Hat tip to IBM for creating some buzz about its launch of Blueworks Live this weekend. Sneak previews last week for excited analysts! Not BPM software as we had known it, more like the release of the new Harry Potter movie.
I'm reluctant to rain on IBM's parade because much of the vision in Phil Gilbert's gig at BPM2010 is spot on. But it seems to me that Blueworks Live has some serious design flaws. More on these below - let's start with an independent assessment.
Sandy Kemsley has a characteristically sound review on her Column 2 blog, and her verdict on Blueworks Live is downbeat. She's underwhelmed by the Twitter integration, the public and private streams, and the level of integration with the Blueprint BPMS engine.
Nimbus has been delivering cloud-based BPM for years so this sticks in the craw a bit - but Sandy is right that the real significance of Blueworks Live is what it signals about BPM cloud solutions:
"They’re not the first to have cloud-based process execution, but they are IBM, and that lends some credibility to the whole notion of running your business processes outside the firewall. Like the entry of other large players into the cloud BPM marketplace, I believe that this will be a benefit to all cloud BPM providers since it will validate and enlarge the market. This validation of cloud-based BPM is a real game-changer, if not Blueworks Live itself."
My own doubts about Blueworks Live center on two areas:
It's Overly Democratic. Every enterprise wants its people engaged with process excellence and continuous improvement. But it's a step too far to simply say that everyone should therefore be able to create and automate process. And it's no excuse to say that the activities to be automated in this way are 'the long tail' of SG&A activities - in HR and Finance, for instance - that will never be reached by automation programs. Even if that were true, it denies the benefits of process expertise and process integration, the value of an enterprise process architecture - actually, the value that the IT mindset brings.
It's a Governance Headache. I don't see how governance, compliance, risks and controls can be efficiently managed in a world where an organization's processes are automated in such a decentralized way by people who may be experts in their field but are novices in process design and management.
In a nutshell: operational excellence, continuous improvement and good governance depend upon joined-up thinking, an enterprise-wide collaborative framework that enables each part to make its specialist contribution to the whole.
And this requirement for a holistic process platform, owned and embraced by the entire business, will become even more crucial in the future as the enterprise becomes ever more virtual and multisourced.
In that context, it's difficult to see how Blueworks Live helps at all. Would you dine at a restaurant where each chef did their own thing? A good menu has coherence. A good restaurant creates the perfect customer experience by harnessing the genius and creativity of each chef de partie and their team together with the skills of the maitre d' and his or her team. I don't want beetroot for a pudding, or filet be boeuf Wellington served Thai style and with popcorn.
Related Posts
8 November 2010 BPM 2.0 on a Napkin
25 October 2010 Social BPM Yes. But Socialist BPM Won't Work
5 March 2010 Process Democracy: Be Careful What You Wish For

Blueworks is just a tool. That said to get the best out of it, good people and processes are still a must. If your business values the IT organization then they know who to contact for the assistance they need.
Posted by: Stacie Kenney | 22 November 2010 at 02:00 PM
Mike, they appear to be targeting ad hoc processes that are now done with email in the lightweight process management, so governance likely isn't an issue: these processes aren't governed now, and no one is looking at risks, controls, etc. for them in many cases. I suppose the goal is that if some of these become widely used, they can be adopted into a broader process governance strategy, but that doesn't seem to be the initial goal. Those completely uncontrolled processes are currently completely democratic, in that anyone can send off an email as part of a "process".
btw, had to look up "curate's egg" :)
Posted by: Skemsley | 22 November 2010 at 02:21 PM
Not being a curate at the court of the bishop (btw, I also needed to lookup wikipedia, but I thought because I'm not native English speaker), I would dare say that anyway I really see very low innovation in the currently available version of the tool.. See my detailed opinion here in my recent post on Much ado about nothing.
Posted by: Marco Brambilla | 22 November 2010 at 05:39 PM
Hi Mike,
First… a sincere thank you for this post. I love a debate and you've identified a couple of real places we should be challenged on. It's the rare individual (or product) that is 100% correct so we're watching the Blueworks Live discussion closely and, hopefully, we'll learn more and adapt as a result.
So in that spirit, I wanted to respond with my thoughts on the one minor and two major points you've made.
The Twitter stream: Who knows if this is going to be a success? However, you missed one critical piece of the stream: it's not just following people, it's also constantly updating for the relevant topical hashtags. I do believe that Twitter represents THE public community (or should I say is the metadata of the public discussions) happening on every topic these days. Our curator is trying to make sure the BPM Twitter stream is the most current, most useful representation of that community. I don't know anyone who has the time to keep their Twitter feed up to date… but again, we'll see if people find value in this.
Too democratic: As Sandy points out above, the processes that Blueworks Live means to automate are today ungoverned and anarchic. With Blueworks Live, those processes would be visible! I agree that if a given process owner has access to a process expert, then the process expert should assist (which is why Blueworks Live also allows you to blueprint a process in detail). But most of these processes-over-email don't have this exposure or prominence. Our assertion is simply this: Blueworks Live is better than process-over-email. And I would also say: A process documented and instrumented at even higher degrees of fidelity is better than Blueworks Live. Moving to the left on the long tail chart always results in higher productivity, re-use and repeatability. But there's not enough money or talent in a given organization to move every process to the left… so for those on the right, Blueworks Live is better than email.
Bad governance: Given the state of processes run over email today, any tracking and visibility is an improvement, even if it's not perfect. With Blueworks Live, you have 100% visibility into those processes for the first time. You may then choose to govern them or not, but the centralized information and reports are there in the product. I would suggest that bad governance is to continue to run those processes over email, knowing that they exist; and good governance would be to respond with Blueworks Live. Further, one of the charts that Blueworks Live generates is actually your own "long tail" chart. It shows clearly which of those processes are most used and which have the most instances running through them. This gives you a god indication which of the Blueworks Live processes should be taken away from Blueworks Live and raised to an even higher level of maturity… Inside the tool you can directly push these processes forward for blueprinting by process experts… and then push them into IBM's on-premise BPM suite for even better automation and understanding (or to anyone else's tooling, via XPDL, if you choose).
No company has the resources to provide process experts to oversee 100% of the processes in their company. I am the first to argue that process understanding and improvement requires a lot of data and expertise… especially as you move up the maturity curve. To move, say, a 2 sigma process to even 3 sigma will not be done solely by lay people. But the majority of the processes we run today are 0 sigma, they are run over email. I'm not talking about HR and SG&A processes. I'm talking about portions of almost every process involving people: like sending a special quote around for approval… right in the heart of the sales process. There are many examples. Moving these processes out of email so that they can even be visible as processes is the primary goal of the automation capabilities of Blueworks Live. The reason we chose automation, as opposed to simple documentation, is that most people in the business don't have the time, and are not incented, to simply document their processes. They have work to do. So they use email to get it done. Blueworks Live means to make their daily lives better by using a better way to get their work done without email.
In the past, the industry focused on documentation in order to get to execution. For these hidden processes, Blueworks Live turns that notion on its ear, and suggests that for these immature process areas, we can execute in order to get documentation. Just by making their personal jobs simpler, via Blueworks Live, people can get more done, and compamies get the benefit of transparency and documentation.
I hope this helps… again, thanks for the discussion.
Cheers,
Phil
Posted by: Phil Gilbert | 22 November 2010 at 07:43 PM
update: in re-reading my comment I noticed I used the term "simple documentation" and that was not my intent. Documentation and understanding are mostly not "simple." I should have said that we chose "simple automation instead of simple documentation" meaning: if a non-process expert is going to do something, we went the "automate part of your job" instead of the "document part of your job" route with that part of Blueworks Live. Sorry if it came across wrong.
Phil
Posted by: Phil Gilbert | 22 November 2010 at 07:49 PM