business process management
Forrester just released their new report on human-centric BPMS. Hmmmm, the title is actually “The Forrester Wave?: Human-Centric BPM For Java Platforms, Q3 2007”.
BPM for Java? Good way to totally alienate the business community from ever wanting to be involved in selecting a BPMS, guys.
Metastorm CEO Bob Farrell gave a briefing this morning about their acquisition of Proforma; not much that wasn’t covered in the press release except the following:
Metastorm announced today that they’ve acquired Proforma. Strangely, the Proforma URL already remaps directly to the Metastorm site and the products are already relabelled as “Metastorm ProVision”; most acquired companies keep their own site and brand visible for a while so as to not freak out customers who haven’t heard about the acquisition yet.
I covered the Proforma user conference last year; I’ve always been impressed with their modelling tool since it goes beyond process modelling to full enterprise architecture modelling, and they seem to be moving in the right direction with increasing their browser-based modelling capabilities to allow this to be rolled out to a greater user base within an organization. They’re currently leaders in both the BPA space and the EA modelling space.
There’s still the round-tripping problem, however: I haven’t been briefed on this by either party, but based on my past conversations with Proforma, I’m suspecting that it’s a one-way trip from Proforma’s modelling environment to Metastorm’s process execution environment, since you likely have to tweak the modelled process significantly in order to make it run, and they likely aren’t supporting an extensible interchange format that would allow those changes to stay with the model if it were moved back to the modelling environment. I’m just guessing on this, of course, and would love to hear different. And, although Metastorm was one of the first vendors to offer a free downloadable process modelling tool, I can’t find that on their site any more, so they may be putting all of their process modelling eggs in the ProVision basket.
Metastorm bought CommerceQuest in late 2005 to improve their integration-centric BPM capabilities, and this acquisition rounds out the front end of their product suite — Forrester gave them a black mark in last year’s report for relying too much on third-party software, and this directly addresses that concern. The trick, as with any acquisition, will be how seamlessly that they’re able to integrate Proforma’s products.
I booked my travel for the Gartner BPM conference on September 16-19 in Orlando: I have a bad habit of forgetting to do that until late, then finding that the airfare is $$$ and I’ve missed all the hotel rooms, so thought that I’d do it early this time. I’ll be live-blogging from the conference as usual. I won’t be staying on for the Event Processing Summit that they’re running for the remainder of the week; although it looks interesting, I’m speaking at the Forrester IT Leadership Forum the following week in Carlsbad, and wanted to spend a few days at home in between, plus I want to be back in Toronto on the evening of the 19th to make it to our second Girl Geek dinner.
There was a lot of vendor grumbling about this conference coming so soon after the one in the spring, but I’m sure that all the usual suspects will show up nonetheless.
Disclaimer: Gartner provides me with a press pass to the event, although I am not otherwise compensated for expenses or my time.
I had a chance to talk to Kevin Hakman of TIBCO late last week about their Ajax Challenge (Kevin co-founded General Interface, which was acquired by TIBCO a couple of years back), the goal of which is to build the world’s largest mashup. You have to use General Interface to build it, but more interestingly, you have to use PageBus, a JavaScript client-side message bus that TIBCO just contributed to the OpenAjax Alliance as open source to become part of the OpenAjax Hub 1.0 installation.
The contest is splashy (win an oversized TV! win a video iPod!) but PageBus is the real news here: it provides a message bus for mashups in an attempt to eliminate the spaghetti mess of point-to-point integrations that we’re already starting to see emerge. In the enterprise world, this is why ESBs have become an essential part of any sizable application integration effort: without a message bus, you’re creating a unique integration between each pair of applications. Okay when you have two applications, but not when you have 10. [To be fair, usually you don’t have every application interact with every other application in a complex integration: each one may only interact with a couple of others, but that just shifts the pain point, it doesn’t eliminate it.]
Getting back to mashups and the OpenAjax Hub, the PageBus exposes the basic functions of messaging — publish, subscribe, unsubscribe — all in less than 5k of JavaScript, so that multiple Ajax components on an HTML page can share data using these standard methods. This allows the development of a mashup to be more easily split up between multiple developers since each can focus on their specific component and not on the interface between components; it will also allow for easier “no programing required” assembly of components within a PageBus-enabled mashup framework.
This is a pretty important step in mashup-land: I’m starting to see a lot of things referred to as mashups that are actually portals, where the components don’t intercommunicate, but the fundamental benefit of mashups is that they are an integration, not just components that happen to coexist on the same page.
TIBCO is apparently already using this in their BPM product for things such as task list publication, which means (I think) that you could create a mashup between your iProcess task list and some other component or data source — a real BPM mashup. Although many vendors are starting to provide RSS feeds of task lists/inboxes (I hope that my past year of nagging about this has had some contribution to those efforts), this is the first truly mashup-enabled BPM environment of which I’m aware.
The full OpenAjax Hub specification is about 4-6 weeks away from release, but the project is already on SourceForge. TIBCO will continue to develop the source and contribute to the open source efforts in the future; their press release about PageBus is here.
Next up was a panel of BPM vendors: Phil Gilbert (Lombardi), Angel Diaz (IBM), Marco ten Vanholt (SAP BPX), Burley Kawasaki (Microsoft), Scott Byrnes (Handysoft) and David Shaffer (Oracle). Derek Miers moderated, and posed a series of questions to the panelists rather than having the panelists do short presentations as we saw on previous panels — a much better panel format, in my opinion, and it even generated some conversation between the panelists directly.
Phil mixed it up right away by agreeing with the other panelists that standards are important (duh), but said that the first thing that we need to standardize is the meaning of the term BPM. He also thinks that OSM (Organizational Structure Metamodel) is going to be one of the most significant standards in the coming months, next to BPMN. In other words, people are going to start modelling their business, not just their processes. Marco added that there’s going to be an increasing interest in the processes that span organizations, and standards that support that will become more important. They all seem to agree that business users don’t really care about standards explicitly, but that standards are an implicit part of things that the business types to want: portability of models and reusability of skills, for example.
One question was whether BPM offered via SaaS is reducing the barriers to entry to what is still a complex implementation. Burley feels that it will make a difference for departmental applications that just can’t justify the spend, and for cross-organizational choreographic processes where no one organization is “in charge”, but that there will still be a strong market for on-premise solutions especially at an enterprise level. Angel added that standards are going to play a strong role here, since there’s likely to be a hybrid approach that uses both on-premise and on-demand systems within the same processes. Marco made the statement that some industries will “never, ever have software as a service”; it will be interesting to come back in a few years and see if he has to eat his words. Many organizations already have their data centres outsourced, including those that require advanced security, and I think that SaaS is just a small step beyond that from a security standpoint even though it might be perceived as being something entirely different. Scott things that a template-driven, simpler type of BPM functionality could be adopted by the SMB market. David pointed out that there’s a difference between having BPM embedded in a SaaS application and offering BPM directly as a SaaS, and feels that the latter is going to see much lower adoption. Phil stated that their Blueprint product is a tactic in their way to building a cloud capability, implying that we’ll see some hybrid on-premise/on-demand functionality from Lombardi in the future
They then discussed mechanisms for supporting more collaboration and deeper embedding into a worker’s environment. Scott talked about being able to share, for example, information about the experts for a specific process, and be able to IM them directly. Marco talked about being able to do some collaborative Visio diagramming in a wiki-type plug-in (presumably on BPX); I’m not sure if this something that they have with a browser design interface, or if it’s a place to upload Visio diagrams. He also pointed out that wikis, forums and IM are going to be start to be built into applications for collaboration, further pushing the need for standards since none of us want the BPM vendors to build their own wiki or IM software.
A question from the audience asked whether the vendors are getting inquiries from other vendors to embed/OEM their BPM functionality inside another product, whether SaaS or not. David, Burley and Angel spoke up that they are seeing this; not surprising since Oracle, Microsoft and IBM are all “platform” BPM vendors that tend to offer components rather than a more cohesive suite. Although I haven’t written up my notes from the BPEL roundtable yesterday, this is one of the areas where standards like BPEL will help to facilitate that type of integration. Phil added that they’re seeing this as well, but more from the standpoint of embedding more of their suite rather than just the engine.
Another question was on the distinction between modelling processes for business improvement purposes, and modelling processes as a visual coding/RAD tool. Phil responded that if you’re just buying BPM as a RAD tool, don’t buy it: stick with Java or .Net.
There was a discussion on the role of large vendors in standards, and how large vendors can sometimes take a standard off into their own organization and develop it 80% of the way and bring it back to the standards group: sometimes this works well, and sometimes it allows the vendor to just mould the standard to their own product agenda. We also came back around to the comment that Phil made at the beginning of the panel, where we need to define what BPM is in the market: the vendors all seemed to agree that they all have their own definition of BPM that coincidentally matches completely with their product functionality, and they all agreed on the buzzphrase “BPM is all about the business”. 🙂 The analysts also all have their own definitions, although they all seem to be congealing around the Gartner definition of BPM as a management practice, which doesn’t at all help the issue when the BPM vendors define it in terms of the technology capabilities. Bruce Silver lobbed a small incendiary device from the audience by stating that from the viewpoint of BPM as a management discipline, the vendor products are all exactly the same, and that customers may just see them as snake oil salesmen trying to sell the same thing in a different way. Not sure that we’re going to solve this one today.
It’s interesting watching a vendor panel like this, where the panelists are not allowed to do any product pitches, and where they’re all pretty smart guys: the discussion is a complex weave of philosophy, techno-geekery and thinly-veiled nudging towards their own specific agendas. This is part of what I like about the BPM Think Tank: there’s much more open collaboration between vendors than at other conferences, although there’s always a strong streak of friendly competition throughout the interactions.
I woke up this morning with the thought that there was an easier way to access the BPM Events calendar that I created yesterday, so I created this page here on Column 2 that embeds the calendar. You can always find it using the “BPM Events” link under the Pages section at the top of the sidebar on the left.
I’ve heard from a few people lately about how hard it is to schedule BPM events since there’s so many calendar conflicts. I joked yesterday that we needed a global Google calendar to which everyone can add their events, and today I’m not laughing, I’m creating it. Click here to subscribe to the calendar in Google calendar using your Gmail login, click here to access the calendar directly without a login, or here for the RSS feed.
I’ll add events to it as I hear about them, and if you want to be able to add your own events, just email me and I’ll add you to the list of people who can add events to the calendar. It would be great if a lot of the BPM vendors could get on this calendar so that people who are interested in events can come to one place to find out where and when things are.
This is the first public Google calendar that I’ve created, so if I’ve done something dumb here, or could make it better in some way, add a comment or drop me an email and I’ll do what I can.
The definition of “BPM event” is a bit loose, so any event that would be of interest to the community is welcome here.
We had two case studies about BPM projects in the real world, the first of which was Lance Gibbs talking about AFLAC using BPM in their administrative services. They started with their invoice reconciliation process, since they dealt with over 500,000 remittances each month, and 30% of them didn’t match the invoice so couldn’t be automatically processed — a problem that I see in many of my own customers in a wide variety of industries. In their case, a mismatched invoice may correspond to a change in coverage, which required changes by their Policy Servicing group, and someone needed to figure out what was wrong with the invoice and start it through the remediation process. There was no visibility into the process, no prioritization to ensure that higher priority invoices were processed first, no coordination between the groups, and a lot of paper flying around. They really needed to get an invoice reconciliation turned around in less than 30 days, or else another incorrect invoice would be sent out for the next month’s billing.
To address this problem, they focussed on straight-through business process automation, human-facing workflow, and business activity monitoring. They started by scanning and indexing documents up front in the mail room, and early sorting and triage of transactions, which cut more than a day off their cycle time and allowed their reconciliation specialists to start work right away. They created web services that access their mainframe billing system using webMethods to be consumed by BPM. They had just achieved CMM level 3, and had to work at using agile BPM development processes that didn’t violate their CMM methodologies. Although it’s less than perfect, they now have more than 85% “one and done” for fixing invoices.
To establish KPIs, they looked at which were critical to the customer, and generally fell into three categories: quality, delivery and cost. They went from more general, hard to measure needs through to specific, measurable KPIs. For example, good customer service (a need) is a result of knowledgeable and friendly CSRs and a short wait time (drivers); knowledgeable CSRs can be measured by whether they were able to answer a question without further research (KPI, or what they call CTC for critical to customer).
They had a core process team of about five people, with other integration specialists being accesses as required. They went through a 4-1/2 month vendor selection process with a full bake-off between the two short-listed vendors (which he didn’t name, probably having been told not to mention vendors specifically from the podium). Update: Lance told me later than the BPMS is Lombardi.
The next case study was Clear Channel, with someone else pinch-hitting for David Jemeyson who took ill suddenly, discussing how they interfaced with Google to place ads through Clear Channel’s radio stations. The basic process is that a customer makes an ad purchase and uploads their audio spot with Google, who dispatches it to Clear Channel where it is prepared for broadcast and sent on to the radio station. At each step, they’re using loosely-coupled web services to move the data through the process, and they’ve modelled the entire thing in BPMN which he showed on the slides complete with drill-down to the detailed process with swimlanes. The result is that the entire process can be executed without manual intervention, including sending on the relevant information to their financial systems, and they were able to develop it all in about six months.
They used BizTalk and SharePoint for most of the SOA infrastructure.
The contrast between the two case studies was interesting: AFLAC was optimizing an existing business process; whereas Clear Channel was opening a completely new market enabled by the technology, which is one of those elusive ROI factors that we talk about but rarely see.
Phil Gilbert kicked off morning with welcome and logistics before turning it over to Paul Harmon, who gave a keynote entitled “Does the OMG have any business getting involved in business process management?” I love a little controversy first thing in the morning.
He started out with a fairly standard view of the history of BPM and process improvement, from Rummler-Brache and TQM in the 80’s to BPR in the 90’s to BPM in the 00’s. He pointed out that BPM has become a somewhat meaningless term, since it means process improvement, the software used to automate processes, a management philosophy of organizing businesses around their processes (the most recent Gartner viewpoint) and a variety of other things.
He broke down BPM into enterprise level, process level and implementation level concerns (with a nice pyramid graphic), and gave some examples of each. For example, at the enterprise level, we have frameworks such as SCOR (for supply chain) and high-level organizational issues such as the Business Process Maturity Model (BPMM); Harmon questions whether OMG should be involved at this level since its primary focus is on technology standards. Process-level concerns are more about modelling, documenting and improving processes, and spreading that process culture throughout the organization. Implementation-level concerns includes the automation of processes, including execution and monitoring, plus the training required to support these new processes.
He made an interesting distinction between stable processes,which need to be efficient and productive, and dynamic processes, which need to be flexible. Processes that are newer or need to be changed frequently are in the dynamic range; in my opinion, these tend to be the processes that are competitive differentiators for an organization. IBM has recently thrown the concept of “value nets” into the mix as an alternative to value chains, but Harmon feels that both are valid concepts: possibly using value chains for stable processes, which might even be candidates for outsourcing, and value nets for more dynamic processes.
He also made a distinction between process improvement, process redesign and process reengineering, a division that I find a bit false since it’s more of a spectrum than he shows.
There was an interesting bit on model-driven architecture (MDA) and how it moves from platform-independent models (in BPMN) to platform-specific models (also in BPMN) to implementation (e.g., J2EE); for example, there may be parts of a process modelled at the platform-independent level that are never automated, hence aren’t directly mapped to the platform-specific level.
He put forward the idea that process is where business mangers and IT meet, and that different organizations may have the implementation level being driven by either the business side or the IT side, and that there’s often poor coordination at this level.
He then discussed BPMS and came up with yet another taxonomy: integration-centric, employee-centric, document-centric, decision-centric and monitoring-centric. Do we need another way to categorize BPMS? Are these divisions all that meaningful, since the vendors all keep jostling for space in the segment that they think that the analysts are presenting as most critical? More importantly, Harmon sees that also the BPM suites vendors (those that combined process execution/automation with modelling, BAM, rules and all the other shiny things) are leading the market now, the platform vendors (IBM, Microsoft, etc.) will grow to dominate the market in years to come. I’m not sure that I agree with that unless those platform vendors seriously improve their offerings, which are currently disjointed and much less functional than the BPM suites.
Harmon’s slides will be available under OMG-BPM on the BPTrends site. There’s definitely some good stuff in here, particularly in the standards and practices that fit into each level of the pyramid.
Good thing that I’m blogging offline in Windows Live Writer, since the T-mobile connectivity keeps dropping, and isn’t smart enough to keep a cookie to stay logged in, but requires a new login each time that its crappy service cuts out. Posting may come in chunks since it will likely require me to dash out to the lobby to get a decent signal.
