Many approaches to business process management (BPM) have centred around the use of standard workflow technology. However, many business processes straddle two or more business applications, and as such, the workflow needs to straddle the applications, and the data, as well. For example, taking an order from a web customer may well straddle the web site itself, a customer relationship management application, and an enterprise resource management application.
Therefore, BPM has historically faced the need to use both workflow and enterprise application integration (EAI). That the data the process is trying to work against may be "dirty", as in a record for a customer named "John Smith" may be in one data store as "J Smith" and another as "J Smythe" means that it pays dividends to use data cleansing and master data management technologies as well.
As processes become more complex, the need for modelling tools, for business activity management (BAM), and for full process management tools becomes more apparent.
The move to the use of composite applications—made up from an aggregation of functions created in a service-oriented architecture (SOA) environment—can add to the complexity of the overall system. When combined with the massive costs for the associated licences for all the various bits and piece, this will either create fragile solutions, or one where the organisation involved will just throw its hands up in the air and decide that struggling on as it has been doing makes a great deal more sense than it had thought.
For many of the existing BPM vendors, the big problem is in dealing with the many existing customers out there. Indeed, this legacy has been the downfall of many a product—Lotus 1-2-3 and Wordperfect ruled the roost before Windows came along, and it was only because Microsoft had so few customers for Microsoft Word that it could afford to throw its existing architecture away and start from scratch again, leaving Lotus and Wordperfect to lose their battles in trying to keep existing users happy.
Companies such as TIBCO, which acquired Staffware a few years back, and IBM, which acquired FileNet, have a high number of existing customers who they have to keep happy. Any massive change to approach tends not go down well as the users try to maintain existing integrations, process models and so on.
They are trying to change themselves, bringing in extra functions to become business process management systems (BPMS), rather than just plain BPM tools, but in some cases, this just seems to be papering over many large cracks. For a small company that is relatively new on the block, this can offer a good opportunity.
One such company is Cordys. Started by Jan Baan (of supply chain/ERP fame), Cordys has tried to pull all the requisite needs together to form a single, out-of-the-box BPM services solution. Finding Cordys a bit stuck for the best way to describe it, ex-Staffware CTO and now Cordys chief strategy officer Jon Pyke has named it a business operations platform (BOP).
The idea of the BOP is to provide a consolidated environment with a complete SOA grid (a scalable, virtualised means of layering on physical hardware to provide a highly available SOA platform), along with a composite application framework as well as master data management, with BAM providing the orchestration required to complete an overall BPMS. The idea is that a single stack reduces complexity and cost, and guarantees integration of all touchpoints.
By pulling everything together, Cordys hopes to create a far more flexible system where the various cycles within an organisation (such as hardware and software purchases, strategic and organisational cycles) can continue to be disparate, and yet the process capabilities overlying these can respond to the changes needed within any single cycle.
In theory, it all looks good, but will it all work in practice? Jon Pyke is a true evangelist on the subject, and he provides compelling reasons as to why Cordys should do well. However, Cordys is not a name that springs to mind when looking at BPM, and it will struggle to get on all the shopping lists it needs to compete with the big vendors. Its portfolio also seems to bring it into competition with others outside of the BPM space, which may create pressures that could defocus it.
More pressing is the problem that many will not see Cordys as a BPMS play at all, seeing the simplicity of the offer as yet more complexity (why do we need an SOA grid, a composite application framework, MDM and so on?). In writing this article, the complexity of what is on offer has become apparent, and Cordys will have to hide this technical complexity successfully through concentrating more on the direct business value messages to be successful. A large amount of education is required as to what the business value propositions are that Cordys brings to the game.
Overall, Cordys does seem to offer something that can provide a lot of value to the business—a highly flexible, responsive and easy to use means of dealing with task and process issues within an organisation. If it can position itself correctly, it could begin to make larger waves in the BPMS space. Whether this then leads to Cordys becoming a takeover target remains to be seen, but at least it should lead to interesting times for the company.