This article first appeared on Workato’s website.
The proliferation of best-of-breed SaaS applications has enabled HR leaders to create business processes and HR experiences tailored to their business needs. However, it has also introduced a new challenge for organizations looking to leverage the best that each unique solution has to offer, leading to an ecosystem that’s continuously in flux.
It’s tempting to constantly upgrade or replace applications with ones that seem to provide incremental value. While this accelerates the pace of business process innovation, it still takes planning and effort to adopt new systems and business processes and migrate data into each new application.
Every time an old application is sunset and replaced with a new one, the HR and IT teams need to ensure that this change does not disrupt the employee experience. On day one of the transition, employees need to have appropriate systems access, and all demographics-related data needs to be correctly populated. Historically, teams created custom integrations to connect each new application directly to the HRIS. This model works, but as companies adopt more best-of-breed solutions, it creates a proliferation of point-to-point integrations. Building and managing these integrations can cause a significant bottleneck for teams that can prevent them from getting to more exciting automation projects that elevate the employee experience.
Of course, people much smarter than I have already identified this fundamental problem as an opportunity to develop modern integration and automation platforms, such as Workato. The automated infrastructure scalability from iPaaS, coupled with increased velocity in low-code development environments, presents the perfect means to begin addressing the proliferation problem that the SaaS application world has created. But it is never as simple as purchasing a magical piece of software that will solve all your problems with the flip of a switch. The fundamental practices around proper data management architecture become more important than ever as they need to account for flexibility and scalability, while still maintaining the performance standards offered by traditional ecosystems.
The challenge then becomes identifying the optimal architecture for developing a scalable integration and data management strategy.
Developing A Scalable Model
In my opinion, a microservices approach is best positioned to address the problem. The fundamentals of this model alleviate the core pain caused by application proliferation with its maintainable, loosely coupled and independently deployable design. Rather than viewing each integration project as a unique problem with a unique solution, you can short-circuit this approach by combining production ready components and applying the minimum required customization to quickly get to value. So how do you implement this approach using a modern iPaaS solution?
The core architecture is to develop a publish-subscribe messaging model for the integration by generalizing the funnel out of your HRIS and building specific components for user and demographics data management directly into target applications.
Step one is to understand how broad the data funnel out of your HRIS needs to be to cover the requirements of your downstream applications and where the published data will reside. The typical goal is to identify as much of the data needs upfront to minimize the chance of re-opening that core data publishing funnel; however, I would caution against over engineering this piece since it will increase your time to value for your tactical integration needs. Utilizing a solution like Workato reduces the complexity and risk with upgrading old work later down the line if additional data requirements are identified as new applications are onboarded into the ecosystem. You will also need to identify the publishing mechanism. In our experience, we have successfully implemented this pattern using tools such as a SQL database, Snowflake, Okta, and Workato’s PubSub feature to name a few, and the selection comes down to your organization’s specific infrastructure and patterns.
Once the funnel to publish the HR data has been built, you can focus your attention on delivering the data into downstream applications. Since your publishing funnel should be configured to be a simple data feed straight from your HRIS, identification and solutioning for data transformation requirements will occur within the subscribe processes. You will need to understand what each application requires from a structural perspective and build the component to ingest the published data, transform and structure the data, and deliver it to the application for processing. Platforms with pre-built connector libraries will simplify the experience when interfacing to applications, and integration service providers like Dispatch Integration can assist with connecting to niche applications or accelerating complex use cases.
While this all addresses the functional work related to this microservices model, the importance of following proper integration project practices should not be overlooked. Structured testing and process documentation is as important as design and development with the microservices model built on the concept of reusability. You are developing a one-to-many model to move data and have a single point of failure that will affect all subscribing applications. Thorough unit as well as user acceptance testing ensures that the solution has been properly validated from a technical as well as business perspective. Paired with clear documentation outlining where and when components can be reused, you can de-risk the current and future integrations.
Committing to the Digital Transformation
Ultimately, the best-of-breed SaaS application model is here to stay, bringing the ever-expanding complexity of organizations’ application ecosystems along with it. By introducing a microservices model for your core HR data management needs, you decrease your time to value and unlock availability to focus on transformational automation projects which elevate your organization’s overall employee experience.
The effort to move to this model is not to be understated as it requires a fundamental paradigm shift on how to structure data flow across the organization. Dispatch has experience successfully implementing this model and can bring valuable expertise around best practices and navigating the nuances of your organization’s specific requirements.
Contact us today to learn more about automating across the employee journey – from onboarding through offboarding.
Dispatch Integration is a software development and professional services firm that develops, delivers, and manages advanced data integration and workflow automation solutions. We exist to help organizations effectively deal with the complex and ever-changing need to integrate data and optimize end to end workflows between cloud-based, mission-critical applications.