Phil Gilbert | Perspectives in Process
Business process management requires a new set of technologies. By 2010, these will replace ERP as the primary focus of solution engineering at companies large and small. By 2020, managing process through technology will be second nature to senior executives, and the transactional systems we use today will be like mainframes. My blog talks about BPM today, tomorrow and where we'll be in 2020. Welcome.
  Home About Me About The Blog  Search  

« Who's In Charge? | Main | The "How" Company »

Like Humans Do

There are really two points James McGovern makes in his post What Enterprise Architects Should Be Thinking About BPM...

First: Human-based processes are rampant, are EAs really thinking about them (and does BPM technology address them). In James' words:

"SOA is wonderful for system-to-system iterations but doesn't really address what architects should be thinking about which I believe is the simple fact that the vast majority of true business processes occur between two individuals without a system in the middle..."

Second: Is anyone working on a BPM reference architecture?

Human-based processes

A while back we were working on this problem and the fact is that there is a "long tail" to processes inside companies. Our depiction looked like this:

Humanprocessdistribution
(Double click to enlarge)

A lot of humans are involved in a few large processes, and a lot more humans are involved in a lot of small processes. Having tools that empower the discovery, creation and monitoring of all those "long tail" processes is crucial to doing business in the 21st century.

In the aggregate, many more humans are affected by small processes (processes that contain only a few people), and mostly don't touch systems, than are affected by complex, large processes. Our studies show that probably 70% of your people are in this category. It doesn't take a lot of work to validate by "smell test" these numbers. How many of your employees routinely use <fill in the blank ERP or legacy system>?

The implications of this speak volumes about the state of "Business Process Management" vendors who talk about automation and rules and BPEL and SOA. They are destined to remain niche players in a declining space! Integration of systems will commoditize, and in the not so distant future. SOA is a great step and is the foundation that insures this.

But business process management (lower case) is NOT SOA! It's not part of the "SOA stack." Business process management is about the business, and the processes that are key to the business have human interactions all along the way - and not just on the run-time user side of things. On the development (or process modeling) face, on the reporting and visibility face. Everywhere. The technologies of a thoughtful business process management strategy and implementation are consumers of the SOA interfaces, when humans need information (or need to kick off automated processes).

One interesting but overlooked aspect of our research is depicted on the dotted lines. These represent the user profiles of the people who are needed to author the processes that the solid lines reflect. As you can see, different approaches - multiple approaches - to modeling are required, if your goal is to embed more technology into the business. Your modeling environment (or environments) must be structured enough to insure transferability of drawing as well as transferability of the model. For example, if you use Visio to draw the picture, just try giving that drawing to 5 people and see how many can understand it, versus how many will simply start over. Even if the objects are transferable, in the long tail you need to communicate to humans, and structured diagrams with simple objects are key.

Another implication is "where do these users live?" That is, if you are a member of a few 10-person processes... do you really need to log into the corporate portal all the time to see if you have a task? Isn't this sort of like "browser 2000"? We need to move to an exception-based world for these users, and deliver process content (like tasks waiting) to portals of THEIR choice, not your BPM vendor's choice! This means, all of a sudden, that Outlook, Notes, Word, Excel and PowerPoint just became more important parts of your process architecture....

And, finally, all this means that not only do we have to share run-time experiences beyond what most of today's so-called BPM vendors provide, we also have to deliver modeling and authoring tools that they can use. These need to be highly portable, highly graphical, etc. If you have to understand, say, a rules ontology then you've missed the market. And I am not talking about my market (the BPM space), but your market, inside your company.

A BPM Reference Architecture

First the positive: OMG is working on this and there are some great vendors participating in this. Axway, Mega, IBM, EDS, Enix Consulting, IDS Scheer, Deere & Co., Sterling, Popkin and, oh yeah, Lombardi. (I am sure I missed some... it's been a long day and I'll add by comment tomorrow). I am not talking simply about the standards per se, but also about the "stack." I think that work on this has been slowed a bit by the merger, but I think you will see movement on this, and probably some good discussion at OMG's BPM Think Tank in May.

However, we have only a little participation from the non-vendor community. So I'll challenge you, James: join OMG and participate in the BPM discussion! We don't have a clear path to doing this... but I think we can have a program in place by Think Tank and I'll push to include this as a topic on the agenda.

Now the negative: You quoted from the Sun web site a book from someone I don't know and who has never spoken with Lombardi... so... I would guess his level of confidence in his statement is low. I say that not because I or Lombardi has all the answers, but I would guess there are other BPM vendors similarly unchecked... His statement would be challenged, for example, by Intalio.

But more to the point, even if he listed all the vendors that checked his boxes, it wouldn't be architecture. It would be speeds and feeds. A Mazerati and a Buick are built to the exact same safety standards of the US Government. It has no bearing on which car is right for a given problem.

In conclusion, I agree with you completely: "Today...[w]e have business process tools that cannot describe how business processes really work, we have workflow systems that ignore the real nature of work and we manage human beings with systems built on the mathematics of automata."

Process is not a math or a rules problem. It is a problem of visibility, management and control. It is a human problem. Humans discover, humans author, humans work, humans improve. The people or systems in the middle of all that simply need to have the information they need to work, presented in context, on the tools they most effectively use.

And, then, good things can happen.

I'm breathin in
I'm breathin out
So slip inside this funky house
Dishes in the sink
The TV's in repair
Don't look at the floor
Don't go up the stairs

I'm achin
I'm shakin
I'm breakin

Like Humans Do


-
David Byrne, Like Humans Do

Technorati Tags: , , , , ,

Comments

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Working...
Your comment could not be posted. Error type:
Your comment has been posted. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.

Working...

Post a comment

Recent Posts

Archives

Categories

Blogroll

Etc.

    Subscribe

    Squidoo BPM Lens

Lijit Search