A post by Don Dodge about Craiglist mentioned The Mythical Man Month by Fred Brooks and launched me into this post. The second-system effect is my favorite theme in this book and I often think about it on SAP HCM implementation projects.
The second-system effect in essence states that an engineer will be sorely tempted to add those features to her second system that she wasn't allowed to on her first system development project. Bells and whistles and all sorts of desirable features will want to be introduced. The first system was successful in no small part because its scope and budget were strenuously controlled, but the temptation for the engineer is now to make the second system even more successful by delving into that wishlist. Thereby missing an important fact about the success of the first.
SAP HCM implementations, especially when they include Benefits, Time Management and Payroll Management are technically demanding. They present plenty of opportunity for over-engineering technical solutions - e.g. for automating the handling of an exception that affects a small number of employees. The more experienced a systems integrator is with SAP HCM, the more tempting it becomes to automate just because she knows it's possible in SAP. Something to be wary of since this automation will take up time that could have been spent on something with better value-add.
That aside, it got me thinking about how we're scoping HCM projects today and how we might in future. As an example, today employee self-service is typically part of the initial roll-out. This wasn't always the case even after employee self-service technology became available. But now it is a leading practise and an integral part of improving HCM business processes.
Now, at least from what I see here in the US, is that most wall-to-wall SAP HCM implementations include OM, PA, BN Administration, Time and Payroll + ESS and maybe some MSS as an initial objective. If an organization is interested in SAP as a wholistic HCM solution then this picture might change in the not-so-distant future. We might instead be looking at e.g. OM, PA, ESS, BI, eRecruitment, Performance Management as the initial goal. (Plus MDM to keep SAP and remaining legacy databases in sync.) Benefits, Time and Payroll Management will come next - if these aren't outsourced.
One of the appealing things about such a development is that customers would get the strategic value-add of these functions relatively quickly, instead of having to wait until all the core transactional functions have also been implemented. From a project sponsorship perspective there's the advantage that this would allow the implementation of subsequent functions to occur under the protective umbrella of an early success.
Way down the road SAP customers might be looking at configuration-before-contract options - as Hasso Plattner divined in his Sapphire 2007 keynote.