Process automation has become a strategic priority for companies. However, many projects continue to offer returns below expectations. Not because of a lack of technology or ambition, but because of a less visible problem: the integration between process engines and document systems.
Various analyzes of the sector agree on a relevant figure: up to 30% of the total cost of an automation project can be allocated to integrating the workflow engine with document systems and other corporate applications. This is not an exception, but rather a constant in projects where contracts, invoices, files or regulatory evidence form a central part of the process.
The reason is structural. “The majority of generalist BPM suites were designed to orchestrate tasks, not to govern documents. When the process depends on real files with versions, permissions, signatures, auditing and legal retention, integration is no longer trivial. Metadata must be synchronized, security logic replicated, guarantee traceability and ensure that the state of the document and that of the process are not misaligned. Each adjustment adds cost, complexity and risk,” explains Josep Llort, Chief Technology Officer of OpenKM.
Frictions between process and repository
This technical effort is rarely seen in the initial phases of the project. On paper, the flow is well designed. In practice, problems arise when the document changes version during the process, when a permission is not replicated correctly or when an audit requires reconstructing who accessed what information and at what time. In many IT departments, a relevant part of subsequent maintenance is dedicated precisely to resolving friction between the process and the document repository.
It is also worth clarifying a key aspect: a workflow does not always depend on documents. In many cases, files are made up of several documents, one or even none, depending on the type of process. Modern document systems manage structured information, metadata and business logic, not just files.
In this context, OpenKM proposes a different approach with OKMFlow, its generalist workflow engine. The difference is not in the nature of the engine, which is comparable to other process engines on the market, but rather in that the integration with the document manager comes standard, eliminating much of the usual effort in document automation projects.
Although OKMFlow and OpenKM are separate applications that communicate using REST services, being designed together significantly reduces common interoperability problems. The workflow can work directly with versions, metadata, permissions and auditing without the need to duplicate logic or maintain intermediate layers.
It is important to emphasize that the workflow engine is not limited to OpenKM. It can be integrated with other corporate applications and even with other document managers. Likewise, if the objective is to replace a traditional BPM platform, the integration capacity will continue to exist, regardless of whether or not there is a document manager in the architecture.
Native AI capabilities
The OpenKM document manager incorporates native AI capabilities to classify, extract and analyze information, expanded through MCP services that allow connecting external models. Being integrated into the platform, the workflow engine can use them directly within the process logic, enabling automatic decisions and content-based automation without additional developments.
Modern document systems manage structured information, metadata and business logic, not just files
When the objective is to orchestrate complex processes between multiple applications without a relevant documentary dependency, these platforms remain a valid option. However, in many organizations the critical processes continue to be documentary, and that is where the cost of integration skyrockets.
Reducing that 30% is not just a budgetary issue. It also has a direct impact on implementation deadlines, operational stability and regulatory compliance. “Less integration means fewer points of failure, less technological dependence and greater capacity to audit end-to-end processes,” says Josep Llort.
In a context where artificial intelligence is multiplying documents, versions and automated flows, this challenge tends to intensify. Automating without controlling the document layer not only makes projects more expensive, but also limits their scalability and increases operational risk.
“Experience shows that many automation projects do not fail because of the design of the process, but because of the invisible weight of documentary integration. Identifying this hidden cost and deciding how to reduce it is becoming one of the key strategic decisions for CIOs and those responsible for digital transformation,” concludes the CTO of OpenKM.
