Process automation inside enterprise environments is becoming more popular than ever. Bernd Ruecker, chief technologist and co-founder of Camunda, shares three key benefits that process orchestration brings to automation evolution.
Market reports project that the category will grow 16% a year, up to $37 billion in 2030Opens a new window . In Camunda’s 2023 State of Process Orchestration reportOpens a new window , 93% of IT leaders surveyed credit process automation technology for driving business growth, and 91% expect to invest more in the technology over the next two years.
But under the surface, there are signs that automation is headed for trouble.
A vivid example can be found in the banking sector. All banks rely on automation to handle many processes, including account openings. But some banks carry out the tasks much more efficiently than others. New, digital banks have set up processes that allow customers to open an account in just a few minutes, completely online. But many conventional banks are struggling to get accounts open even in several days.Â
Why? In recent years, many traditional banks have primarily digitized documents and automated individual tasks, often without creating an overarching strategy and, above all, without focusing on the end-to-end process. Digital competitors have thought through the account opening process from start to finish and intelligently automated it in its entirety.Â
So although individual tasks, such as credit checks or comparisons with sanctions lists, are automated in both scenarios, traditional banks tend to have isolated solutions that are not connected to one another. This leads to faulty or interrupted processes, which of course, has an impact on the throughput time. In addition, processes in traditional banks often involve people, which leads to additional delays.
It may seem like automating a series of tasks should be enough to reduce the overall process time significantly, but it doesn’t always work that way. In order to improve throughput time, banks need to look at how those tasks are connected and make sure handoffs are working smoothly. This allows for optimizing the end-to-end process, as some tasks can run in parallel, but also makes sure all complexities are properly taken into account.Â
For example, many tasks can only be triggered after previous ones have been completed, and the process must be able to deal with exceptions, such as terminations or subsequent changes. So if processes aren’t mapped end to end, with individual tasks meshing seamlessly, the overall automation initiative will fall short of the bank’s goals. And that will disappoint customers.Â
These scenarios are playing out across industries. As consumerization takes hold, customers expect more, and companies react by pushing forward with automation initiatives designed to provide better, faster service. As organizations look to automate more tasks, a number of factors have made it more difficult to wring true value out of process automation.Â
See More: Building a Network Engineer-friendly Approach to Automation
Three Hurdles in Automation
Here are three pain points enterprises are facing as they try to master the art of automation.
1) Too many endpoints need to be connected
Every process is made up of a series of individual tasks. Each task is executed by an “endpoint.†These endpoints include a diverse mix of people, systems, and devices – which all need to work together to perform the end-to-end process.
It would be much easier if a set of processes was carried out fully by a single endpoint, like a plant foreman or a back-end ERP system. But endpoints are diverse, varied and interconnected. You might have a process that moves from a human task (like a manual approval workflow) to a software-driven task to a set of system tasks (like microservices or SaaS applications). Then, at several points along the journey, additional steps would be triggered by a handheld scanner or IoT device.Â
The number of endpoints is exploding. According to the State of Process Orchestration report cited above, a typical automated process uses 5.68 different endpoints. These include everything from customer portals, package enterprise applications and mobile apps to RPA bots, chatbots and telco voice interactions.
Endpoint diversity can lead to more fragmented and disjointed silos in business processes. These disconnects can, in turn, make it difficult to gain enough visibility into overall processes to measure success. They also make it harder to change end-to-end processes since a change in one area would require changes across many different endpoints.Â
2) Processes themselves are becoming more complex
Not only have the number of endpoints increased – processes themselves have become more complex. This trend was reflected in the State of Process Orchestration report, where 72% of IT leaders said that processes in their organizations are becoming more complex and harder to maintain. They cited a long list of issues that contributed to the complexity, including having to deal with human logic, branching and/or conditional logic that spans complex business rules, and having sub-processes or systems being owned by another team.
Processes themselves rarely follow a straightforward sequence of steps. They don’t follow a linear pattern; they incorporate numerous tasks (endpoints as described in 1.), some running in parallel while others either wait for changes in events or business conditions. To operate smoothly, activities need to be able to be interrupted at any time. These processes are difficult to implement, hard to scope and complicated to define among multiple stakeholders.Â
Individual steps can be automated by tools (such as RPA bots) or simple integration flows (moving data from one application to another using an iPaaS tool.) But, complex processes follow advanced workflow patterns, which require a more advanced level of process logic.Â
3) It’s hard to visualize and constantly improve end-to-end processes
While enterprises clearly have a lot of process automation under their roofs, many don’t understand what’s going on where. In the State of Process Orchestration report, 68% of the surveyed IT leaders say they don’t have a good way to see all the processes in one place. This can have disastrous effects. Not knowing what processes run within a company, how these processes connect to each other and where bottlenecks occur can slow the business down. Having processes run haphazardly without monitoring them and improving them constantly can put the business at risk.
Nearly three-quarters of respondents to the State of Process Orchestration survey say disconnects between teams cause wrong things to be built and rolled out. It’s important to be able to communicate many different aspects of a process model. For a process to complete, pieces might have to close in short order. There might be SLAs around them, and team members may need to figure out quickly where processes are stuck, incomplete or on hold.
Using a standards-based language such as BPMN to design and monitor processes, IT and business teams can not only get more visibility into processes – they can work together to make processes better. Speaking the same language, they can communicate and collaborate more effectively. That gives them the ability to monitor processes and then update them if regulations or business conditions change.
Future-proof Transformations Need Orchestration
As processes become more complex and more endpoints get integrated into the end product, organizations need to be able to holistically coordinate all of these processes. What’s needed is a set of technologies that can orchestrate the end-to-end processes that are automated throughout an organization.Â
Orchestrating processes establishes continuity across all process endpoints – whether it’s being originated by a person, system, or device. Orchestration can bring together disparate, siloed processes to ensure that they operate in a coordinated, predictable manner. It can also create more visibility into process effectiveness, giving organizations better oversight and more ability to optimize processes on an ongoing basis.
How are you targeting the pain points of automation with process orchestration? Share with us on FacebookOpens a new window , TwitterOpens a new window , and LinkedInOpens a new window . We’d love to hear from you!
Image Source: Shutterstock