COGNITIVE WORLD

View Original

How low-code platforms can aid intelligent business process management

Image: Depositphotos

Source: CogWorld Think Tank on VentureBeat

The potential for low-code/no-code platforms is enormous. Low-code increases the productivity of IT developers — sometimes by several orders of magnitude. And no-code empowers experts and subject matter experts primarily on the business or operations side (as opposed to IT) to become “citizen developers.

But as I explained in a previous article, low-code and no-code platforms are not a panacea; they face challenges.

Given the broad spectrum of low-code and no-code platforms, how should enterprises find the best options for their specific needs? And what are the use cases for using multiple low-code/no-code platforms?

I will address these questions in a series of articles to help you navigate this transformational landscape while avoiding the pitfalls. Specifically, I will be looking at low-code/no-code related to intelligent business process management (BPM), intelligent databases, automated integration, and a number of other areas.

In this first installment, I will be focusing on low-code/no-code in the context of intelligent BPM, or iBPM.

What is iBPM?

iBPM’s core value proposition is the collaboration and orchestration of people, applications, connected devices, and trading partners to achieve and continuously improve business objectives. Intelligence and automation are two essential conjuncts for BPM. Intelligence for BPM comes in many forms: digitizing business rules, intelligent virtual assistants, and increasingly process mining. A BPM solution will involve fully automated robotic process automation sub-processes for repetitive tasks that do not need human intervention and automated tasks assigned to human participants. Thus, increasingly RPA is becoming part of the complete intelligent BPM platform.

Here is a simple order-to-cash process example:

Some tasks will be performed by humans — for instance, approving the orders. Others could involve automation with RPA — for example, receiving the goods. There will also be tasks accessing systems of record — for instance, preparing and paying the invoice.

An iBPM platform will model, execute, monitor, and improve the end-to-end process. Other terms are also often used to denote end-to-end processes. These include “workflow” and “case management.”

Intelligent BPM is much more than technology. At its core, it is a transformational management discipline that helps organizations achieve their strategic goals. Automation is a crucial component of iBPM solutions. As a discipline, BPM drives the operations of enterprises. It includes several iterative phases from design to execution to monitoring and continuous improvement.

There is a remarkably close affinity between low-code/no-code and BPM. As far as back in 2005 or earlier, BPM suites were touted as platforms for model-driven development, which is akin to what we now call low-code/no-code. What were the “models?” Well, check the next section on how low-code/no-code manifests itself in iBPM platforms.

Low-code/no-code in iBPM

Low-code/no-code iBPM platforms handle:

  1. Modeling the workflow or the process

  2. The user experience or screens for the human participants, and

  3. The process analytics dashboards — for continuous improvement.

There are other components of a complete iBPM low-code/no-code platform — such as the decisioning (aka business rules), integration, and data model — but I won’t be getting to those in this post.

The following is a simple purchase request process model from Bizagi, using shapes from the BPMN graphical notation for business processes (the de-facto standard):

The swim lanes represent the participants in the process. The rectangular shapes are tasks or activities. The diamond shape is for a decision, and the circles represent the start and end of the process. There are many other shapes in the BPMN standard, but these three are the most common.

If human participants, such as Boss, Requester, etc, are involved in a particular workflow, the low-code/no-code BPM modeling also supports the creation of UI forms to enable that interaction, and these are pretty easy to model. This “drag and drop” paradigm of building user experience is common and similar across all low-code/no-code platforms that support Web or mobile applications. The following figure illustrates a simple user experience designer from Kissflow. There will be elements such as buttons, input fields, drop-downs, images, etc., that a non-technical developer can use to create the user experience. The elements are then connected to the properties or fields in the business process being modeled and automated.

The interface builder of the iBPM platform is robust enough to allow the designer to build a user experience — preferably without any code.

Once the application is deployed, the various participants can then monitor the performance of the activities through interactive analytics. These are actionable analytics dashboards, which means that if there is a bottleneck or issue, the stakeholder can take action, such as escalating or re-assigning a task. The analytics dashboards will typically have pre-built analytics that also support low-code/no-code customization. Here is an example of an actionable business process analytics dashboard from Nintex:

iBPM low-code/no-code recommendations

Why is it important for organizations to be able to model, automate, monitor, and improve their business processes without coding? An organization is a collection of business processes for production, marketing, sales, service, and support functions. So any optimizations and improvements of the most critical processes will enhance the bottom line: cost savings, revenue generation, and compliance. These are called operational excellence (OE) improvements. iBPM low-code/no-code platforms are an enabling technology for OE.

Here are my recommendations for iBPM low-code/no-code.

Prioritize your improvements: There will typically be many mission-critical and support processes that need improvement. By balancing the complexity of implementation with business value, you will identify the low-hanging fruit. (For more details, check out this explanation of four intelligent automation methodologies). The result will be a list of automation and OE business processes that you can optimize through an iBPM low-code/no-code platform.

Make sure you start with process mining: To find the top priority low-hanging fruit, you need to know the most common process paths, the bottlenecks, the variations, and improvement opportunities. In other words, you need to understand what processes your transactional data is subject to and then improve them. That is precisely the domain of process mining. Do not automate bad processes. The figure below illustrates the OE reference architecture with iBPM low-code/no-code. At the bottom, you have the systems of record that generate the transactions for specific processes. After aggregating and cleaning the transactional data, a process mining tool — such as Celonis — can then identify the most common process path and the variations and the root causes for the issues. Like data mining, process mining algorithmically “mines” and discovers the processes from the transactional data, including the variations and bottlenecks. Based on these, a iBPM low-code/no-code platform is used to improve, implement, and automate the processes, leveraging workflows with human participants and robotic process automation.

Create and fund an operational excellence competency center: iBPM low-code/no-code — and all other low-code/no-code, for that matter — is technology. As noted above, it is also a management discipline for operational excellence. For organizations that use this approach, it is a good idea to have a competency center that does three things at a minimum: balances innovation through iBPM low-code/no-code with best practices for security and reliability, enables non-technical subject matter experts to leverage iBPM low-code/no-code and become participants in development, and governs the continuous improvement from process mining to automation.

Understand the landscape and leverage experts: There is quite a bit of confusion when it comes to classifying what solutions are BPM solutions. Some analysts classify these platforms as “workflow,” “business process,” or “case management” solutions. For example, see these classification schemes:

There are also low-code/no-code development platforms that are closely affiliated with the BPM space but that might be classed into other low-code/no-code categories:

The low-code/no-code ecosystem is constantly evolving. There are hundreds of platforms — and new ones are entering the market all the time. Sometimes inexpensive and straightforward low-code/no-code tools will be sufficient for your needs. Do not pay for what you will rarely use. Also, avoid vendor lock-in. There are emerging new and innovative low-code/no-code platforms that support plug-ins and add-ons, including those that address process mapping.


Dr. Setrag Khoshafian is a cofounder at Startup Assistant and Principal and Chief Scientist at Khosh Consulting. He was previously VP of BPM Technology at Pega, Senior VP of Technology at Savvion, and CTO at Portfolio Technologies and is a member of the Cognitive World Think Tank on enterprise AI.