Do femtocells offload the type of data traffic that would make a difference to the macrocellular network? We have been witnessing an explosion in mobile broadband data traffic over the last 12 months. Donglemania has taken root. Whether in Western Europe, Asia or elsewhere, you can often see laptop users out and about with a USB Dongle, accessing the web, email and internet services wherever they are.
 
Flat rate subscriptions from $20/month have brought this into comparable range to fixed line DSL, but without the restriction of working only within the home. This is ideal for business and personal users who are nomadic, using their laptop in different locations during the day, or who are not able to commit to a 12 month contract.
 
Performance can vary, but this compares with DSL lines that are becoming more congested and where headline peak rates are often not achieved (typical achieved rates in the UK are 4Mbit/s compared with the advertised "up to 8Mbit/s").
 
Operators report that the volume of data traffic (in bits) on their networks now exceeds that of voice (i.e. more than 50% of bits sent over mobile networks are now for data services rather than voice). Recent annual financial reports have trumpeted the success and growing profits from these data services. However, voice revenues continue to dominant with more than 70% - the price per bit of data is very much less than that for voice.
 
This transition is revenue mix has already changed for fixed line operators. BT's voice revenues are only around 10% of its total income - it has diversified into a services organisation, and rents copper lines, broadband and other fixed price telecoms services. Until recently, most mobile operators made 80% from voice and 20% from SMS with just 1-2% from data and all other services.
 
So will femtocells solve this problem by offloading much of the low price, high volume data traffic from the outdoor cellular network onto customers own equipment?
 
I'd argue that the current explosion in mobile broadband data traffic is of the wrong type of traffic to make this jump:
- Outdoor use, where there is no fixed broadband alterative
- Transient use, when travelling through a transport hub
- Fixed/Mobile Substitution, where mobile broadband replaces the need for a fixed broadband (e.g. short term rented home, houseboat)
- Low data usage, where prepaid mobile broadband data compares favourably with fixed broadband subscription.
- Prepaid users who don't want or aren't able to get a fixed broadband service.
 
Indoor laptop wireless data is currently well served using WiFi in the home. Sure, if the same system worked inside and out, users may find it easier to remain using the mobile data broadband system. Some may find or perceive it to be more secure that WiFi (it defaults to encrypted transmission over the air interface and is much more difficult to eavesdrop).
 
There is a growing percentage of mobile data use in smarter phones and devices. High profile devices such as the Blackberry and iPhone can claim to have driven some mobile data usage, but are actually very efficient devices and don't need huge volumes of data to display text email messages or small sized pictures. Mobile TV does not appear to have taken off, and although many phones incorporate good quality cameras, pictures are rarely sent wirelessly. Recent figures compared 1 Billion SMS with 7 Million MMS picture messages, less than 1%.
 
What is needed is perhaps a continued growth in the availability and use of smarter devices, combined with a growing confidence in users that they understand the price and capability of using data services on it. Where before we saw only very compact (mainly voice) mobile phones and large screen laptops, we are now seeing a convergence between ultra-mobile PCs (ie very small laptops) and high end smartphones. Rather than take minutes to open up and start applications on a laptop (mine takes 6 minutes to boot up), a more convenient connected mobile device might be used for mini-surfing which will interwork and synchronise with other internet applications for daily use.  
 
Other suggested applications include downloading podcasts or music, which previously might have been sideloaded from a PC onto a mobile device. There is added convenience of having the music delivered seamlessly.
 
In summary, don't assume all the data traffic growth seen on mobile networks today is suitable for offload to femtocells. Instead, a new generation of data traffic will evolve over the next few years which together with some of the data we are seeing should justify the operator's business case for femtocells.