BY Capgemini
This page was produced by FT2, the advertising department of the Financial Times. The news and editorial staff of the Financial Times had no role in its preparation.


IT trends spotted and checked by experts

Founder and CTO, Cityzen Data

Mathias Herberts is CTO of Cityzen Data, a company he co-founded in 2013. Cityzen Data provides a platform for the Internet of Things and sensor data. Before this venture,...


Going it alone in the Internet of Things sector is illusionary

18 Jan, 2016 10:41 am

For Mathias Herberts, cofounder of Cityzen Data, mastering the connected objects sector requires skills such as managing data storage, handling data collection networks and guaranteeing the longevity of services. These are tasks that a company, for its own sake, should delegate to others.

The challenges of the internet of things sector are far greater than those so far presented by the world of big data. With big data, social security administrators have learnt how to process one billion healthcare-related forms per year, the RATP (the Paris region public transport authorities) have been able to transport 5 million passengers a day and banks recorded 8 million bank card payments in 2013.

With the internet of things, passenger plane black boxes record 300 different parameters per second for around 100,000 flights per day, and the electricity meters of the future will collect information relative to 35 million households every ten minutes, that is to say 5 billion measurements per day. With the internet of things, it will no longer be enough to quantify each of our activities: we will need to break them down into smaller chunks of information.

It is not possible to have an "end-to-end" approach 
All of this leads to data analysis challenges. Current providers of connected objects often try - and wrongly so - to have an "end-to-end" approach, that is to say they want to possess the data produced by their own objects and associated services, believing this approach will give them a competitive advantage. In reality, however, only access to data which can be analysed is important.

Possessing data adds a useless layer of complexity. Platforms need to be created in order to deal with several technical details: a high quality data intake process, data storage, storage scalability and, of course, a sufficient level of security. The provision of these services to an optimal level - one that is compatible with the world of the internet of things - is a sector all of its own.

Connected object manufacturers need to be aware of the fact that it is in their best interests to outsource the task of storing their data to platform providers who have been emerging recently and who propose IoTic (Internet of Things analytical) ecosystems. Not only are these companies innately developed for the internet of things sector, but they also offer a large number of analysis possibilities for the connected object provider, who can also offer a large number of consumer-oriented services.

Plurality benefits the consumer
The emergence of such platforms place consumer interests at the heart of their activities, as they are "agnostic": they can collect data originating from different objects. This means that a user of several connected objects, each produced by a different manufacturer, can, thanks to these platforms, benefit from third party applications offering cross-cutting services. These services will generate supplementary data which, ultimately, will enrich any data that the connected object provider would have been able to obtain through their own services alone.  

Furthermore, these agnostic data collection sites, which are sometimes dedicated to specific sectors of activity (like insurance, for example), enable consumers to formally authorise the use of their data. This is important because, let us not forget, this data remains the consumers' data. If companies forget this, they risk facing accusations that they did not act in the consumers' best interests.

Finding new data collection network partners
Not having to deal with data management will enable connected object providers to concentrate on their core business activities. And, in this domain, there is already a lot to do, starting with choosing the way in which data will be collected. Although the networks that are currently in place can be used immediately, they are not set up to accommodate new types of usage. It is therefore necessary to study possible alternatives and find partners capable of managing them.

Among these alternatives, we can cite the possibility of using smaller cells (Femtocells, or even picocells, such as those that JC Decaux and Vodaphone are looking to install in urban properties). Another option is to opt for different types of technology, such as Bouygues Telecom's Lora network or that of newcomer Sigfox.   

Developing a connected object is a sector all of its own
Even before that, the first challenge of the internet of things is above all for companies to develop devices with captors and actuators that can take in situ measurements then trigger actions in other geographical locations.

The problem is that it is difficult to make modifications based on experience when dealing with objects that are found all over the world. It is therefore important to have a vision as early as possible of how the services linked to an object could evolve throughout the life span of the object in question. What's more, this must be done without the overall cost of the object spinning out of control. This, once again, belongs to a sector all of its own. 

These challenges, within themselves, create myriad opportunities that visionary entrepreneurs will no doubt seize - entrepreneurs who will become the privileged partners of the already well-established companies. This will only happen, however, if the players already in place do not try and opt for too sovereign an approach.