Skip to main content
SAP, business rules and decision management

(Posted by guest blogger, James Taylor)

As anyone watching the business rules space will have noticed, SAP recently bought Yasu, one of the small rules vendors. Gartner wrote about this here and made a couple of useful comments on the acquisition and it made me think I should blog some kind of response to answer the question "what does this mean for business rules and decision management in general?"

At one level this is clearly a threat to existing business rules management system vendors - whenever a big player like SAP starts including some feature in their broad product set, vendors of those features have to worry. Some SAP customers who might in the past have considered buying a BRMS to add some flexibility to their installation might now just use the embedded Yasu technology. Similarly the folks who were selling their BPM tools as a way to enhance SAP now have to worry a little more also, as the SAP platform will now offer embedded business rules. As the folks at Gartner said:

"This smart, if late, acquisition will challenge the status quo of the BRE and business process management technology markets"

However, I don't think this is the main consequence of the purchase. Overall I think it is positive for the rules market for a very simple reason. Today a very small percentage of the projects that should be done using business rules are done using business rules. This is the main challenge facing those of us who espouse a new way to developing systems - getting the non-believers to try a new approach. SAP adopting business rules makes the technology much more mainstream and helps to solidify its position as a serious technology. Just as the growing strength of the open source rules business shows a maturing market, so does this. Broader use and understanding of business rules technology can only help.

The issue here is the difference between taking a rules-based, declarative approach to "coding" (in a process or system) and actually managing decisions as a corporate asset. Just as BPM vendors provide some support for rules (and I agree with Gartner that this will force "BPM suite vendors to make additional BRE acquisitions as they satisfy their desires for direct ownership of this crucial technology"), this is not the same as supporting decision management. Decision Management means externalizing the high volume, operational decisions that drive your business processes and managing them as a separate asset. Just as BPM vendors correctly point out that managing your processes outside your ERP/CRM products can add value, so can managing your decisions. I could argue, and have, that managing decisions separate is even more important. While a process might be delivered using a single BPM environment, many decisions must be shared across processes, systems and channels. No matter how good the individual platforms you use are at managing rules, managing a cross-platform, cross-channel, cross-process decision requires you to step outside that platform. Most companies have long since realized that they cannot do everything on a single platform. If you have SAP and other software running your business, you need to manage decisions in a way that supports both. Indeed SAP's increasing positioning of NetWeaver as an integrated composition environment reinforces this. If the right place to manage something is outside of SAP then Netweaver will let you compose it into the solution you need. If you building decision services using a powerful BRMS like Fair Isaac's Blaze Advisor then Netweaver will let you integrate them into SAP but you will still be able to reuse those services and rules in other areas.

So, what should SAP do?

  1. Integrate Yasu into Netweaver to support declarative programming of relevant constructs.
    This is a better way to "code" many things and this should be extended widely throughout NetWeaver. This will help grow the acceptance of business rules as a paradigm.
  2. Make it easy to build real decision services on Netweaver (not quite the same as 1 above) and publish the APIs so that Fair Isaac, ILOG et al can develop compatible deployments.
    This will allow easy integration with products built on other rules platforms as well as making for easy upgrades for companies that start off embedding rules in their SAP platform and want to move to true decision management over time.
  3. Take the modules of SAP that are hardest to configure using SAP's traditional table-driven approach and embed the rules engine so that rules can be used to configure these instead
    Put some skin in the game. If rules are better, and they are, then some of SAP's application teams will need to show they believe.
  4. Encourage third parties to develop rules-based decision services for Netweaver
    It's a great way to add functionality to SAP installations without a lot of database or user interface change, which minimizes both development costs and deployment disruption. Major analytic application vendors like Fair Isaac are already well on the way to offering decision services-based engines that would really enhance SAP installations.
  5. Think about analytics in terms of decision services
    SAP is total distracted on the analytic front with its acquisition of Business Objects. Once it gets done with that it should be talking to its data mining and analytics partners about how to bring analytically-enhanced decisioning to NetWeaver.

Some other posts you might find interesting include:

Finally, these are my personal opinions not necessarily those of Fair Isaac Corporation.


Visit my Smart (Enough) Systems Blog(RSS) or my ebizQ blog (RSS). Buy the bookor visit the companion wiki

related posts