Welcome to the API jungle – part II – what Brands should do

Welcome to the API jungle or why developers must learn how to find, select, integrate APIs and contribute to their improvement and evolution (part II)

by Martin Duval, CEO, Bluenove

The API landscape is extremely dynamic. The following 2 diagrams taken from ProgrammableWeb describe the most common APIs which are used in order to build mash-ups. They show the dominance of major historic players (check the “see all time” diagram) but also the more recent rise of new players such as Twilio in the recent past (last 14 days), as well as the convergence of Cloud Computing and Telecom.

There is yet more evidence of the emergence of this new ecosystem; it is indeed interesting to notice the emergence of new players offering the technical support to manage API infrastructures such as the new Application Enablement Services Business Unit from Alcatel-Lucent (the one that acquired ProgrammableWeb in 2010), Mashery, Apigee, Aepona or 3Scale.

But let’s come back to the relationship between Apps and APIs.

After all, aren’t applications mere channels? To support their promotion, we can reasonably bet that Brands will also create APIs in the future, with probably a more obvious way to demonstrate Return On Investment through the number of innovative apps created by third party creative developers than through the number of downloads KPI[7] of their own app. Brands could therefore propose APIs in order to extend the reach of their products and services.

Here are some suggestions for a few popular Brands[8].

  • Nike could create a “Just Size It” API that gives the perfect shoe size from the photo of your feet,
  • Evian could create a hydration API that calculates the quantity of water a person needs to drink daily and reminds her when rehydration is needed,
  • Netflix has proposed an API to tap into its customers’ creative capabilities, and even organized a contest [9] so as to crowdsource ideas leading to the improvement of the algorithms of its movie recommendation engine,
  • French off-licence chain Nicolas could create an API that allows its customers to find and leave recommendations about the wine they buy.

How could these companies support the use of their APIs, and therefore the promotion of their Brand? This would be done by the developers who would make sure to make APIs accessible by the end users on different interfaces, and who would find ways to remunerate themselves through the proposition of new business models.

Of course Brands can still develop some specific applications themselves, but the decision to propose an Open API will offer an unparalleled way to boost exponentially the reach of their promotion.

A lot of marketing managers are sometimes the victims of the ‘gadget syndrome’: they follow the trend getting on board the last fashionable feature to include into their marketing plans. One year it is the ‘Facebook Page’, or the ‘Twitter account’, and the year after the ‘Mobile App’.

But as part of a more sustainable marketing and innovation strategy, the best solution may very well not be an application but rather an Open API.

Another trend to take into account as a booster for the number of APIs, is Open Data. The opening of public data by the administrations (After initiatives in the US with Data.gov and in the UK with Data.gov.uk, Etalab[10] is also about to launch the Data.gouv.fr portal of data sets in December 2011) and French cities such as Rennes[11], Paris[12] or Montpellier[13] have already exposed some data sets with some of them as APIs.

Open Data for businessese

The concept also appeals to businesses as shown by the Bluenove white paper (in French)  entitled “Open Data: what are the issues and the opportunities for the enterprise?” with sponsors such as French railways SNCF, French Post Office Group La Poste, SUEZ ENVIRONNEMENT and the French confectionary giant Poult group. The Civil Service, local governments as well as businesses will have to learn how to attract, engage and manage a community of developers but also of entrepreneurs, researchers, academics, students and companies from other industries to motivate them to use their APIs and boost their innovation.

as a conclusion: the fundamental role of developers

Martin Duval, CEO, Bluenove

One the one hand major platforms continue relentlessly to open themselves to to more and more end users thanks to more open developments. On the other hand, developers will try to invent new applications but will also have to use an increasing number of available APIs and use new skills to detect, select, integrate them but also contribute to improve them and even ask for new ones.

One sees new types of requirements, services and skills emerging which keep the collaboration and innovation momentum going between the members of these complex ecosystems among which developers have a fundamental role to play.


[7] KPI : Key Performance Indicator

[8] Examples from this article on Mashable by Adam Kleinberg : http://mashable.com/2011/01/04/brand-open-api-developers/

[9] One of ’the 12 levers of Open Innovation’ : see http://www.slideshare.net/Bluenove

[10] EtaLab : http://www.etalab.gouv.fr/

[11] Rennes Open Data : http://www.data.rennes-metropole.fr/

[12] Paris : http://www.bluenove.com/publications/revue-de-presse/bluenove-fait-parler-les-donnees-de-la-ville-de-paris-et-le-web/

[13] Montpellier Open Data : http://opendata.montpelliernumerique.fr/Le-projet

Welcome to the API jungle – part I – a surfeit of APIs

Welcome to the API jungle or why developers must learn how to find, select, integrate APIs and contribute to their improvement and evolution (part I)

by Martin Duval, CEO, Bluenove

As I started writing this article at the beginning of October 2011 , the http://www.programmableweb.com/[1] web site indicated on its home page that it has identified 4007 APIs and 6175 mash-ups on a global footprint: At the moment you are reading this piece on the Visionary Marketing, I am certain these numbers are completely outdated. I agree that, Dear Developers, these numbers are still very far from your ‘Ocean of Apps’ but this new ecosystem nevertheless starts to look like a ‘Jungle of APIs’.

First and foremost, let’s take the time to put this notion of Open APIs back into its context. ‘Application Programming Interfaces’ do enable the connection between different IT platforms and the integration of different application and services through the creation of a ‘mash-up’. Open APIs proposed by a mobile or web player aim at helping the creation of an ecosystem around a common platform, therefore forming a dynamic community of creative developers who are given the opportunity to innovate faster and in many more directions, than if  they wished to do it on their own. This is therefore a genuine Open Innovation strategy in which the various players will have to initiate and maintain a long-term bond of trust, based on elements such as stability, sustainability, ease of use of the platform and the APIs, but also based on a win-win relationship with its community of partner developers.

Beyond the major web platforms (Google, Facebook, Ebay, Twitter, Amazon, etc.) that propose to the developers a big set of APIs and of course the main mobile OSes (Iphone, Android, Windows Phone, etc.) offering their SDKs[2] to support the development of mobile applications, similar open programs exist as well in the Telecom industry. Telecom operators such as Orange (with Orange API[3]), Telefonica (with their BlueVia[4] program) or Telenor (with Mobilt Bedriftsnett[5]) also allow access to third parties to some of their network assets such as SMS, click-to-call, location, storage, billing, etc. in order to facilitate the emergence of new services through the innovation potential from developers, start-ups and brands.

A signal demonstrating the need for rationalisation and standardisation in this ‘jungle of APIs’ came up with the GSMA ‘One API’[6] initiative: a success still to be confirmed.


[1] ProgrammableWeb has been aquired by Alcatel-Lucent in 2010

[2] SDK : Software Development Kit

[3] Orange API : http://api.orange.com/en

[4] BlueVia from Telefonica : https://bluevia.com/en/

[5] Telenor Mobilt Bedriftsnett : http://www.telenor.com/en/news-and-media/press-releases/2009/Telenor-opens-new-interface-to-third-parties

[6] GSMA ‘One API’ : http://www.gsmworld.com/oneapi/

Duval Suggests Killing More ideas Fosters More Innovation

note: this article was originally compiled and written for the Orange Business Services Live blog

Bluenove’s Martin Duval is not only a successful entrepreneur and an open-minded innovator. He is also a controversial business writer with French flair who can deliver straight to the point conclusions. Whereas most would-be innovators will lay a stress on the production of new ideas – the ideation process – Duval knows, like most hardened innovators, that the truth lays not in that process but in the more delicate art of rejection and … killing innovations. Here are his thoughts on the subject, for the benefit of our readers:

Lately, I had a chance to describe the challenges faced by French start-ups with regard to the financing of the early innovation phase and the managing of partnerships with major corporations, and that piece was published by the French High Tech weekly 01 Informatique. In that article, entitled all players in the innovation chain should play their role!, I was stating that start-ups should only focus and partner with those corporations, which have implemented a structured and proactive business incubation and partnership programme such as NOVA External Venturing, part of the manufacturing industry behemoth Saint-Gobain, or the ‘Veolia Innovation Accelerator’. Amongst the new ‘Open Innovation’ processes which have been designed and implemented by those major corporations, I did point out the ability and the value of killing innovation and potential partnerships. I know that this may sound strange coming from a proponent of innovation but I insist, one has to learn how to say NO if one wants to get to YES.

What I mean by that is that start-ups by nature have limited resources and time to work their way through the complexity of a large organisations and handle their long-winded decision-making cycles. Therefore, when a large organisation is able to implement a process to efficiently filter out potential partners within a reasonable period of time, it is in fact sending out a positive rather than negative message. I would advise a 1 month or a 6 weeks-delay at the most as a fair period for a large organisation to get back to a candidate partner with a positive or negative answer. Delaying the response for any length of time and keeping start-up owners on tenterhooks is simply not on in my eyes. The start-up in question can then decide to keep trying to partner through another part of the organisation at its own risk, or to change what needs to be changed within its project structure or even look for another partner.

Ideally, the more a negative feedback is explained and detailed, the higher the value that is delivered: such explanations can help highlight the weaknesses within the original project, so as to better identify the target market position for the new solution. As a consequence, it is an easy way for corporations to deliver value and improve their reputation within their innovation ecosystem. It is certain though that rejecting an application actually requires a lot of preparation as well as some process and resources in order to produce the analysis and manage the follow-up the within 1 month to 1 ½ month.

As a matter of fact, rejection can take place at each gate within the open innovation stage-gate process with a higher probably and more preparation needed early in the process than ever after: from screening before initial contact, right after the initial contact and/or meeting, after the feasibility study, during the partnership negotiation and after the test. To a certain extent, I even believe that a good quality rejection process delivers more attraction and better corporate image than piecemeal success stories. Besides, that kind of process applies not only to start-ups but also to universities, government-owned or private R&D labs, suppliers and customers involved in crowd-sourcing initiatives etc.

At the end of the day, large organisations can derive a real high-end competitive edge from the management of rejections regarding innovation proposals from start-ups and other innovators. The tougher the process is, the more desirable those selective large organisations therefore become in the eyes of smaller players.

Similarly, when it comes to managing internal innovation processes, too little attention and effort is devoted to killing projects in my eyes. Oftentimes, a standard innovation pipeline is contrived – as part of an innovation process – with a wide ideation spout on the left side and a narrow tube on the right, from which successful projects emerge. Once again, there is so much value in killing projects efficiently at each stage-gate of the innovation process and here are a few examples of the expected benefits:

  • re-allocating resources to other, more promising projects,
  • learning from trial and errors and capitalising on best practices across projects,
  • developing a culture of innovation – learning from errors, aiming at success – in order to foster motivation and encourage new daring ideas,
  • simplifying project portfolio management,
  • reducing overlap if not competition between projects.

Innovation processes are becoming more and more collaborative with the help of enterprise 2.0 platforms supporting ideas and project management. Thus it ensues that sharing thoughts about innovations that should not be accepted and projects that should be stopped is an absolute must-have. Once again, let us emphasise the fact that the proper number of resources should be allocated to the screening of projects and that pruning weak ideas should be an area of focus.

Both the rejection and even the killing of bad ideas/innovations can actually deliver benefits from a competitive edge viewpoint. Are you – and your company – ready to reap those benefits and image improvement from saying NO and for killing more innovations?

Bluenove is a consulting firm specialized in Open & Collaborative Innovation http://www.bluenove.com

Add to FacebookAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to TwitterAdd to TechnoratiAdd to Yahoo BuzzAdd to Newsvine