How to Connect Multiple ERP, PLM, CRM and Other Enterprise Systems in 2024

The reality of every engineering team or manufacturing enterprise is multiple systems. For the last 20+ years, the question of system integration was one of the most popular and asked questions in any PLM system implementation I’ve been involved. The question of integrations between PLM and ERP was critical for decisions related to PLM implementation back in 2000s, and it is still one of the most often asked question in 2024 when companies are making decisions about usage of multiple SaaS systems.

Despite many theoretical debates, the reality is that manufacturing systems will live with multiple systems in the foreseeable future. I read this comment by Martin Eigner in a LinkedIn publication by Prof. Dr. Jörg W. Fischer. The question he asks is about ERP’s future and the need for ERP to have the master data. I will leave this question for another article and focus on how to integrate multiple systems, which, in my view, is becoming even more critical these days than before.

I found comments to this article interesting bringing interesting questions about what is the right way to integrate multiple systems, how the strategies of enterprise software companies are changing and evolving from the ideas of dominant ERP software position to ideas of connected systems and services. in my article today, I’d like to share my perspective of where I see integrations between systems are going in 2024 and what to expect in the future of building integrated business processes between product lifecycle management, CRM solutions enterprise resource planning (ERP), integration of customer data and creation of single source of truth. How business management software will be able to collectively integrate sales and marketing teams, build a supply chain model, organize product development processes and manufacturing processes using multiple applications. How customer service teams will be able to find best solutions in customer interactions, connect to a sales team, integrate and connect to engineering and manufacturing teams? To answer all these and many other systems you need to find the right way to integrate 100s and sometimes even more applications and data elements.

Integration Patterns

Integrations were around as long as we’ve been developing software. It is a method that required for both business and process organization. Integrations can simplify data transfer, but also can facilitate business development. Integrations were used by engineering and manufacturing organizations for a very long time to streamline processes. Integrations were used by businesses and software vendors to simplify introduction of systems to manufacturing organizations.

I can identify the following three patterns of building integrations:

Direct Synchronization: This pattern involves point-to-point integration, where data is synchronized directly between two systems. This method is straightforward but can become complex and difficult to maintain as the number of systems increases. The technology behind direct sync can be different. The simplest sync can be organized by export/import file (still very popular). Usage of APIs made it simpler and for the last 20+ years, we’ve seen many API development that helped to simplify direct synchronization (from integrations libraries, proprietary APIs, Web Services, REST APIs, etc.)

Hub/Bus Architecture with Connectors: This approach uses a central hub or bus to manage communications between systems. It is a way to prevent from developing a large number of direct integrations. Connectors are used to translate data from each system into a common format or data model. This architecture simplifies adding or removing systems and is more scalable than direct synchronization. The technological process here is also notable and introducing of a new software system only requires adding of a single connector to make integration faster.

Data Centric and Master Data Management: This integration pattern is focusing on how to extract data from multiple systems and create a central repository for critical business data, ensuring consistency and accuracy across all systems. Once this repository is built, it can be used for multiple purposes such as data search and navigation, optimizing customer focused tasks in CRM software, concentrating data from many systems in a central ERP system or integrating multiple ERP systems used by a company. This pattern supports better data governance and can help maintain data integrity in complex environments. Technology for such a data centric integrations was evolving from both directions- connecting enterprise systems and developing of global scalable web platforms.

Each of these patterns evolved over the last 20+ years based on the technology available and business process development.

Connected Systems Concept and Vision

The concept of connecting systems is becoming increasingly popular as organizations have a higher demand for integrations. The technological ability to make integration easier is advancing. Modern integration platforms and tools are designed to simplify the process, enabling businesses to connect disparate systems quickly and efficiently.

In my view, “connected systems” is more a vision rather a specific technical recommendation. And this vision is critical because it confirms that reality of multiple systems integrated to build business processes.

Industry Standards and Integration

The idea of using industry standards for successful integration was developed for the last two decades and achieved success, especially for enterprise companies that looking how to be compliant to specific standards (eg. STEP). Besides that, I can see a huge success in standard approaches organizing APIs such as REST API, GraphQL and others. Semantic Web is continue to be an important source of standards to organizing enterprise (and not only data).

Two Critical Aspects of Integration

Regardless on what integration pattern your organization will be choosing, there are two critical aspects of building any integrations:

Data: Ensuring data consistency, accuracy, and synchronization across systems is vital. This involves handling data transformations, mapping, and validation to maintain data integrity.

Process: Integrating systems also means aligning business processes. It is essential to ensure that workflows are coordinated, and processes are streamlined to avoid bottlenecks and inefficiencies.

The 2024 Integration Landscape

The technology and vision behind system integration is advancing. Modern SaaS systems makes deployment and integration easier and faster (eg. REST API, GraphQL, etc.) Advanced data management technologies can provide a better mechanisms (eg. Graph Models, Graph Databases, Semantic Web technologies). In 2024, I foresee several trends in the integration landscape:

Continuous Exploration of Multiple Applications and Services: Companies will continue to leverage various applications and services to meet their unique business needs. The diversity of tools will drive the demand for robust integration solutions. Shift towards SaaS make it even easier for each company to use multiple services for their needs. And integration need will only expand.

Simpler and Better Integration Technologies: Integration technologies are becoming more user-friendly and powerful. REST API standard makes any connection between applications to be implemented as easy as writing a Python app. We will see AI-generated connectors sooner rather than later, reducing the complexity and time required to integrate systems.

Advanced Data Management Technologies: Data modeling and representation technologies are advancing. Graph models and graph databases for example, offer more flexible and powerful ways to manage relationships between data points, enhancing the capabilities of integration solutions. The solutions to build a holistic graph data model to support a digital thread will prevail, in my view.

What is my conclusion?

Despite theories of how one system (or platform) will dominant, the reality is opposite. Even small manufacturing companies these days are bringing multiple SaaS services to manage CAD files, organize product structures and bill of materials, connecting it to a CRM solution, MRP or ERP software, migrating their supply chain models from Excels to SaaS applications. It is unlikely that companies will shift to using a ‘single application’ for everything. I think, the battle for a single application is lost. The reality is how to integrate platforms. However, the next battle for enterprise companies is for master data, which I will discuss separately (stay tuned). Just my thoughts…

Best, Oleg

Disclaimer: I’m co-founder and CEO of OpenBOM developing a digital-thread platform with cloud-native PDM & PLM capabilities to manage product data lifecycle and connect manufacturers, construction companies, and their supply chain networks. My opinion can be unintentionally biased.