At the request of Keith Swenson, I created a SlideCast of my presentation from the Process.gov event in April 2010. The basic premise is that Adaptive Case Management is as much about approach as it is about technology. I’d love to hear your feedback in the comments section below!
The past couple weeks have been all about three P’s; Predictability, Practicality, and Passion.
You might have heard that the new Adaptive Case Management (ACM) book, “Mastering the Unpredictable,” was launched last week. One of the primary tenets of ACM is that knowledge work is unpredictable. This has generated a surprising amount of discussion and, in many cases, disagreement. Keith Swenson asked recently for every blog to have a glossary, which is a fantastic idea and one that I’ll get around to in the future. For now though, let me clarify what I’m saying when I use the term unpredictable.
Here’s what Reference.com defines predict:
–verb (used with object)
1. to declare or tell in advance; prophesy; foretell: to predict the weather; to predict the fall of a civilization.
–verb (used without object)
2. to foretell the future; make a prediction.
And using that, here’s how Reference.com defines unpredictable.
1. not predictable; not to be foreseen or foretold: an unpredictable occurrence.
2. something that is unpredictable: the unpredictables of life.
Doesn’t that second one violate the whole rule about not using a word to define itself? In any case, the root word, predict, is what I’m focusing on here, and the first definition is the most important. The reality of many business processes is that you simply cannot predict with a high degree of certainty the order of events. Yes, you can approximate the order of the “happy path”, the path most often taken, in some scenarios. And yes, you can likely come to some distilled version of the process through a lot of compromise. But in the end, many of the problems that knowledge workers face need to be flexible enough that trying to predefine them via a process map is an exercise in futility. I say this not because I don’t believe in structured processes, but because I believe in the right tool for the right task, and structured processes are often too rigid. I’m sure this concept will get some folks up in arms, which is fine because it leads into our next P.
What’s been fascinating about watching discussions around the nature of knowledge work is the degree of passion and conviction people display. I referenced a prime example of this between Keith Swenson, one of the co-authors of the ACM book, and JJ Dubray here during my talk last week at Process.gov. I’ll admit that I lost the plot a bit during some of the more theoretical / academic aspects of the conversation. With that said, what I took away from the conversation was two things. First, the fundamental disagreement was around whether it was possible to model the entire universe as a series of states and transitions, effectively predefining and predicting everything (more on this one in a moment). And second was that peoples’ beliefs around approach and theory are deeply rooted and contradictory opinions can evoke very strong responses. Hence, Passion.
Passion is good, because a strong conviction and excitement can be contagious and can help spread knowledge. The most articulate people I know aren’t simply good communicators because they speak and present well, but also because they are passionate about the topics on which the speak. This is one of the reasons I am active in the ACM community, because I feel strongly that applying ACM can make a substantial impact on how companies do business and how people get their jobs done more effectively.
Where passion goes wrong is when belief in a singular concept or approach drives people to close their eyes to alternatives. I say this not thinking about process improvement alone, but also about life in general. I’m reminded of a favorite book that I read a few years back, “Who Moved My Cheese” by Spencer Johnson, M.D. It’s not really about passion per se, but rather about change in general. There’s a preview of the video for the book here that’s worth a quick watch if you haven’t read the book (or even if you have).
My point is that as ACM evolves, peoples’ world views will change. Folks that have invested significant time and energy into things like BPMN and various process modeling tools may not immediately gravitate towards ACM because at some level it requires a shift in approach. We won’t all agree all the time, but my hope is that over time, these discussions will help people see the value that an ACM approach can offer. Which brings me to the final P for today, Practicality.
I’ll be posting a screencast of my presentation from Process.gov in a couple of days, but one of the themes I focused on was practicality. Through all of the discussion of predictability and structured / unstructured, there is one theme that I see that I think is most important. It’s simply not practical to predefine every possible choice or event for complex business problems. Sure, you probably could get to a fully predefined, rules-driven process map, but at what cost? My preference is to structure and control what can be, but provide capabilities to deal with the variability of business instead of spending months of time and excessive amounts of money to implement a “perfect” system that will, in all likelihood, soon be out of date. More on this concept in my slidecast, which I’ll post later this week or early next.
So, with that, thoughts? Comments? Criticisms? Suggestions or experiences?
This week we launch “Mastering the Unpredictable” at the Process.gov event in Reston, VA. The official site for the book is live at www.masteringtheunpredictable.com, and you should be able to order the book on Amazon shortly. A full description of the book can be found on the site, but here’s a snippet:
The facilitation of knowledge work or what is increasingly known as “Case Management” represents the next imperative in office automation. The desire to fully support knowledge workers within the workplace is not new. What’s new is that recent advances in Information Technology now make the management of unpredictable circumstances a practical reality. There’s now a groundswell of interest in a more flexible, dynamic approach to supporting knowledge work.
The collection of authors represents a broad cross section of industry experts in the fields of Adaptive Case Management and Business Process Management. The foreword for the book was written by Connie Moore, Research Vice President, of Forrester, who states that “I think a sea change is coming in the process world.”
The chapter I wrote is titled “Moving from Anticipation to Adaptation” and discusses the fundamental shift from predefined business process models required by conventional model-centric BPM to a new world of adapting to business in real time. Here’s the official description of the chapter:
Using examples of work from an insurance company, the qualities of emergent processes are examined to find that they are constantly changing. To handle this, tasks should not be rigidly fixed in an immutable process definition, but instead should be planned as the work proceeds. The planned tasks act as a guardrail to keep you from going off the road accidentally but can be changed as necessary during the work itself. This is the essence of “adaptability,” which guides work and allows the plan to be modified at any time, but it does not enforce a particular pattern.
My colleague Dana Khoyi, Vice President of Development at Global 360, also contributed to the book, writing a chapter “Data Orientation” as well as co-authoring a chapter “Templates, not Programs.”
A special thanks to Keith Swenson who singlehandedly shepherded this project through to completion. Without Keith, none of this would have been happened!
Stop by the launch event in Reston on April 14th and say hi!
Like everyone else, I have a perception bias, and that leads me to see the world of BPM through the looking glass of Case Management. Maybe it’s because I’m a slightly unstructured sort of person and because I tend to do multiple tasks in parallel. Whatever the case (no pun intended), I see significant value in capturing many business processes through a CM style of BPM. Despite that bias, I can objectively say that there’s been a lot of interest in Case Management lately (see Connie Moore’s post here as one example).
But it hasn’t been limited to Case Management. There’s also been a lot of talk about Dynamic Business Applications and Unstructured Business Processes. And before all that there was Collaborative BPM. Clay Richardson makes some great points about the intersection of Social Media and BPM here. His example of Google Wave tied to a process instance is fantastic.
I believe we’re hitting a point where companies are becoming more interested in improving business processes that aren’t quite as straightforward and well defined. Further, I believe that whatever name you choose to describe those processes, dynamic, unstructured, collaborative, adaptive, etc., companies are looking for solutions that can solve those problems. I for one am excited, because it means that we’re getting to the “hard stuff”.
Think of all those meetings you’ve sat in arguing about how a process should behave. Did you have to compromise in the end, sacrificing everyone’s position to find a common ground? I’d bet in many cases you did. I’d argue that having to compromise like that just creates more problems. It forces people to find workarounds because the solution isn’t really a solution at all. And I’d say that in many cases, that compromise was unnecessary, because the right solution would have let you deal with the “art inside the science”. For some types of business problems, specifically those that involve human judgement, we need to trust in the people that do the work, more often than not they know what they’re doing. Provide them with guidelines and best practices, sure, and provide a mechanism to see how they’re doing (performance metrics, reports, etc). But give them the flexibility to adapt to the real world as it happens.
It’s an dynamic world out there, and life happens. Don’t you think we should accept that the same can be true of the processes our companies follow?
Interesting article by David Mitchell on Business Productivity at CIO Today. David leads with some good statistics on worker productivity and costs, and moves into a comparison of the BPM industry with the Discrete Manufacturing industry in the 1950’s.
One of my favorite bits is as follows:
That’s why I believe that business process improvement today has hit a wall. When it comes to process management initiatives, organizations today over-respect the importance of process automation –- how work moves through an organization -– and under-respect the contributions of workers –- how work gets done.
The BPM industry has spent the better part of the last several years making better and better “modeling” tools, but how many people does that really help as a percentage of the total user population? I’d argue pretty low, so the overall productivity gains are limited.
In any case, good article, one worth checking out.
As those of you I’ve worked with know, the concept of Case Management is of particular interest to me. I believe it solves some very real, challenging business problems that conventional model-driven BPM isn’t well suited to handle. As companies get more sophisticated and begin to run out of “low hanging fruit”, the more complex and less structured processes are the ones they need to get under control. The challenge many companies face is not being able to describe those problems accurately, much less to implement a system to manage and monitor them.
At the June meeting of the Object Management Group, the possibility of a Case Management extension for BPMN 2.0 was a hot topic and one that resulted in a number of very passionate (ok, as passionate as software gets) posts on various blogs. The buzz around Case Management has been increasing steadily over the past year, and folks like Bruce Silver of BPMSWatch, Henk De Man of BPTrends, and Jon Pyke of Cordys have been talking and writing more about the benefits and challenges of Case Management in general.
One of the more interesting developments to come out of these discussions and the OMG meeting is Bruce Silver’s decision to start a community site dedicated to Case Management and BPMN. You can find it here or in my links to the right side of this page. Bruce has invited me to contribute to the site, which I’m excited to do.
More to come on this topic soon!
In my last post I wrote a bit about the perception bias that companies and people face when trying to solve a business problem, the idea that they view the problem with the filters of their experience in place. Experience can be a great teacher, but the potential downside is that this bias can lead to an attempt to solve problems with a pre-determined toolset or solution in mind. You’ll often see a manifestation of this in the RFP / RFI / RFQ process where it becomes clear that the author had a particular tool or technology in mind when writing the document. I also talked a bit about the feeling of déjà vu that I get sometimes when speaking to customers and prospects about many of the problems they’ve tried to solve in the past and still face today.
So where do we go from here? Good question. Clearly our approach and direction should be based on what we’re trying to accomplish, so let’s try and define that first.
We want to automate, and ultimately improve, the processes that drive our business. For a health insurer, we could be talking about an end-to-end process that spans multiple systems, for example Quote to Enrollment.
Ideally we’d need a graphical tool to capture the process, an execution engine to run it, and probably some simulation and optimization tools to improve it. We also need to keep track of all the content associated with that process including enrollment forms, customer and possibly group data, documents related to establishing identity, and miscellaneous other policy related information and documents. We’d want to be able to easily re-use processes, forms, and various “bits and pieces” without having to create copies or pre-define every possible combination or choice in our process. We certainly want to know all about how we are doing relative to our business goals and service level agreements we’re obligated to meet for our customers. And of course we want it all easily modified by business people who really understand their goals and challenges so that we can react to changes in our business rapidly.
If you think back to the perception bias of each of the solution providers I mentioned in my last post, you might realize that it’s not very difficult for any of them to say “I can meet most of those needs; after all, I have the best of breed Content / Process / Rules technology. We can deal with the others during implementation as a consulting effort”. For companies (in this case the health insurer) that are trying to solve their problems, it might sound slightly different. “Well, I tried this with an enterprise content management platform a couple years ago and it went ok, but not quite as well as we hoped. Maybe I would be better off using a Business Process Management Suite?”
In the example above, we’d like the modeling, simulation and execution capabilities of a BPMS. And we’d also want the versioning, content integration and audit capabilities of a Content Management tool. A Business Rules product would help us make the solution more flexible and might help us re-use components of our solution without significant rework. For the “how are we doing” part of the equation, we’re probably looking at some of the more advanced capabilities of a BPMS or Business Activity Monitoring solution. And it’s the combination of all of these components, implemented using best practices established by hundreds or thousands of implementations, that drives the flexibility and ability to make changes.
In short, we need a blended solution, something that brings all of the best of these solutions together. I’m not suggesting a “best of breed” approach where you bring in each of these solutions and stitch them together but rather finding a platform that incorporates most of these capabilities in a single configurable tool. The key word in that last statement is “most” because I don’t think any single vendor offers “best of breed” capabilities in a unified platform. I do believe that there are solutions that incorporate several of the key characteristics in a single tool, for example strong process and document management capabilities combined with robust modeling and analytics. The key is trying to find the right combination for your specific requirements.
The Insurance industry is a fascinating ecosystem. Not so much the products as the actual business model and problems the companies face. I’ve had the fortune to work with all manner of carriers (Health, Life and assorted other segments) over the past ten years, and been able to observe the similarities and the differences across those segments. I’ve always been on the outside, a solution provider looking in, trying to better understand the business problems facing the various companies I’ve worked with. And in that time I’ve worked for a number of different software companies spanning disciplines like Content Management, Configuration and Rules, and Business Process Management, each of which experienced some degree of success in one of the many segments of the Insurance industry.
Spend enough time in the software business and you might start to feel like Bill Murray in “Groundhog Day”, doomed to repeat the same day over and over. I recall trying to help a Health insurer find a way to simplify the enrollment process (both individual and group) back in the late 90’s with a configuration and rules solution. And then I was in the same situation again around 2004 but this time with an enterprise content management company. Not surprisingly, I discussed a similar business problem again recently.
In each of these projects there were challenges meeting specific functional needs. The challenges varied based on the focus of the software we were using to craft the solution. For example, when using a rules-based technology to define the enrollment process it was difficult to do anything more than track the status of a particular application using an attribute which limited visibility into the actual process. Using the electronic forms capabilities of most traditional content management solutions didn’t quite get us there because the enrollment wasn’t about the document itself but rather about the customer and all their supporting information housed in many other systems. And most pure-play BPM solutions ignore paper entirely and focus on the process definition, glossing over the challenge that the content itself presents. And these same problems apply not just for enrollment but for many other core processes as well.
The difficulties inherent in using any one of these individual technologies might seem fairly obvious in hindsight. So what was it that prevented the teams (both the customers and software provider) from knowing that in advance? I believe it’s a phenomenon that I think of as a perception bias, the idea that we all see the world and the problems we are presented based on our past experience and from within our sometimes limited frame of reference. The phrase “when all you have is a hammer, everything looks like a nail” comes to mind. This mode of thinking unfortunately forces a high degree of customization in order for the solution to be implemented.
Despite the hurdles each project team faced, all of the projects were ultimately successful. Maybe not as wildly so as the return on investment calculations predicted, but each one rolled out and enjoyed a period of success. Still, I’ve heard over time that some of those solutions have since been replaced, and in some cases replaced again, each time with a different technology solution. Which brings me back to the title of this post, haven’t we been here before?
I have a theory how we can avoid repeating today again tomorrow that I’ll share in my next posting. Until that point, I’d be interested in your feedback and thoughts.
When I think about what my first blog post should be, it’s tempting to fall back on cliched quotes like “The journey of a thousand miles begins with the first step.” The fact is though, as cliched as the quote might be, there is some truth in it. I have a concept for where I want this site to head, but as I’ve found many times over in life, planning is a great concept and one that the world laughs at. What I will say is that I’ll try and stick to topic, which is focusing on improving business processes. I can’t promise I won’t head down a side path now and again, as I tend to do that now and again. So, with the dreaded “first post” behind me, on to more interesting topics!