I got my monthly issue of HBR this weekend and was really pleased to see the article "The Next Revolution in Productivity" by Ric Merrifield, Jack Calhoun and Dennis Stevens. For those of you who've followed my posts on Business Architecture and Capabilities, this article will be a welcome and easy going read with a strong connection to MSBA, though the terminology used is a little different.

The article makes strong points that "most companies that have embraced SOA have applied it without first rethinking the design of their businesses. This omission means they have overlooked SOA’s greatest value: the opportunity to create much more focused, efficient, and flexible organizational structures." And that "few companies are using SOA to create more productive and focused organizations or to slash costs by purging duplicative operations and technologies. They are not revisiting the fundamental design of their operations."

The idea of plug-and-play businesses is mentioned several times and builds upon an understanding of how new operating models can be developed from some key classes of capabilities or activities:

• primary - i.e. core capabilities in MSBA terms - are activities that constitute the core competency of a company.
• shared activities are re-useable across divisions.
• shifted - i.e. outsourced capabilities in MSBA terms - are activities that can be performed by or transferred to a third party including partners, specialists and even customers.  
• automated activities are capabilities implemented as software services that can be incorporated into a SOA.

The article doesn't address what I see as a challenging gap between a capability definition and what you actually implement as a service in a SOA. The authors' use of the term activity rather than capability made it sound (to me at least) plausible if not easy to implement services in a straightforward manner provided activity analysis has been performed thoroughly. Getting to processes is then just a logical next step. But, having said this challenge is more or less glossed over, it's acceptable to omit tough technical design issues like this in HBR given its business-oriented target readership. The most important point that good SOA begins with good business architecture design and can lead to great improvements in productivity is made very well and with some great case studies too.

Finally, this paper is definitely worth reading and you ought to mail it to your CxOs immediately.