Jim Holincheck

This post is a follow up to the presentation Jeff Woods and I did at Gartner Symposium/ITxpo on Wednesday, October 15.  We did not get a chance to answer any of the questions submitted from the audience so we thought we would use this blog post to provide answers to the attendees (and others)

• During your presentation, you present arguments for both sides (ERP suites and Best of Breed – BoB — solutions).  Is it possible to take a clear position on some of them (for example, TCO, Time to Implement)?

It just is not clear cut.  The TCO argument depends on the purchasing model (perpetual license vs. subscription), delivery model (on-premise vs. hosted), maintenance model (self-maintained vs. vendor-maintained), and the timeframe of the analysis (SaaS may provide a short-term cost savings, but be more expensive in a longer-term compared to a perpetual license, on-premise installed, self-maintained solution that is fully depreciated).  That is why we tell you to not blindly believe vendor assertions.  You have to do the analysis yourselves based on your particular circumstances.

I am also somewhat skeptical of claims that SaaS has a much shorter time to implement.  Yes, it can be quicker to start a project because the environment is already up and running (you do not have provision the infrastructure).  However, when comparing "apples to apples", most of the things that drive the time to implement are vendor (and delivery model) independent.  It includes factors such as:

• how much process reengineering will be done
• how much customization is required (this is a vendor and deliver model-specific factor — SaaS allows limited customization so this can be an advantage assuming it meets requirements)
• difficulty of data conversion and interfaces
• report creation
• training/rollout strategy

The challenge for ERP solutions is that a particular area is not implemented in a vacuum.  There are dependencies.   I can make a change in module of the application that has a downstream impact on the usage of other models.  In addition, the scope is often not "apples to apples".  A specific module may be implemented as part of a broader ERP solution implementation (a larger, more complex project).  So, best of breed may have an edge because it is a smaller, self-contained project.

• In user acceptance testing, I saw that often users simply want to avoid the governance issues/complexities or shared services associated with ERP.  Is this a common issue?

The short answer is yes.  What I advise clients is that you have to define the WIIFM (what’s in it for me) for users to change.  If they do not see the benefits to themselves personally, they will often revert to the "old way" of doing things.

• Hasn’t ERP (for example, SAP) become more modular (synchronized/integrated) than engineered?

This is an interesting question.  Many ERP solutions are becoming more componentized (and service-oriented).  However, SAP’s ECC 6.0 would still fall in the engineered suite category.  Oracle PeopleSoft would be an example of a synchronized suite (it followed a "pillars" approach for Finance, HCM, etc. that allowed independent implementation and upgrade, but still synchronized data).  However, the direction seems to be moving somewhat away from engineered as the large ERP vendors do more acquisitions and service-enablement.

• Does BoB or ERP better support integration with external systems and processes such as income tax management or payroll tax management via a third-party BPMS?

I am not sure I understand this one.  Most BoB and ERP vendors do not use a third-party BPMS to either orchestrate processes or to integrate with external systems.  So, if you submitted this question, please feel to clarify via a comment.

• Which module do you recommend implementing first (HR, Finance, etc.)?

All things being equal, usually Finance is the best place to start.  All financially related transactions ultimately end up there.  In addition, management and statutory information requirements are often defined as well (for example, accounting key design).  However, it is certainly possible to start with HR and integrate with an existing financial system if there is a more compelling business case.

• What can you do with a combination of both ERP and BoB?  What is the best strategy to manage?

As we discussed in the presentation, the best way to manage it is to think of it as a portfolio.  You will have a combination of ERP, BoB, and custom solutions.  As business requirements evolve in your organization, look for opportunities to move commodity processes to the ERP (remember leading edge practices become more commodity over time).  Leverage BoB for emerging practices or to meet business requirements that cannot be met in ERP.  If it is truly strategic and differentiating, do not forget to consider custom development because it is likely there is not a packaged solution to support your needs.  Revisit you application portfolio mix regularly and adjust as needed.

• With ERP how do you avoid being held over a barrel regarding maintenance cost?  You’re tightly coupled with the vendor.

This is the lock-in issue we discussed.  One way to avoid lock-in is to put a fence around your ERP usage that is well-defined.  Another way is to use multiple ERP vendors so that you keep alive competition over time.  A third way is to make sure if you buy a suite, that you get itemized pricing so that you discontinue paying maintenance on a module if you stop using it.