can provide solutions in multiple markets is particularly attractive).

 

On the other hand, there are less mature potential uses of the tower, areas which could present new revenue streams but for now are very much in their infancy. Some ideas that fall into this category include high resolution sophisticated GPS, thermal cameras for collecting data on the climate or potential wildfires, pollution monitoring, and potentially nests for drones to take off and recharge. Such services could co-exist with the co-location of network providers on towers.  

 

 

The movement into private networks was a very easy decision for Cellnex. We were used to providing mission critical networks in the public safety space, and the technical requirements bare a strong similarity to those in the provision of private networks. Both are dedicated networks with their own spectrum and strict SLAs, the difference is rather the customer, either a public body or a relevant enterprise. Our decision to move into private networks meant only an extension of our customer base.

 

As such, when we acquired the private network player EDZCOM, it was less about acquiring the technical capabilities to operate private networks and more about leveraging their experience in the enterprise space. In such a space, every sector or vertical is different, and comes with their own set of challenges. Companies do not decide that they want a private 5G network because of connectivity issues, rather they have a specific business challenge or set of challenges which they need to address, for which a 5G private network could offer a solution.

 

Cellnex has chosen to focus on private networks for heavy industries, ones where the installation is large and complex (for example utilities, mines, ports and logistics). This is where we feel we can add the most value and differentiate ourselves from others in the market. 

 

Our preferred business model is for Cellnex to make the investment and manage the network, charging a service fee to the enterprise customer. In this scenario, Cellnex designs the system in partnership with a technology vendor and the customer, takes on the investment cycle as the network moves from 4G to 5G, 6G and so on, and operates the network.

 

Whilst some of our competitors prefer private networks where there is the potential to act as a neutral host and offer, for example, a public network using the private network infrastructure, in the industries in which we operate the client generally wants no interaction with other networks. As such, in 99.9% of cases we serve just a single customer. With the mission-critical nature of their operations, the clients tend to have very strict requirements that everything must be on premises, closed and protected. Due to the nature of the industries, any interference could have drastic consequences. 

 

When it comes to the spectrum required for private networks, Cellnex is not willing to acquire licensed spectrum at all – that is a red line that we will not cross. So how do we address the spectrum issue? In some countries there are already shared spectrum policies which are being established by the regulator, mostly because of some industry demand. Where such policies exist means the customer (a port, a manufacturing plant, a mine, etc.)  can apply to access such frequencies in specific locations. Commercially speaking, however, it is always our preferred option to work with an MNO that is active in this space.

 

 

Traditionally, datacentres have been large buildings built in a centralised location. For example, in Spain most of the country’s large datacentres are in Madrid – in Madrid there are around 20 datacentres, whereas Barcelona has just two. There has been a clear strategy to locate the datacentres in a central place. From these central locations, they serve other regions and cities through fibre and cabling.

 

With the increase in data consumption and the lower latency required, there has been the decision to partially move some of this activity from a central location to a regional level. Eventually the trend will be for this to move to a local level and then to a site level. The idea sounds exciting – that every tower could have a datacentre providing for example, gaming solutions – but in reality, we are a long way from this. That demand doesn’t exist yet. 

 

What is however important to understand is that the idea of co-locating a datacentre at a tower is to enable the content to go through the radio interface of the tower. Plugging the content in with the radio access network is where the real value sits, and is how you reduce and minimise the latency. Again, we are not yet there. 

 

Yes, there are some towers with datacentres, and Cellnex has some, but in these instances, we are working with retailers who solely wanted a good location with access, security, power and cooling. We are providing a co-location service for a regular datacentre, there is no interface between the datacentre and the radio equipment on the sites. 

 

Edge computing will come from integrating the content of the content generators (the OTTs, Amazons, Googles, Microsofts of this world) with the distribution networks of the MNOs. This is the big opportunity and the big picture that everyone dreams about. Whilst there are some examples in Europe, we are not yet seeing widespread demand to integrate the content with radio access networks at towers. 

 

At the time that this happens, the companies who will be best positioned to benefit from the opportunities will be those with a pan-European network of sites - the big content players will want a solution in multiple countries. Companies providing edge computing will also need to offer an orchestration or control service layer. For example, think of a major city like Barcelona. Let's say at 6PM in the evening you have lots of teenagers gaming, at this point you would need to establish an automatic software process that loads all the gaming software to the servers. At 8PM, the teenagers stop gaming, and you then need to dismantle this software and install, for example streaming services. There is a layer of software to manage the transition from one service to another and to make this active equipment to many players – to MNOs, content players and whoever. This orchestration layer will make it much more interesting, and customers will pay per usage. It is a nice story, and it will happen, but it is not yet a reality.

 

So, what is the reality today? Because of the convergence of fixed and wireless networks on the MNO side, and the traffic they get from the fibre-to-the-home (FTTH) networks, MNOs require some central metropolitan offices to interchange the traffic. Cellnex has some of these metropolitan and central offices in France with the bulk of the traffic coming from our partner, Bouygues Telecom. There is however some remaining capacity in these datacentres which at some point could be used by content distribution players which need much more density of locations. This is not yet massively deployed but holds potential in the future. 

 

 

If you have a netco, one very real opportunity is managing the radio access network. To date, RANsharing has not been promoted by the towerco, it has been established through agreements between the MNOs. Having an enabler in between that can manage the RANsharing makes things much easier and cheaper. It is an opportunity for a towerco, but it is a complex opportunity because you are then responsible for radio performance. You need engineers, you need NOCs, you need to manage SLAs and you need a field force; it is a natural evolution, but it is a different business.

 

Other areas linked to the tower space that could take special attention could be V2X (Vehicle to Everything) communications. Every day cars are becoming more autonomous and although the cycle to make all cars completely autonomous may take 15+ years, there needs to be a major increase in telecom infrastructure rollout to prepare for this. Initially we will see the infrastructure being required along the main corridors and highways, but eventually this will move to regular roads. The infrastructure will need to be flexible, lean and intelligent. There is no continuous power grid along roads and so renewable energy will be required. It does not make sense to have the site on through the night when there is no traffic and so there will need to be a management system with edge computing to control the intelligent switching on and off of the site. You also need to consider the different stakeholders that could interact with these networks, for example, alerts from emergency service vehicles that move cars out of the way when they are coming through or a notification about special offers at an upcoming service station restaurant. There are many things to explore and we think that V2X is a very interesting space. At a basic level it could just be MNOs deploying more 5G sites, which is good for us, or maybe there is the opportunity to deploy dedicated networks for this purpose which would be more complex but also hold higher opportunities. There is activity starting in the US and Europe, the European Commission has been very active on promoting projects to provide connectivity in key transport corridors along cross border zones (extending to public transport networks as well as roads), as a starting point.

 

Another area to think about in the future is quantum communications. Currently quantum communications are provided by fibre which transmits data that has been encrypted with numbers created by a quantum computer. Such steps improve the privacy and security of communications. If we want to set this layer of security to mobile networks what does this service look like so that MNOs can provide it to the enterprises and end users. Who are the end users? There will be many potential customers, from government to financial services to healthcare and utilities. This isn’t just a pipedream; it is reality and there is huge investment happening in Europe around quantum communications. I’m not talking about quantum computing – this will come later, but rather the encrypting of data with numbers created by a quantum computer. It is something that is already within the scope of our innovation team’s focus – it may not be a short-term opportunity, but it is something to which we are already devoting  some attention. it is important to be close to those who are building the new quantum technologies – this is not your traditional tech companies, it is for example public technical research institutes.

 

As a business, we see a host of other exciting opportunities ahead!