Welcome to part 1 of an ongoing series of blog posts demystifying "process intelligence" - the terms, the technologies, and how and why process intelligence can help improve your business. For our first blog post we will be looking at the original process discovery methodology - manual interviews - and why this historical approach to understanding processes and workflows is probably not the best tool for the job.
Company leaders have long looked to manual process discovery when they were tasked with transforming a business process. The idea goes that when you bring external expertise and perspectives into a process redesign initiative, you get an independent evaluation of what you’re doing. You also get fresh ideas that may help the process better meet its goals and objectives.
So, what is manual process discovery? What does it look like? And most importantly, why doesn't it work?
When enterprises explore manual process discovery, they often start by hiring a consulting firm that conducts an internal business assessment. Some companies may look to inside resources instead, like an internal audit group or business process excellence team. Whoever conducts the manual process discovery, the team usually starts with walkthroughs with the workers who execute the process. They may then directly observe the process and perform some testing, to validate its design and execution.
Business analysts then take information from the interviews, observations, and testing and build process narratives and maps. These as-is process documents become the starting point as they set out to design a to-be process that will deliver more efficiently and effectively by implementing new best practices and renewing commitments to existing process requirements.
So, where does manual process discovery go wrong?
Manual process discovery binds you to our limits, as people. Humans cannot always put our thoughts into words when we describe how a process works. We sometimes miss the “big picture” and focus on the island of what we do instead of the ocean of a whole process.
Also, manual process discovery takes time away from people who would otherwise perform the process. That puts pressure on budgets, deadlines, and other company initiatives. Lastly, by the time manual process discovery delivers its results, those results may no longer be relevant or even useful. Let’s look at four shortcomings of manual process discovery in greater detail:
“We know more than we can tell,” says Polanyi’s Paradox. This cognitive phenomenon explains that while humans know how to perform a task, they may struggle to explain why they perform it, or why certain steps exist and in a certain order.
In manual process discovery, examples of Polanyi’s Paradox can surface when we ask a claims handler to explain how they identify red flags that require fraud unit investigation or how they can sense when a direct-deal claim may be slipping into a contested claim with attorney representation. In a customer service process, an example of Polanyi’s Paradox could emerge when a customer service representative explains the steps they take to calm an irate customer. Even though experienced employees may intuitively understand how they perform certain process steps, they may struggle to explain why they perform them, or how they may adapt them—based on how an interaction unfolds.
Polanyi’s Paradox says that humans are imperfect reporters. When interviewers conduct manual process discovery, they often come back with incomplete details.
When you talk with a worker, you get that worker’s description of their role in the process. And when you talk with another, you get the next step and so on. In insurance underwriting, that may mean learning about how a risk submission arrives in the office, how a data entry person enters the risk into the system, and how the risk gets analyzed, in three distinct walk-throughs.
But, what happens in between those steps, or if an exception comes up? Would a different worker describe a different version of the process? How do you capture procedural differences (process variants) among a team of workers performing the same function?
When you’re performing manual process discovery, your observations may only be as good as the walk-throughs you have the time and resources to complete. Due to time and logistics, interviews often involve only a small subset of employees who provide narrow snapshots of their role in a process.
You hire claims handlers to handle claims, underwriters to underwrite, and appraisers to appraise damage and repair costs. When they sit down with an interviewer charged with completing a manual process discovery, they lose productive time that could have gone toward completing their work responsibilities.
The fact that manual process reviews often target a function’s most experienced, knowledgeable workers exacerbates this shortcoming even more. Interviewing employees takes them away from their day-to-day jobs and makes it harder for them to fulfill their responsibilities.
Consulting projects can easily take months to complete. Walk-throughs happen early in manual process discovery and the static information that reviewers gather from your workers may be months old before it can drive process insights and changes.
By then, new people, priorities, or regulations can completely shift a claims or underwriting process. Static data and changing processes can lead to outdated analyses, findings, and observations by the time that the results of the manual process discovery are ready to be shared with process stakeholders.
Wherever you are conducting your process discovery, AI can offer new ways to reimagine your business processes.AI’s insights can lead to a deeper understanding of processes, a more accurate view of how your process really works and how the work does (or does not) get done, and help leaders reach a higher level of operational intelligence and in less time than manual process discovery requires. When you set about revolutionizing a process, the first step is to understand it—and all of it.
Manual process discovery was a great tool that’s had its time. Today, business leaders see the need to explore and embrace the innovations that automation and AI promises. New tools are available to model, simulate, and measure as-is processes and to-be processes without losing so much time to obsolescence or inefficiencies.
Today, the key to success is adopting a data-driven discipline to your process optimization efforts. But the question is...where does one start? Skan’s actionable Process Intelligence Playbook explains approaches to process discovery, how to get started with process intelligence, and illustrates process intelligence in action through real-world applications. Check it out today!
“One of the things you don’t ever want to do is to automate a bad process. You are just going to make bad things happen faster, and that is not what anyone wants.”