I've never seen it formally done, although I agree that it is a
critical factor. I attended a talk on reuse way back in 1998 at SD
East in Washington D.C., and the speaker had said that the one
differentiating factor in a successful reuse program was marketing.
One challenge I see to applying marketing inside the enterprise is
marketing the right stuff. For this reason, a good place to start
may actually be on the infrastructure side of things. To be
successful, the infrastructure team must have a clear marketing
message, clear services to be marketing, and clear processes for on-
boarding new consumers. All too often, an infrastructure item only
gets into production when the first customer needs to use it. As a
result, the focus is on that customer and no one has a clue how to
make it available to someone else. For example, let's suppose the HR
department funds the purchase of a Portal product for the company
intranet. Is that infrastructure team now in a position where portal
technology can be marketing to other teams that may want to set up a
departmental portal? If it gets marketed by the team, and then the
department goes through hell trying to go live, they've lost all
credibility. If it doesn't get marketed, some department may go and
invest in redundant technology. If the CIO makes a mandate, it still
doesn't mean the team responsible is prepared to take on new
consumers. The key change to the culture is a customer-centric focus
from the team providing the service, starting with an understanding
of what the service is in the first place.
I've encouraged teams to think of themselves as a vendor to the rest
of the organization. We all know that some vendors choose a growth
model where marketing is pushed very heavily, sometimes too heavily,
but with a hope that enough funds come in early on to allow the
product to catch up. Other vendors may invest heavily in
engineering, with hopes that they're able to get a few big wins and
let their customers do the marketing for them. It seems that the
best marketed products usually win out.
I mentioned that infrastructure areas may be a good place to start,
and this is only because the demand for their services is likely to
be greater. I think development teams face a bigger challenge, as the
number of potential consumers of their services is far less. If we
are going to market business services from the development teams, it
can't be AOL-style marketing without an appropriate target audience.
People will begin to ignore it all, as the majority of things won't
be appropriate to them. The only way to get around this is to
educate the development teams on the business rather than on
technology, and help them to know where the potential exists.
-tb
On Feb 5, 2007, at 6:13 PM, Steve Jones wrote:
Now this got me thinking about marketing and SOA, namely if you
don't market a service effectively it won't be used effectively.
> IEEE International Conference on Web Services (ICWS 2007)
> Salt Lake City, Utah, July 9-13, 2007
Part of the issue I see these days with registry products is that
they assume that all services are equal and that marketing is just
a description. This conference in part suffers from that by
focusing on its content v its perception. (Lets face it if that
said Oahu, Hawaii we'd be queueing up). So as a question. Has
anyone seen an SOA effort that has actually focused on marketing of
services as an important exercise?
The reason I ask is that the old "UDDI search and find" which is
clearly rubbish has died, so the future has to be "what I value"
which is as much about marketing as it is about reality. I've not
seen that yet, but I've seen the behaviours.
Has anyone else seen service marketing emerging as a formal
approach to reuse/use?
