11436 SSO

An Agile and Adaptive Cycle for Business via APIs & Big Data

Chet Kapoor
Aug 18, 2014

The proliferation of smartphones, tablets, and internet-enabled sensor-controlled devices is changing the way consumers adopt and use technology, and the way customers interact with enterprises in every sector.

Every business needs to know the customer and do business wherever that customer wants to; it needs to be nimble and agile; it needs to go to market and onboard partners more quickly than ever before; and it needs to tap into a broad base of innovation inside and outside the enterprise.

What does it take for enterprises to participate at scale and create a sustainable competitive advantage in this mobile- and data-driven, cloud-powered era of IT?

Old tools won’t solve new problems

Becoming a successful digital business requires a whole new approach that marries two major shifts already underway in some enterprises—the shift to a modern API-based infrastructure and new powerful data and analytics capabilities. Web browsers and web infrastructures have ceded ground to application and API infrastructures, which have become the foundational technology for the development and deployment of scalable enterprise applications. In this app-centric mobile world, machine access has taken over from human browser-based access; as a consequence, the human “think time” that was built into browser-era processes becomes obsolete. Machines need no think time! This reality drives important new requirements for enterprise infrastructure, including OAuth, rate limiting, quota, caching, and sophisticated API and application analytics.

Secondly, as a vast amount of data is generated by increasingly connected customers, the volume, velocity, and variety of data is driving enterprises to look for new ways of working with data and the emergence of new kinds of digital services.

Beyond the 3Vs—volume, velocity, variety—the Intelligence from big data is proving critical to understanding user engagement and for optimizing the experience on multiple channels via applications. Informed by big data analytics, enterprises can create individualized information and services, tailored for who customers are, where they are, and what they intend or need at a given moment. Additionally enterprises need to be able to take advantage of multi-dimensional data that includes time-sequenced (or temporal) information. Coupling temporal data with transactional and contextual data enables an enterprise to make informed decisions as business events unfold.

Making big data analytics a first-class citizen in enterprise infrastructure

Applications provide an enterprise’s data and services to customers, capture a myriad of data from users, and provide it back to the enterprise. To be where customers are, enterprises must develop and adopt new channels and ensure alignment across channels without a major retooling of their infrastructure; they must deliver sophisticated and relevant experiences on every channel and support customers doing business at the edge.

Doing this well demands that historical and contextual information be augmented with real-time and time-sequenced data, both of which hold valuable clues about in-the-moment intent and needs—critical information for an enterprise when designing individualized and just-in-time actions and decisions.

Not only are applications delivered by APIs, and powered by data; they get better with data. In an API-centric architecture, APIs gather real-time data across all channels, and every application is responsible for exposing its data in a structured way via APIs. The API-based push and pull of application data enables a complete feedback loop—relevant data can be shared easily with analytics systems and predictive intelligence engines, and in turn, deliver back data-driven, contextually-relevant actions based on real-time feedback loops driven from those same analytics systems.

There are two dimensions for data in today’s enterprise: operational and business.

Operational statistics provide an enterprise the information to adjust operational levers for APIs and applications. Enterprises get real-time data on the health and performance of their APIs, can reach beyond APIs to monitor mobile and web applications, and can identify performance bottlenecks in applications by network, carrier, and application version. Predictive analytics can model an enterprise’s growth and help with capacity planning, can use risk scores for risk modeling, can predict genuine as opposed to bogus transactions and so improve security, and much more.

Likewise, predictive analytics takes the guesswork out of business outcomes, enabling the enterprise to understand and predict business and market changes. Analytics are critical for the understanding and optimization of the entire digital value chain and especially the experience for the human being—after all, at the end of the enterprise’s digital value chain sits a person with a mobile device.

Enabling an agile, adaptive cycle for business

APIs have become the de facto standard for exposing and consuming data and an increasing percentage of customer interactions flow through APIs. Every interaction generates digital intelligence, and applications that learn from this intelligence to anticipate and provide the right experience on the right device, at the right time, and for the right person, are the key to customer satisfaction and, ultimately, to turning a business from reactive to proactive.

Building and operating adaptive applications and the APIs that power them means separating the signal from the noise of big data, gleaning actionable insights from patterns in the data (and not just in sample sets), and taking action on those insights. It means collecting and analyzing the enterprise’s transactional data along with real-time and time-sequenced data to predict outcomes. It means delivering that digital intelligence in a feedback loop to optimize and improve APIs and applications.

Given the tremendous pay-off for the business and IT of using the history and intelligence gleaned across channels to predict a customer’s intent and next action, and of customizing experiences based on transactional and behavioral history and propensity, it is imperative for enterprises to have API, application, and predictive analytics capabilities built into one digital business platform.

Trying to bolt on the predictive intelligence dimension through a system that is separate from your API platform will be slow, expensive, and inaccurate. Through one digital business platform, enterprises will enable an agile, adaptive cycle for business—a cycle of collecting big data, analyzing it, anticipating and predicting outcomes, and acting to improve apps and APIs and so improve their bottom and top lines.


Scaling Microservices