This is the first part of our series on How Enterprises Foster Innovation and Grow Developer Communities.
During European technical director for SOA Software Andy Jones' APIcon talk (see the video below), he said that even in his first work with the aerospace industry, “The nature of innovation is important to how the system works.” He went on to say, “This idea of innovation being through either large steps or through many small steps kind of influences how, in my view, you have to go around building your community for development of applications.”
One way to address it is through the aggregation of marginal gains. Jones talked about how British Olympic cycling coach Dave Brailsford used performance management to go gold by making these many small, incremental improvements of about one percent each that collectively had a huge impact on overall performance. “By improving every single thing around the team, he could take the basic talents of the individuals and you deliver the best possible outcomes for them.”
The next step comes by working on the innovation value chain. This value chain links together the essential work of both management and developers. Jones admits that enterprises trying to engage with the developer community is “not necessarily a natural alliance” because “the natural practices of a tightly run business stifle innovation.” Perhaps because of that, he suggests that the best way to do it is by creating a framework of excellence and innovation, which he says requires inspiration but can be then addressed in a logical structure.
This Innovation Value Chain has three phases:
- Idea Generation: These ideas can come from development, can come from the end-user customers, and they can come from any part of your business that uses or accumulates data and would like to use it in a different way.
- Idea Conversion: Then selecting which ideas to follow through in the second step of Idea Conversion.
- Idea Diffusion: Finally, as this funnels down, it’s time for Idea Diffusion so that ideas are deployed to their maximum advantage.
Business value doesn’t accrue until the idea is turned into production and then properly presented to developers in a way that engages them.
When something successfully runs to the top of the market to the surprise of everyone, Jones says that either we didn’t understand the market to begin with or we didn’t understand how the idea could be executed. He says this Innovation Value Chain doesn’t work if along the way:
- there are too few ideas
- or you can’t determine which ideas should move forward
- or can’t decide which are the good ideas
- the ideas can’t be executed
- you didn’t create the right constraints to make it a success
- or, by the time it gets to be the time to diffuse, you simply can’t collaborate enough to make it happen
- and you might not be able to collaborate with the different geographic or sector markets
“I think it’s worth remembering that APIs enabling a lot of these business functions is going to be transformational for these businesses and therefore all the normal disciplines of change management are going to have to pop up somewhere,” Jones said.
For an enterprise working with an Innovation Value Chain, is is natural to ask what is the business case behind APIs.
- outsource model — increased productivity, more possibility of creating a good idea
- organizational flexibility
- reuse, recycle — use historical data within current system; but this comes with the risk of exposing an internal system that was never intended to be exposed
Jones warns that if you are using an API to better access parts like the external customers, by the nature of the enterprise, the API itself can become just as entangled as other supposedly less agile approaches.
Following traditional business techniques of measuring how an API is successful and how the market receives the API becomes essential toward having long-term innovation via a long-term development program. And, of course, depending on your type of business, the influence of those business techniques and notion of corporate governance will vary.
If it’s a highly regulated industry, like pharma, healthcare and point of sale, that framework of innovation and relative heavy-handedness has to exist. This is a situation where bottom-up innovation can never fully exist because there is an enterprise-level risk that needs to be assumed at all times.
You must constantly focus on “What it is you really do and is it actually adding value.”
“Fundamentally, you are looking for them [developers] to innovate on your business model, so you must expose the business model as simply as possible, particularly if it’s built in the language of the customer if it’s at all possible.” Jones continued that if you are inviting non-specialists in your industry to access your service, you have to strip away complications. He goes on to admit that “Things can be made as simple as possible, but no simpler,” in order to make sure it is rolled out as a business benefit.
Collaboration must be done with the acknowledgement of external business influences.
Before ever building anything, ask your external partners, “If we had an API that looked like this, what would you do with it?” They may have no need for it at all, they may have a totally different want or use for it, or it may be just what they need. You don’t know until you ask.
Create a virtual version of your business and let people play with it.
Jones said “By creating the appropriate abstract model of the business, by understanding how that maps onto the reality with the systems that you have, you’re then ready to have that conversation to say, if the idea is good enough, we can make that happen by these kinds of advice and changes.”
This is part of our series on How Enterprises Foster Innovation and Grow Developer Communities. In part 2, we will look at how developers can move innovative ideas forward within an enterprise.