Skip to content

What is custom code? To answer this question, let’s begin with a metaphor. Imagine that code is a vehicle: a blue Honda Civic.

You might say to yourself, I want this blue Honda Civic. So I will put the car together, build the engine, paint it, install all the electronics, get it entirely road-legal and become responsible for driving it and maintaining too. In other words, if the car breaks down, requires enhancements or fixes, then you’d be solely responsible for fixing the issue or would have to hire someone to take that responsibility. That’s the definition of custom code.

On the other hand, imagine that you received that Honda Civic – or code – as a pre-packaged car that is ready to drive. Then, you are told that if anything goes wrong with the car, you can bring up the issue to the nearest accredited Honda Civic Repair center for free, they will fix it and do any maintenance necessary at any time. In this scenario, the vehicle is not custom code.

Custom code, as you can see, is time-consuming and complicated to create, maintain and change. Its advantage is that it can be used to support any use case or

Custom code: Point-to-point integration

Custom code comes in many shapes and forms. Often, custom code is used to connect two systems together, which is known as point-to-point integration. Point-to-point integration requires developers to build custom code between numerous applications, systems, data and devices within the enterprise.

Although custom coded point-to-point integration quickly and easily enables communication between data, it is not future-proof and does not scale. Point-to-point integration create one-to-one connections between data. This leads to “spaghetti architecture” and creates a system that is tangled, fragile and difficult to maintain.

spaghetti architecture

As business requirements and initiatives inevitably change, this spaghetti architecture makes minor modifications very strenuous and time-consuming. Although custom coded, point-to-point integration may seem like a quick, short-term remedy to the challenge of integration, custom coding and relying on standalone integration tools creates complications in the long-term. This leads to scalability issues and impacts business agility.

Beyond custom coded point-to-point integration

Enterprises must move beyond custom coded, point-to-point integration and, instead consider a new integration approach based on reusable, purpose-built APIs, otherwise known as API-led connectivity.

This approach facilitates plug-and-play integration and makes adding or changing systems as simple as connecting LEGO blocks. Data is connected to applications through modern, managed APIs  that can be easily discovered through self-service and controlled through governance.

With API-led connectivity, retailers are able to integrate data from disparate applications, systems and devices and expose that data to create new applications or extract more value from existing applications.

Unlike custom-coded point-to-point integration, API-led connectivity makes it easy makes change easy by using APIs to make connecting assets, systems, or data simple ; if something needs to be added, changed, or subtracted, assets can be plugged or unplugged with ease.

Extend your AI capabilities with MuleSoft.

Start your trial.

Try MuleSoft Anypoint Platform free for 30 days. No credit card, no installations.

Try for free

Talk to an expert.

Tell us a bit more so the right person can reach out faster.

Request a call