I was lucky enough to attend the API Strategy and Practice Conference this week in NYC and was struck by the wide diversity of attendees. Sure, I expected to see the young guns heading up the industry’s hot new companies, but US Postal Service? Walgreens? AT&T? Target? An even bigger surprise was how relevant and interesting all of the discussions were. But it was obvious there is still the same old gap between the enterprise and the start-up that we have come to expect. For as long as I’ve been in the industry, the difference has always been around governance and process vs rapid innovation. Most large companies attempt innovation programs to find and nurture new ideas, but few take that innovation all the way to their production environments without putting them into the assembly line of Process. In fact, many of the large enterprises miss the point entirely by putting governance around their innovation programs.
And I have to ask… is that a Bad Thing? I have worked in both environments, usually because the start-ups I go to get acquired by larger companies, and there are downsides and upsides to both. What was different about the last couple of days is this: all of the technologists at this conference were there with innovation in mind. They are at the forefront of what has been dubbed the API Revolution and they are all doing compelling things with this technology, even the Big Companies. But the Big Companies have different external pressures. They are driven by the demands of their customers, who insist on reliability and quality above speed and creativity. Laura Merling, VP Ecosystems and Solutions at AT&T, kicked off the conference with a discussion about governance and then proceeded to run through slides that showed an API development process that was larger than most of the companies in the room. But that’s her reality – AT&T can’t ‘play’; their audience isn’t as forgiving (and hey, most of us are their audience and I definitely don’t want them screwing around with the backbone I rely on).
But to Merling’s point, governance does not have to be a dirty word in the API revolution. We see examples every day of APIs that have been widely adopted and now have the power to bring down multiple applications or websites at the same time. Wouldn’t you prefer to have those APIs governed to ensure they are well-designed, well-tested and monitored for performance? I know I would. Alistair Farquharson, CTO of SOA Software also presented about the importance of a governance model, which again spoke to the enterprise reality of a large customer base and the correlating visibility of issues that can arise due to lack of governance.
But as I said in my talk at this same conference, I have never seen the software industry have so much fun as it is right now. Developers can tap into the work of other developers and help each other build applications that are rich, full, and enticing from Day One. We are collaborating and cooperating as never before and the result is Innovation with a capital I. That was absolutely apparent at this conference – the breaks were filled with brainstorming discussions between attendees from different companies and I think we all came home with a million new ideas swirling in our heads.
So, I admit I find myself with a foot in each camp – I want to freely innovate without any more process than scribbling some thoughts on a napkin in a bar but I also want to make sure the industry on a whole doesn’t crash on its head because it has no process and quality controls in place. I have been talking for a while about the importance of ensuring quality in your APIs and not taking risks with your application by adopting unstable third-party APIs (I do carry some Big Company blood in my veins). But I struggle with how to marry the two things: governance and innovation together could make the perfect partners if we could figure out how to combine them.
Here are some ideas for driving creative thought in your organization without losing control:
- Provide time and budget to play with your technology. Often the most successful innovation projects happen outside a company’s walls where they are unburdened by the process inside the corporation. Making your APIs available for public use is a good way to foster that energy and experimentation but don’t let those good ideas get away from you. Allowing your employees to create and explore without having to get approval or prove their business case will very often result in concepts you can’t conceive in a boardroom. Idea generation by its very nature needs to be free and without boundaries. As a result, I don’t believe governance belongs at the beginning of the innovation cycle – I would venture to say that true governance should really only come in after the prototyping stage. When an idea really looks viable, then it makes sense to harness its potential and put it into a process that guarantees delivery of that concept as a well-engineered, well-tested product.
- Host your own hack-a-thons. I think more companies need to invest more freely in this type of exercise – we see it happening at smaller software companies who host internal hack days that allow employees to come together and experiment for a day or two or three with no restrictions on what they can create. While it doesn’t come free, it’s money well-spent and it is perhaps the best survival trick of all in during this API-driven era. You could end up with the beginnings of a feature that your product managers could never have thought of on their own. Having your own developers involved means you can easily bring those new ideas into your normal development process and get them to market quickly but with the same quality and approval gates you use for the features that were designed through the normal flow.
- Forget your customers for a minute. I don’t mean forever – I’m a big advocate of building what the customer needs and not what the developer thinks is cool. But… if you are truly being innovative, you have to let the cool thing have its moment. Customer requests are usually constrained by the box you’ve created with your current software product – a customer knows what they wish it would do (or not do) and usually the requests they make are formed in that vernacular. But innovation is really about forging roads where nobody thought to go and finding new customers as a result. I’ve worked in too many environments where new ideas were greeted with “But are customers asking for that?” Well, no, but that could be because they haven’t thought of it. When you let that idea grow into something, you have plenty of time to put it into your standard vetting process to see if it resonates with the market.
All in all, this is a very exciting time in the software industry as we reinvent how we design, build, launch, and engage. I hope we can find a way to keep the enthusiasm high while still maintaining a level of quality for the users of our products. Merling’s lesson is one we need to listen to eventually: Governance is not a dirty word… we just have to find the right balance.