Understanding the business intelligence (BI) portfolio approachDate: Apr 29, 2010
Building a formal business intelligence (BI) strategy may help your organization get more value out of BI software. But how do you go about creating an effective and manageable BI plan? One possible option is a BI portfolio approach. In this video interview shot at the recent TDWI World Conference and BI Executive Summit in Las Vegas, Jill Dyche, a partner and co-founder of Baseline Consulting and an expert on BI and data management technologies, explains the portfolio concept and discusses how using it to prioritize the deployment of different BI applications can make life easier for organizations that are implementing new BI tools.
In the six-minute video, viewers will learn about the basics of the BI portfolio approach and hear Dyche’s take on why companies should consider it as a BI project management technique and as a component of their overall BI strategies. Viewers will also learn how widely BI portfolios have been adopted and get actionable tips for managing a BI portfolio, including advice on selling the concept to business executives and on how to overcome potential issues such as dealing with multiple best-of-breed applications, eliminating data silos and avoiding lengthy project backlogs. And if you haven’t done so yet, check out part one of our video interview with Dyche, with tips and advice on building a BI strategy.
Table of Contents:
• Understanding the business intelligence (BI) portfolio approach
Read the full text transcript from this video below. Please note the full transcript is for reference only and may include limited inaccuracies. To suggest a transcript correction, contact email@example.com.
Understanding the business intelligence (BI) portfolio approach
Craig Stedman: What's the BI portfolio approach, and why do you think it's the way to go?
Jill Dyche: Basically we came up with this concept of the BI
is essentially the business capabilities incrementally deployed to an
organization over time. What that means is that it is not just about
information, it is not just about BI tools, but it is the coupling of the
information with the business capabilities deployed to a cross-section
of the business. What we are essentially doing is we are defining a BI
strategy around a set of business capabilities, then we are letting
those capabilities and the data that enables those capabilities inform
our pipeline, so essentially, the strategy becomes the development
pipeline. Executive sponsorship, the system development life cycle for BI,
even things like capacity planning at a very low level become very
obvious because we are actually incrementally deploying these
applications in the portfolio, and planning along with them, resourcing
along with them, and budgeting along with them. It becomes a very
straightforward incremental development approach, and it also fosters
this idea that business intelligence is ongoing.
A lot of executives still see BI as a project or a set of very finite
projects, and they want to know, "when are we going to be done?" The fact
is that if BI is really, really effective in an organization, it will
never be done; it will evolve as the organization evolves and as its
strategies evolve. That is the goal that a lot of companies want to
Craig Stedman: How should companies get started on creating a BI portfolio approach?
Jill Dyche: If you are interested in it and you want to start, something to do is
to actually engage some cross-section of business stakeholders in the
metrics for success. It does not have to be a BI specific
conversation, but what we are advocating is let us define the metrics
for success, for successful initiatives in this company. If we
understand what makes individual programs successful and our company's
very culturally specific, very historical in nature, we can actually
then align BI to those success metrics. It is a great way to reengage
and recollaborate with the business, so that is a great first step.
Craig Stedman: Is it hard to manage a BI portfolio process?
Jill Dyche: The Portfolio should actually simplify some of those decisions, to
your point about the different best-of-breed tools, for instance. What
the Portfolio does is business requirements are no longer defined at
the level of the data warehouse or no longer defined at the level of
the platform; they are defined at the level of the application, so
each application may need its own toolset. One of the discoveries we
are making with all this market consolidation is best-of-breed does
not necessarily mean multi-vendor anymore. With all that consolidation
going on, maybe you have a mixture of Cognos capabilities and SPSS
capabilities, from IBM's Smart Analytics stacks, for instance. It does
not necessarily need to be multi-vendor, but multi-function is really
the secret to that. In terms of your question about the backlog, the other advantage of this Portfolio approach to BI planning is that it allows us to make very deliberate choices about
what we are going to resource. The only reason we would have a backlog
is if we did not have the funding to actually resource the
applications that were in our pipeline, so there would be a backlog.
Otherwise, we can work on more than one at the same time using that
common data infrastructure on our data warehouse.
I think the key with the Portfolio and the key to BI projects in
general is that we are using that common, reusable, clean, consistent
information that has been modeled, designed and that coexists
together in an integrated way across all of these business processes
and applications, so there is huge economy to scale.
Craig Stedman: How do you get business executives to be onboard with the BI portfolio approach?
Jill Dyche: The first way is to show them the money. Show them, first of all, what
the cost savings are going to be by using that common, reusable,
sharable data, and how much money that can save, not only their
departments but the organization at large. The other way is to
actually show them that you will be measuring value on their behalf.
Value and return-on-investment is no longer going to be measured just
at the level of the platform but at the level of every application.
In the case of a marketing executive who says, "We have our own
marketing data mart and we are not really interested in playing." You
can say, "With the information we have on the data warehouse, not only
can we tell you what a customer's cost to serve is, we can tell you
their profitability and lifetime value with this additional
information. That will results in marketing return on investment,
result in return on marketing investment and a higher campaign
response rate, so that is new revenue. You guys are being measured on
revenue, and we can guarantee that revenue because you are going to be
using a cross-section of enterprise data, not just your own siloed
Craig Stedman: Any final thoughts on business intelligence and the BI portfolio approach?
Jill Dyche: It is interesting. This week, at TDWI, we are hearing a lot about
executive sponsorship again, and that is a concept I think that had
gone away for a little while. There was a lot of cynicism around
executive sponsorship, but I think that companies are actually
starting to reengage the business and reengage some decision makers
around BI. With the Portfolio, one of the other things we find, is
that the executive sponsor needs to be specific to the application,
and that if a successful BI program will actually have multiple
executive sponsors over time, and that has been a real 'a-ha' moment
with a lot of the attendees in these workshops this week.
Craig Stedman: That is all the time we have today. Thank you again to Jill for
speaking with us, and as always, thank you to our viewers for turning
in. Until the next time, this is Craig Stedman for TechTarget