Legacy Transformation with API Led Approach

Ashish Kalve, Director of Software Development, American Tower

Ashish Kalve, Director of Software Development, American Tower

Organizations have been investing in some form of digital transformation program to reinvent their web and mobile presence which they have been trying to achieve by investing in various technologies but have not really achieved or changed with significant impact in today’s competitive marketplace. In order to bring in evolutionary and ground up transformation, they need to transform organizational thought process and not just technology. However, not all organizations are ready for that, many organizations today are still building solutions with technology investment and solution approaches of the past. Today, Walmart has done a great job in reinventing on both technology and business front but there are other retail giants still struggling to redefine themselves with the time.

As companies evolve driven by business strategy to grow at a rapid pace organically or through takeovers, acquisitions and mergers, they also need to evolve their IT strategy that aligns with the business strategy in order to maintain scale and ability to adopt to growing demands.

Typical legacy IT systems have their own set of challenges. In most legacy companies, a complex IT landscape is due to systems developed in silos over a long period of time by various internal IT teams and/or external contractors/consultants. These systems are often time patched without thorough analysis to meet immediate needs of the business, which further adds to existing complexities resulting in spaghetti IT landscape that’s hard to untangle. All this results in slow delivery due to complex inter-dependencies that in turn leads to a long release cycle affecting cost, time, quality and, finally, a delay in getting final solution ready for the business users and clients. This also leads to added pressure on governance, change/release management, and support, resulting with most of the IT budget spent on maintaining systems rather than innovation and improved services leading to high technical debt over time.

In a fast-growing technology and business landscape, rapid and transformative change within IT is absolutely necessity. Today, transformation is gradually happening as more and more legacy companies adopts buy vs build, modernize or retire legacy systems, and consolidate systems to scale as businesses grow. Companies are also adopting cloud strategy to reduce data center cost, migrating their legacy applications on to SaaS solution to gain process standardization and on demand scalability, in an attempt to achieve agility in supporting a growing business. In an environment like this, integration platform and API strategy can play a major role in defining key aspects of the overall IT strategy that will help IT speed up transformation and be more aligned with the business goals and strategy.

Defining the integration platform starts with identifying the appropriate product as a platform with essential features and functionality that aligns with the IT strategy. Companies should look for product with cloud first and API first API development life cycle built in. The product that has pre-built connectors that can quickly integrate with the SaaS solutions, custom applications deployed in cloud, and an on-prem eco system of modern and legacy applications. Product that supports reusability, flexibility, API management with configurable API policies, such as, throttling, rate limiting, SLA based governance, client ID enforcement, OAuth, various threat protection, SLA based tiers with in-depth analytics and reporting natively-built in to the product.

To set API strategy, understanding API lifecycle concepts is important in creating the right foundation for the API strategy to evolve and be understood and adopted within the company

Product evaluation and selection could be a daunting task, especially in a large organization with geographically distributed teams, but it could be easily achieved with proper upfront preparation for the evaluation kick off. We have seen great success with constructive steps such as identifying key criteria upfront that are essential to the organization, initiating internal team discussions across teams, providing product demos by vendors, identifying a practical case studies and having vendors put together a solution document elaborating case study implementation that meets all technical needs for the API, on-prem workshop sessions, and, finally, rating product based on selected criteria.

After selecting product, bringing the team onboard is the key driver to make it a success. That can be achieved by disseminating information about the integrating platform across teams within the organization, and arranging developer workshops, proper training, and deployment automation processes, etc.

To set API strategy, understanding API lifecycle concepts is important in creating the right foundation for the API strategy to evolve and be understood and adopted within the company. This includes identifying and communicating API development standards, designing lightweight and easily manageable services, simplified lifecycle, reuse through engagement, identifying and applying standard integration patterns, and creating reusable code fragments that can be imported into Open API or Restful API modeling language (RAML) for rapid team onboarding, design and development. The process includes creating security standards common to internal and external customers. It is important to review the process for re-use of published APIs time to time.

API is not just a way to gather and display data. It’s a product that’s delivered to the end consumer who will rely on the API to drive consumer application behavior and experience. It’s important that the API being developed follows industry best-practices to ensure effectiveness and usefulness that can last beyond consumer-application life. Ensure that the consumer-centric API features and offers a canonical model of the data being gathered, analyzed, and presented, and that the model can be orchestrated at a layer above data/system API or above core source data model. The API featuring a semantic model can use multiple data/system level APIs to construct the model. This approach will allow the developer to apply appropriate policies to protect core data in the most effective way and to provide clean data that’s consumer/business friendly.

The API evolution can also drive rapid application development with micro-services architecture that is comprised of independent components, self-contained, built with one or more reusable APIs. The API, along with microservice/micro-component architecture, will pave the path to retire legacy integrations and applications by breaking down silo applications into more manageable fully decoupled pieces that can scale independently and maintain resiliency with reusable APIs. This approach will reduce any disruption within business processes and increase agility and versatility within the IT organization.

As API evolves in organization, it is important to measure its effectiveness with well-defined KPIs, such as, total number of APIs built, API adoption within internal and external applications, application users, user growth rate and user churn.

A strong and relevant API strategy will provide IT with the foundation it needs to affect rapid legacy transformation that will increase agility, scalability, and the ability to respond in time. In addition to achieving agility, it will also put in place the pathway to innovation.