How a single pilot API provided a platform for business continuity

I’ve recently been in touch with a friend who is a lead architect at a large-scale distribution company to understand how COVID-19 has impacted their business and what they are doing to ensure business continuity. Suddenly the global coronavirus pandemic has upended how companies operate and respond to changing consumer behavior and shocks to the supply chain. Companies need to adapt by enhancing their digital capabilities and transforming IT operations to be more responsive, resilient, and efficient.  

As businesses across industries start to re-evaluate their IT infrastructure and processes, the mentioned distribution company is using this situation as an opportunity to learn and turn a tactical approach it implemented in the near term into a strategic part of its long-term business plan. 

Tapping into the large potential of a small-scale API

The distribution company has a large footprint, serving multiple states and markets. It wanted to improve operational efficiency and create a connected digital experience for its sales reps and customers. However, its critical operations were backed by old and manual processes, and it had limited customer-facing applications. In addition, the company had numerous legacy and modern application systems connected via point-to-point integrations. This increased operational inefficiencies and forced teams to reinvent the wheel each time they needed to deliver on a new project or develop a new experience.

To support its end-to-end supply chain, the business deploys a high-touch model to capture orders, returns, and payments. Now with travel restrictions and shelter-in-place mandates enforced in many regions, this manual order-capturing process has led to limited orders and delayed payments — leading to a sharp decrease in sales. 

To address this challenge, the company brought to life an internal and low-priority pilot order capture API with one of its customers. The only function of this API was to persist the order request in the messaging system. The API consumer reused the existing API to quickly deploy a very simple UX application to publish the order requests, helping the business to better serve its online orders. Soon, the company saw the potential to go large scale with this single API to support their wider business goals of increasing revenue. Suddenly, this originally low-priority and small-scale API became one of the most critical assets for providing the business with a platform for business continuity. 

Developing an API strategy to support current and future business goals

The business goals of deploying the order API were to increase operational efficiency, reduce logistics cost, and allow sales reps to focus more on customer acquisition rather than on the manual process of capturing orders. The order API example above shows how organizations can shift to an API strategy to create new opportunities and improve existing products, systems, and operations. The distribution company is now using this learning moment as an opportunity to test their vision and adapt their strategy in the face of disruption. Here are some ways they are addressing new challenges during COVID-19:

Customer onboarding

The enterprise is looking to streamline the customer onboarding process and mature its API program. Here are a couple of ways that they can consider an API-led approach to accomplish this goal: 

  1. Customer-focused contracts: The current consumer-focused API is bundled within the core order processing logic and is not reusable, meaning the end-to-end process is designed to work only with a single customer. Now with multiple consumers signing up, it is becoming difficult to expose the same functionality without duplicating the effort and experiencing downtime from this API. With an API-led approach, the company can follow a consumer-first model and design and deploy in a more manageable microservices-based approach. This approach allows organizations to build an application network where it is easy to plug and play new consumers, while reusing and connecting with the underlying backend systems and services.
  2. Discoverability and self-service: The organization understood the importance of enabling its consumers to discover and self-serve the APIs and integrations needed to deliver projects and innovate faster. By leveraging API-led connectivity, the organization is able to connect data through a series of APIs that are each built to play a specific role — unlock data from systems, unify data into processes, or deliver an experience. The real benefit of this approach is that the APIs themselves become reusable. While a team might invest in unlocking a set of systems with APIs for a specific project, the same team or another team can reuse them across the organization in future projects. Others can also discover and self-serve the existing APIs and microservices versus building from scratch, resulting in quicker time to value. 

Managing the unknown

Maintaining operations is one of the most important tasks for IT, especially during a major disruption or crisis. The company’s current monolithic and on-premises-only-compatible design doesn’t allow service implementations to easily scale. With limited management and monitoring tools, it becomes difficult to block bad traffic that could compromise the stability and availability of the system. With a growing external consumer base and an expansion of its API-backed network, the enterprise needed to manage any unknown that could bring down the entire system (e.g. surge in API calls or a bad deployment causing bad requests). In fact, it actually experienced a bad consumer deployment that brought down the production system – impacting operations and delaying orders for consumers. To manage for the unknowns, the organization needs to consider the following: 

  1. Availability and scalability: Adopting an API-led approach not only enables innovation but also ensures that each service or API can be managed independently and in isolation. For example, with the order API, the processing logic can be scaled without scaling the individual consumer-focused APIs to manage a peak in traffic. This approach decreases the effort and cost when compared to scaling the entire monolithic application. For example, applications developed on MuleSoft’s Anypoint Platform are built on the principle of “build once and deploy anywhere,” which allows you to quickly migrate and scale workloads in the cloud without investment in on-premises servers – bringing down the overall operating cost.
  2. Backend stability and end-to-end API management: As the company experienced an increase in consumer base, it also increased the risk of incidents and called for end-to-end management. The company had to ensure that the legacy backend system was not bombarded with too many requests and needed to track each consumer and API usage to stop bad requests from breaking the end systems – impacting day-to-day operations. They also wanted to gain visibility into end-to-end transaction flow to reduce MTTI and MTTR and to be proactive in responding to any incidents. Features like Anypoint Monitoring and Anypoint Visualizer, as part of Anypoint Platform, helps to provide enriched visibility into the application network. With a holistic view of the organization’s APIs and integrations, users can troubleshoot issues quickly, understand where issues may emerge before they happen, and pinpoint the root cause when they do arise.

Building the foundation for business continuity

If done right, a single API can set an organization on the path towards building a foundation for business continuity. In the distribution company’s case, the single order API use case is now being expanded into a mature API program. With a leadership-sponsored API program, the company is now working on a new initiative focused on an order visibility functionality. This capability will enable the sales reps to better: 

  • Engage with customers and internal teams (e.g. procurement).
  • Gain better visibility into the number of orders fulfilled versus backorders.
  • Effectively use the task force to minimize the backorders.

If you’re looking to maximize the potential of your APIs to support your business goals, check out our API Program Workshops here to get started on your organization’s API-enabled digital transformation.



We'd love to hear your opinion on this post


One Response to “How a single pilot API provided a platform for business continuity”

  1. Efficiency is the best use case of an API. Developers can meet their objectives on time and do more in less time due to API support. The API itself has no implementation, but it specifies how software components should be assembled to develop a program. Basically you dont need to “Reinvent the wheel”
    And specially in times like these it becomes even more critical to have plug and play modules!

    Legacy backend systems can be an issue and needs to be managed accordingly! Good perspective!