I like your [BPMN Method &] style

I recently completed the BPMN Method & Style training, led by noted BPM specialist Bruce Silver and based on the principles found in his book of the same name.

This post is about what I personally got out of the training (which includes certification), the improvements it made to my process modelling and why I highly recommend it. I took the online class between the 5th and 7th September 2012.

Please note that this post expresses my own personal opinions; I’m not affiliated with the training provider, nor have I been financially compensated in any way for the taking the training or writing this post.

What’s the training all about?

Well here’s what it’s not about: it’s not learning the ins and outs of the BPMN 2.0 spec. BPMN’s had some harsh criticism for its ‘complexity’ and I posted about my thoughts on that in a previous article.

Knowing the spec is one thing; knowing how to apply it in the real world so that your business stakeholders understand it is quite something else.

In my last post, I used the technical challenge in the Great British Bake-Off as an analogy for an interviewee using their knowledge and skills to elicit requirements. I’ll use it again here: in the technical challenge, all competitors are given the same outcome to achieve and the ingredients (spec) with which to do so, but the ‘how to’ is deliberately vague – they use their skill and expertise to apply the required techniques to achieve the desired outcome.

Surprise, surprise, the end results are very different. Good process modelling is no different.

Specifications aren’t ‘how to’ guides; this isn’t peculiar to BPMN, no more than a list of ingredients can tell you how to combine them in such a way as to create a culinary masterpiece. My wife can pull together a complex dish at the first attempt with minimal issues; I can try the same and not only will it be a disaster, I’ll have destroyed the kitchen in the process.

What did I get out of it?

For starters, how to get models on one page by using hierarchical modelling. I’ve always tried to do this anyway having read Bruce’s work extensively, but I understand it far better as a result of the method.

The method starts without going anywhere near the notation itself. First, we understand the process scope – how you start it; how you end it (and indeed, the multiple ways in which a process might end). A lot of process mapping I’ve seen just picks a start point and goes straight to the detail from there: bad idea. Why do I say that? The scope’s ill-defined. Ever been sat in a painful workshop that’s a lot of talking and no real outcomes? That’s because boundaries aren’t established and so everyone’s expectations are different. What comes out of it, typically, is another workshop!

Then we have the high-level map, still going nowhere near the notation. What are the key activities between the start and end points? What are the outcomes of each of those activities (this gives us an idea of the number of potential different routes through the process)?

A lot of this is fairly common-sense stuff; scoping and understanding outcomes are the bread-and-butter of everything from project definition through to use case development. But as many will know, the effectiveness of applying that rigour varies from project manager to business analyst.

Then we add in the concepts of pools and lanes, mapping the interaction of our process with external entities and external processes. Understanding all of an organisation’s touchpoints with a customer during an end-to-end engagement is the cornerstone of Customer Journey Mapping.

Top-level end-to-end diagram for my BPMN Method & Style certification exercise

The process model above is the top-level of the Part 2 certification exercise I did after the training. I’ve published the full article that I completed to pass part 2 of the certification, with the permission of Bruce.

What does it tell me?

I can see how it starts (production of an application pack); I can see all of the key activities (the application is processed; the payment is processed) and the associated touch points with the practitioner and the bank; I can see its different end points (they’re either renewed or registration lapses). So in one go, I can see the whole thing. One might argue you can see a whole process if you blow it up onto A0 paper: but just try following it around; my eyes are bleeding just thinking about it.

Yes, there are some different things on there you won’t see on your ‘traditional’ flowchart, but BPMN has specific semantics making its meaning unambiguous and gives the concept of reusability an actual fighting chance. The application of Bruce’s Method & Style means that those specific semantics can be applied to make diagrams correct, clear, complete and consistent.

For details of upcoming online class times, check out the BPMEssentials homepage. The timings are typically 11am – 4pm Eastern Standard Time and 4pm – 9pm GMT (don’t believe the 5pm – 10pm slot: we’re only five hours in front!) and the course runs for three days. I really can’t recommend it highly enough!

16 thoughts on “I like your [BPMN Method &] style

    • Thanks for the reply, Lee – you won’t regret the decision! And speaking of decisions, I’m actually thinking about expanding the exercise at a later date to incorporate an example as to how the Decision Model could be used in association with the business rules tasks shown in the ‘Request Application’ sub-process.

  1. Hi Nick,

    I looked at your diagram and one question came to my mind. I had some discussions with someone about BPMN2 and his take on intermediate/start message event is that it represents an automatic thing. I am not sure if this assumption is correct, the BPMN2 spec never makes this distinction, neither is this assumption made in Method and Style book. But after looking at your diagram, I assumed you consider an intermediate message event as being applicable for both manual and automatic things. What is your take on this? And how does the practitioner submit the registration application? On paper? Thru some web site? I am pointing to Application Received intermediate message event of your diagram…


    • Your colleague’s thoughts mirror those mentioned by Bruce in a couple of his recent blog posts entitled “What is a message?”

      He does distinguish events as being more ‘instantaneous’, that is, once the event occurs, the subsequent activity is then carried out immediately.

      This is the case in this example – the principle is that an incoming application should be processed as soon as it’s received.

      I have used user tasks to send and receive messages where the implication is less instantaneous and this is in line with the training that Bruce provides. So whilst it’s not mentioned in his book, it did crop up in the training.

      As far as the method of application receipt goes, it happens to be a paper application as opposed to a web form.


      • Well…timing is one thing. The other thing is more technical: does message event represent automatic things only or not?

      • An application should be processed right away, but in my opinion reality is different quite often. An application may wait for attention before getting processed…thanks for sharing, Nick.

      • I’m well aware of that – this is modelling the business process and does not reflect specific operational concerns that may be implemented either through a BPMS or some other system.

        When I asked Bruce about this particular configuration, he said it was acceptable to put a comment against the activity to say that there may be some delay between the receipt of the message and the execution of the activity.

    • It’s an expanded sub-process. If you think about how a collapsed sub-process works, you can have additional lanes within that sub-process that don’t appear on the parent level.

      Each of the activities within the sub-process are contained in specific lanes, because they’re the atomic activities carried out by the role that the lane represents.

      But lanes have no semantic meaning in the BPMN spec – they act as convenient containers and are more a layover from ‘traditional’ flowchart swimlanes.


      • I have personally learnt lanes are traps, they make one think in terms of lanes instead of subprocesses. I avoid lanes as much as I can and I use groups instead. Actually lanes work for leaf-suborocesses, in my opinion. Just sharing, I don’t claim I know better 🙂

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s