I spent a few days last week attending, and presenting at the All About the API Conference, in sweltering Las Vegas. As a software developer, one of the things I took away, is that it is time to start pointing traditional enterprise organizations towards APIs as a universal tool to solve their data problems.
I've been hearing about "Big Data" as "The Next Big Thing" for the better part of the past decade. Trends like these always take time to bring to fruition in the corporate world, but, as far as the analytics revolution goes, if it offers what it promises, many organizations will need a bit of help in overcoming the hurdles that stand in their way.
A report published by IDG Enterprise indicates that only 37% of enterprise organizations have implemented a Big Data strategy, while only 22% of medium and small organizations (< 1,000 Employees) are able to take advantage of their data. Why? Many organizations are planning for, or are in the middle of, implementing big data initiatives, but I have seen these programs run into technical and/or process hurdles that leave organizations mired in a half-baked and useless state.
For medium to small organizations, there are two obvious problems that can be overcome:
- Not getting enough meaningful insights. You must be able to perform the kind of analysis and data science needed to make meaningful insights. Fortunately this has become a more commoditized space with vendors ready to step in and handle these technical complexities. The usual enterprise players like IBM, Oracle, and SAP have created feature rich offerings in this space over the past few years and startups like Periscope are popping up with innovative solutions all the time. The only thing they need is access to your data, and that brings us to the second hurdle…
- Not addressing data silos. Departments in many organizations have grown their datasets organically for their own use. 25 years ago, as these systems were being created, the need for an overarching, connected, strategy across departments did not seem so evident. But today, organizations are staring down the barrel of a total I.T. reboot - in order to get all of their systems connected and leveraging one another.
A total rebuild, of any system, can be paralyzing for businesses. Many may think that they are going to be locked out of the analytical revolution, but your organization's I.T. departments can look to the cutting edge of software architecture to find a solution. Spurred by Amazon and Google in the late 00's, the "micro service" architecture completely changed the software development space. Isolated programs, performing single small functions, are connected together using REST API's to perform larger roles. This standard interface between small programs makes change management very easy and provides limitless interoperability across a platform.
If we begin to look at existing Legacy I.T. Infrastructure, it is not a perfect example of a micro service. These systems are often massive, and are performing many diverse operations. Regardless, “micro service” methodology can still be applied to make the legacy mess more manageable. Even though the systems are not ideal micro services, the gains you can see by exposing them through a standard managed endpoint are huge.
For a small upfront cost of creating API endpoints, an organization can also expose all of their data and functionality - not only across their business units - but also to vendors in a simple, standard - and secure - way. This is a huge win.
As time marches on towards whatever the next "Next Big Thing" in I.T. is, there will always be administrators fighting to keep up. Paralyzed by extremely difficult change management, and legacy systems that no longer meet the requirements of todays connected world, these admins need to act. With a diligent API-first mentality, a connected and expressive I.T. ecosystem is closer than it might seem.
Interested in learning more about APIs? Reach out with your questions!