Over the last seven years, I've led and participated in business process management initiatives in Fortune 500, start-up and mid-market companies. Every project teaches you new things, and one of the best places to start learning is by assessing the knowledge base of your audience.
Once, as I was preparing to lead the first ever Six Sigma Green Belt training for a suburban Philadelphia software company, one of the executives told me, "You know, I don't think any of the attendees know what a process actually is. Before you go too far, you'll probably need to define that for them."
The word "process" is used so generically in business, policy and political circles, it's easy to assume the meaning is universally understood. But I heeded the executive's advice, and have kept it in mind ever since. After all, achievement often requires prerequisites. Therefore, before teaching the best ways to manage a process, it's only appropriate to first clarify and confirm what a process actually is.
A process is a systematic and repetitive series of steps, actions or operations used to achieve an outcome. It contains a starting point and a stopping point, with multiple steps in between (in sequential order), transforming one or more inputs into outputs fulfilling a need or requirement. Starting and stopping points determine the scope (or boundaries) of your process, i.e., what's included and what's excluded. It's critical for management to understand these boundaries, in order to identify where one process feeds into another, and to prioritize which processes need immediate improvement.
Organizations strive to make their processes both effective and efficient. But for this to happen, business processes must be identified, defined, mapped, owned and managed. An effective and efficient process is measured continuously, and updated as needed. Failure to do so reduces a process to a mere sequence of steps that lose efficiency over time. Put another way, a process that's not measured continuously is like setting a car on cruise control then falling asleep at the wheel. Everything is fine, until you have to change speed or direction.
In order to address the various needs of customers, it's critical to measure different levels of process. However, developing a variety of stand-alone processes to address different customer requirements often creates inefficiencies. Instead, there should be one overarching process for conducting root cause analysis with the flexibility to address the needs of different customers, through structured sub-processes that are measured, communicated, repeatable and reliable.
For most organizations, a process management program evolves over time. A common path might look something like this:
Step 1: No deliberate process controls or measurements in place. Ad hoc solutions can rarely be replicated. Panic button reactions when process outputs fail. Guesswork, finger-pointing, and band-aid solutions are typical.
Step 2: Spotty, fragile, fragmented process or quality controls exist. Early controls are being imposed on an ad hoc system. Unstable processes are restored to steady state by troubleshooting techniques.
Step 3: Function-based process management implemented via process control techniques and improvement teams managed "top-down." The sense of process for its own merits begins to emerge. Systematic methods are developed to integrate new people into process disciplines, but inconsistent alignment across functions and sub-processes remains.
Step 4: Cross-functional process management and controls aligned to process goals and linked together. Information matures to the point where it can be trusted to drive process management. Planning and decisions are based on "Voice of the Process" facts and data.
Step 5: Process-centered organization, rather than functional hierarchy. Process dashboards used to operate the business, aligned with current strategy. Continuous reengineering of all strategic business processes - and their connecting interfaces. Dramatic performance improvement targets. "Process" and "out-of-the-box" thinking and implementation are pervasive. An institutional shift from reactive to proactive improvements is centered on defect prediction and prevention.
The road through these steps is a journey typically measured in years, because it requires structural and cultural changes to an organization, many of which are done through trial and error due to resource constraints and executive tradeoffs between short-term and long-term priorities. Change of this magnitude can easily fatigue an organization, unless it has clear, consistent, sustained and enthusiastic sponsorship from chief executives capable of repeatedly communicating "big picture" objectives throughout the entire company.
Executive sponsorship is the ultimate hammer needed to pound away whatever opposition may exist against a Business Process Management (BPM) program. Why would anyone oppose a program intended to improve company profitability and individual productivity? Because a good BPM program attempts to quantifiably prove or disprove conventional wisdom through systematic methods of root cause analysis, and people are generally wired to pursue and accept anecdotal evidence ahead of objective measurement systems. How many people are wired this way? Offhand, I'd say roughly 73.3%.
According to inferential statistics developed by Isabel Briggs Myers (who first created the Myers-Briggs Type Indicator personality assessment questionnaire with her mother Katharine Cook Briggs in 1962), approximately 73.3% of the U.S. population has a personality type combination where "Sensing" (i.e., interpreting the world through your five senses) is a more dominant trait than being "Intuitive" (i.e., interpreting the world by recognizing seemingly abstract trends and patterns). If you accept that only 26.7% of the general population is predisposed to embrace an analytical BPM methodology, then you begin to understand why change management of this magnitude takes so long to become institutionalized, and why so many companies falter along the way.
If you don't grant this premise, then you should probably provide baseline data supporting your hypothesis. While we're at it, let's also take a look at your data source, sampling method, degrees of data normality and measurement system variation, as well as your experimental design, testing results and control plan. In other words, show me your process for reaching this conclusion. Hopefully, at least 26.7% of you know what I'm talking about.