< back to Expert Forum overview
The second BPM Expert Forum took place on May 4, 2021, with Gero Decker and Jan vom Brocke.
1. Perhaps the question should not be about modelling BUT about the “TYPE of models”. And perhaps the specific problem is with BPMN… because its rooted largely in Scientific Management principles?
[GeDe] Yes, I couldn’t agree more. Depending on what kind of question you want to answer and what you want to transport, it drives the need for how you want to model, how detailed and what aspects. There’s a lot of BPMN bashing and BPMN has a lot of deficiencies, but it’s still the best that we have. There are many tools available, content, trainings, and a lot of people knowing how to interpret BPMN. Therefore, there’s a lot of inherent value in having an open standard, and everything that would come as a replacement needs to overcome that threshold of adoption. And I’m looking forward to seeing better examples. CMMN was an attempt, and the only attempt that I’ve seen in the last 10 years, but it failed. Maybe discovery graphs of process mining might supersede BPMN at some point, but I don’t see that either, because they’re typically not rich enough. They’re good for exploration purposes, but for target state design purposes, not usable.
2. 10 years ago model-driven software development promised a lot. How do you see the roles of process models today in driving the execution of the processes?
[GeDe] It’s more important than ever. BPMS was the first attempt that didn’t get much adoption back then. Now what you see with RPA, low code and workflow companies like ServiceNow are picking the ball up, companies like Microsoft picking the ball up with their Power Automate platform. Process modeling to drive execution is more prominent and more visible than ever before, so I only see a continued rise in the coming years. We’ve never seen so much like what we have today.
3. Gero, a clear vision that I share perfectly. I have at the University an academic license from Signavio so that my students can work on process modelling. Are there any news from Signavio/SAP for the future on this academic bet?
[GeDe] Of course, we will continue everything that we do on the academic side, and we will even revamp that. SAP is much stronger in terms of academic alliances support, providing content and supporting lectures and professors out there. We’ve started work with them and we’re revamping our activity as we speak. For the winter semester, we should have a ton of cool things to offer.
4. How you would argue with those people saying that modelling is quite expensive? What about SME? I agree that modelling is important also for them, but how convince them to move in such direction?
[GeDe] I can answer with anecdotes. We have tons of SME customers and I know so many examples where within two or three hours of using modelling, people could unlock many hundred thousands and sometimes even millions worth of returns and insights just by talking to each other for the first time, and creating transparency. Not everybody in an organization is process minded per se. The process perspective in and of itself oftentimes unlocks insights and Aha moments that people otherwise don’t have. So yes, for an SME it’s not a luxury. It’s a luxury, if you want to do a full documentation effort, and that’s an unnecessary luxury even for large companies, to be honest.
For example, there was a TELCO company in Alaska, and they were completely brainwashed with project management. For them, everything looked like a project, so they treated customer onboarding, new clients, corporate clients onboarding as a project. They applied project thinking to it, but not process thinking, so very quickly they found that 80-90% of what they’re doing is always the same and super simple, doesn’t require the involvement of 10 people, but it can be done by one person once you agree on what should be done or what the area should be where you ask for additional help, the types of exceptions that you want to escalate. Everything else can be handled straight through by one person. That wouldn’t have been possible without people sitting down and agreeing that this should be the case, because as long as they don’t agree they always need committees of 10 people, signing off stuff. Just as a simple example.
5. Thanks for the inspiring words Gero. Can you tell us, in your opinion, some challenges about Business Process Modeling that you think are still there and that require attention from the community?
[GeDe] Yes, you should know how process modeling works and how it ties into a process improvement or process redesign exercise, but don’t get hung up with the methodology or the modeling per se. It is a mistake that we often see that people are just so fond of describing stuff or mapping things out. They lose sight of how do they actually attach themselves to the center of power, or the center of the point of gravity impact in the organization. Oftentimes it’s connecting the dots. For example, some people are fond of customer experience and process management, and just understanding the process and making a 2% better, and removing obvious friction. While they could use the same time to attach themselves to something much more important for the company and suddenly, they can unlock something much more impactful: It’s about understanding the context and attaching yourself to the things that really matter. That’s probably the number one guidance.
6. Any suggestion for developing a process model for Scalable Control & Acquisition Data in a Machine to Machine IOT Network?
[GeDe] I don’t know.
7. Agree that process models have really been useful across industries and not limited to digital innovation, but don’t you think once models are created they limit/restrict thinking and do not allow for flexibility that should be possible within a process?
[GeDe] Once they are created, they limit and restrict thinking and [trying to include] a lot of flexibility that should be possible within a process. That’s completely a misunderstanding of how you should use a model. There was this great quote by Jeff bezos, the founder of Amazon. He said, the process should never own you, you should own the process, because it’s a powerful tool to bring great customer experience to scale. So the number one thing you should always do with a process is question it, but never reject it in the sense of we don’t need a process. The question always should be how a better process looks like. So, back to your question - where once we’ve created a model it’s outdated, yes, that’s true if it’s only used in a moment of time, but if it becomes part of your DNA in terms of the way we think about change, it comes back to what we need to revise, together with the assumption that nothing is set in stone, never, and things are meant to change. You have a vehicle to define that change, to communicate the change, and to make that change happen. The process model becomes a super powerful tool and, yes, it has to change. One trick is to blend data driven approaches with people. Something that we promote a lot. Because then very naturally, you avoid for those processes that kind of disconnect between model world and reality. You should really see it as a tool that gives you superpowers, nothing less.
8. Where does BPM have to change and evolve to successfully support digital innovation – or is it all done already?
[GeDe] Very broad question, I mean the number one thing is always to remind ourselves that we need to attach ourselves to what matters, and so this notion of how does success look like, what does good mean, this is something that traditionally in the BPM community has been underrepresented, to be honest. Too many people jump too far, too fast, into how do we solve it, than fully understanding the problem.
9. Hello Prof Gero. thank you for the insight. What is your opinion about process model in less structured business process?
[GeDe] Answered in forum. Please view the recording of the forum.
10. Any inputs on how to translate Business Models (strategy) to operational business process models; is it a one step of more step translation? Also, how can we map Customer journey concepts into BPM concepts?
[GeDe] As with most things in life it’s an iterative process. If you have one target picture that you want to achieve in terms of an experience that you want to create, for instance, you have a million options, how to translate it into operational structures. Automated, non-automated, batch, individual, outsource, insourced, digital, analogue - there are so many permutations, and it’s not a function in the sense of there’s only one possible solution, but there are many. You have to be aware of that, make it transparent and lay different alternatives on the table, so you can decide whether it matches up or not.
11. Hello Gero, Jan & Jan, thanks for the great event today. From my point of view modelling was and will be the basis for every activity also in the current phase of digital transformation. Process models are the common language between the different teams of the organization. What we have to manage in future is the combination of process mining software robotic with modelling.
12. Gero, your basic message is that models are pictures to help transporting a message. I agree with that: that’s why pictures are important and that’s also why the modeling technique is not so important. So what is important in a process model to be useful for enabling the innovation/the change you talked about?
[GeDe] Most important, you must know to map out what are the outcomes that you want to achieve, and how you measure those outcomes in terms of success criteria and KPIs. Then you work yourself backwards regarding who needs to be involved, what needs to be done, what decisions need to be taken. But you always start with the outcome that you want to achieve and the qualifier for those outcomes.
13. Process models are often understood as a static version of reality. It seems that people sometimes have a hard time to think beyond this static description and think about new ways to do things. How can process models be complemented to overcome this situation and give people incentives to think something new?
[GeDe] Honestly, this was, I think one of the major contributions that we added as a tool vendor. To make it so easy for people to suggest things, to change things around. And this needs to be a living and breathing thing. It’s great if there’s a lot of engagement and iteration happening, and you want to have a quality indicator. It is not a good sign if something didn’t change for six months or a year. This is what many of our customers say - if we haven’t touched something for three months, it means that we haven’t challenged ourselves for three months, and this is too long.
14. To Gero: Wouldn’t you agree that focussing on process models place concrete to the inside-out perspective neglecting the (always changing) customer focus? Although you and others highlight the value of customer journey - but often as a pure lip service…
[GeDe] I agree, honestly, that process models and especially BPMN puts a lot of emphasis on what I do, and how I solve things, but it doesn’t place enough emphasis of what am I optimizing for. For example, BPMN doesn’t even require you to specify an outcome. It doesn’t require you to specify KPIs for it, but it requires you to specify activities. That’s very odd, right? So yes, the language places emphasis on probably less important things and so that’s why you should never see BPMN and its meta-model as the one and only truth, but it’s around building that practice. This is where the BPM billboard and many other things help to compliment the picture. BPMN solves a lot of the hard questions of defining semantics of very interesting things, and being very expressive, but it neglects some other things for sure.
15. What is the interface and relationship of Master Data Model and Process Mining. How can Master Data Model (Management) can help on Process mining?
[GeDe] There is such a strong relationship between process mining and data. Process mining relies on data. So the moment you’ve understood what kind of data you have available, only then you can do process mining. This is honestly not only a task on the users and the people who use applications, but it’s also up to the vendors big time, because they can expose so many things. We see it at SAP right now. You just talk to the right people and suddenly, they give you native process mining, support and products. This is something that never existed before, because suddenly people realize the value of analyzing processes on top of applications. If only a couple of engineers invest a couple of hours’ worth of work, it removes the friction for practitioners out there spending thousands of days, trying to work around the limitations of those systems. Things are changing, which is good.
16. Thanks Jan for the analogy of a car - means and end. I understand from an end user perspective it is ok and maybe better not to understand the intracacies but if we look at it from a mechanic’s lens wouldn’t he want to know the process flow - and hence the model?
[JvB] Sure, you are right and I think we discussed this already in the call a little. It’s a matter of perspective: of course, there also is the perspective of engineers. My point is that we use means serving the purpose of engineers while talking to c-level. It is important to find the right means serving the ends (and perspective) the people we talk to are interested in.
17. Will Jan’s slides be available to attendees? I have found them very thought provoking
[JvB] This is great, happy to share the slides (see here).
18. There is an “Industry” of management systems such as ISO9001 that is incredibly stuck with “representative” process models that look nice… How would you change/disrupt that? Create a “digital9001”
[JvB] This is a great idea. Changing things is always a socio-political endeavour. Very happy to be part of this. See e.g. https://www.iso.org/developing-standards.html, let me know if there might be a movement we could contribute to.
21. Question for Gero: In my opinion there is an important function of Future state process models: To visualise the dream to people, as a way of pull them to an future state. Some magnetic power.
[GeDe] The really powerful thing about process models is to be able to talk about future state. For current state, we have many different ways to talk about it, but the future state very few and process models fill their critical gap.
22. N/A
23. Thanks for this great topic, great event, and great panelists. Process models are highly related to BPMN. Does BPMN need any changes or improvements to assure future process models’ presence?
[GeDe] Well, for where it came from and what it was supposed to solve, that was to bridge the business IT divide and to have something that is rigorous enough to take things all the way to execution and formal interpretation. BPMN is great and continues to be great. Yes, if you ask automation vendors, they might have asked where BPMN falls short, at some of the more higher-level constructs or concepts, but then the question is, do we want to build that in? Should we intend to standardize customer journey models? If I just look at the history, for most of those more or less precise modeling languages, the standardization efforts have failed, because the requirement to be rigorous is simply not there that much. People do it differently, so BPMN is very good. It was invented for Signavio and Camunda to work together very nicely, right? Even if you don’t want to go down the path of execution, it just happens to be very useful there as well. It’s how it came about.
Long story short, I think where we should rather invest time is frameworks, like a business model canvas. There have been many others on this type of level. There is this innovation topic often: what’s the pain that we’re leaving, and what’s the gain we’re getting? These more high-level conceptual frameworks and also the BPM billboard go into that same direction. I think this is the type of level where a lot of contribution can be made, but the expectation should not be that this becomes a rigorous formally interpretable and verifiable model. Unfortunately that doesn’t allow you to write papers about deadlocks and BPM billboards.
24. When we talk about strategy: What would you think about integrating aggregated business processes and business model canvas Transformation patterns / Ideation patterns?
[GeDe] Some people have looked into this in the past. I still remember that a PhD student at HPI, for example, five years ago, looked into that business model canvas versus process models, but people always looked at this with the ambition to find consistency problems or some kind of formal relationships. I think it’s the wrong angle to look at it. It’s more of a kind of management type of question of what are the useful thinking patterns. If I see something, see a certain pattern in the business model canvas, what should I think about, what should my association be. What is a catalogue of things that have worked in the past versus things that haven’t worked in the past, and what was just a dead end. Those types of things I think will be very interesting, but it’s much harder to do research on it. That’s why it’s less popular than trying to find a deadlock, which is a much more easily solvable problem.
25. What is presented on the Billboard regarding context is based on the study presented in the article “On the role of context in business process management” and the bottom part is based on the framework of the delphi study presented in the article “Business Process Management in the Digital Age” ? Does make sense think about relationships between these two layers? Is it the idea?
[JvB] Yes. This is exactly the intent of the billboard to link prior contributions together and to make them accessable in an easy-to-use one-page representation.
The lower part builds on our six elements model.
The upper part, also relates to the ten principles of good BPM (specifically principles of purpose and principles of context).
Gero is co-founder and CEO of Signavio, a leading software provider for process management with more than 2,000 customers and almost 500 employees worldwide. Once the acquisition of Signavio by SAP closes, he will co-head the Business Process Intelligence business unit at SAP. Gero is a thought leader for Business Process Management and was named Innovator of the Year by MIT Technology Review. He studied Software Engineering and has a PhD in Business Process Management from Hasso-Plattner-Institute in Potsdam, Germany.
He states that: “Of course, digital innovation needs process models. People often get excited by the newest gadget, the cool new way of doing a certain task or taking an automated decision in a way never thought possible before. And such excitement is all good and great. But in the end, innovation is only valuable if it turns into operational and customer value. Process models are the linking pin for exactly that, to put things into context and see the bigger picture. Without proper integration into the operating model, the “innovation” might be cool but have no impact and rather turn into a burden. Therefore, the question is not whether we should use process models or not, but how we can best deliver ideas, insights and recommendations for digital innovation and foster a culture where experimentation and fast iteration is possible.”
Jan vom Brocke is the Hilti Chair of Business Process Management and the Director of the Institute of Information Systems at the University of Liechtenstein. Jan’s work has been published in many of the A, A+ and Financial Times Top 50 journals, such as Management Science, MIS Quarterly, Journal of MIS, and MIT Sloan Management Review. Jan has published over 40 books, including the International Handbook on Business Process Management and the book BPM Cases - Digital Innovation and Business Transformation in Practice. He has teaching experience from 26 universities in 13 countries, and has received over 30 international awards in recognition of this research, teaching and leadership. He is an invited speaker at and trusted advisor to DAX 30 and Fortune 500 companies and governmental institutions as well as digital start-ups across Europe (see: http://www.janvombrocke.com).
He states that: “The traditional focus on process modeling has not done BPM a favor in demonstrating its potential to support innovation. On the contrary, in many ways, conventional process modelling has hindered innovation. Digital innovation in specific thrives on exploration (as opposed to exploitation) and support is needed for divergent thinking (as opposed to line thinking). This is what we focus on in explorative BPM. Will we still need process models at all? Probably yes. But process models must look very different – such as the way we work with them. Process models should enable us to sketch out design alternatives as we go (and think) as well as to engage with stakeholders on our ideas. Once processes will be implemented, much of what was intended with process models, can be done by mining the processes` digital traces. So it’s high time – and an exciting opportunity – to rethink process modeling!”