Interview

The five pillars of project

management infrastructure – An
interview with Glenn Ferrell
Glenn Ferrell
has over two decades of technology and manufacturing experience, the majority of this spent with RR Donnelley’s Technology
Center. He has managed projects in industrial energy, competitive intelligence, robotics, continuous improvement and software
development, including Donnelley’s first production Digital Asset Management system. Additionally, he has served in a series
of Director roles, managing Digital Content Technology, Process Technology, Digital Systems and Automation, and also on RR
Donnelley’s Manufacturing, Continuous Improvement and Marketing Councils. While managing project managers and Black Belts
involved in everything from Software Engineering to Six Sigma and Lean initiatives. Ferrell also served as an internal consultant
in new business development and technology strategy, annually presenting to Donnelley’s executive team. He has degrees in
Architecture, Engineering Science and Computer Science and is a certified Project Management Professional (PMP). He lives in
Glen Ellyn, Illinois, where he consults on Project Management and Process Infrastructure, serves on the Board of Overseers for
Illinois Institute of Technology’s College of Applied Technology, and occasionally plays Old Time music at various historical venues.

ABSTRACT How has project management become especially relevant in the business
context of today? Michael Moon interviews Glenn Ferrell on the history of project man-
agement, strategies for executive decision support and project portfolio management,
possible applications of prediction markets and on the problems of innovation and
knowledge/history retention during an economic downturn.
Journal of Digital Asset Management (2010) 6, 83–96. doi:10.1057/dam.2010.9

Keywords: project management; project management infrastructure; innovation; lean;

Six Sigma; scheduling

MM: We are here with Glenn Ferrell. Let’s companies who are just starting to form their
start off with a little bit of your background. strategies or who are simply struggling with the
GF: Michael, I’ve managed projects across problems of how to keep their projects on track
several different fields including industrial energy and how to get the best return on their overall
systems, software development, industrial project portfolio investment.
automation and process improvement. I’ve also The statistics I’m seeing indicate that only
worked in new business development, about 40 per cent of projects succeed. Given
technology strategy, competitive intelligence and the tremendous demands on a company’s
on marketing and manufacturing councils, as bottom line in this economy, that just doesn’t
well as doing a couple of stints on Balanced work.
Scorecard teams. I spent the majority of my MM: It sounds like a great place to start. Glenn,
career in the RR Donnelley Technology if you would give us a short historical survey of
Center. project management, its roots and perhaps some
MM: Right. Perhaps you can bring us up the of the major phases which it has gone through,
speed in terms of your current practice, what bringing us up to the current day?
you focus on, and the types of engagements and GF: Well, there are a couple of different ways
Correspondence:
the details. to attack that. A history of projects is really sort
Glenn Ferrell GF: Currently I consult on project of the history of civilization. It is a history of
263 Sheffield Lane,
Glen Ellyn, Illinois 60137,
management, project management infrastructure the way people interact in groups. However,
USA and project portfolio management strategies. I’m when project management people talk about the
E-mail: gcferrell@aol.com
particularly interested in working with smaller history of project management, they are usually

© 2010 Macmillan Publishers Ltd. 1743–6540 Journal of Digital Asset Management Vol. 6, 2, 83–96
www.palgrave-journals.com/dam/
 Moon

talking about project management tools and the or stakeholders, we have the associated problems
way project management tools and processes of communication and division of labor in a
have been used. large group.
MM: Before you go further, I’d like to examine So the type of project we are interested in is
the concept that projects represent another temporary, delivers unique results, is
mechanism of culture – how we propagate from progressively elaborated, is cost constrained and
one generation to the next, a set of values, requires coordination of many stakeholders.
norms, expectations, beliefs and behaviors. And MM: Okay.
so, like great art or food or architecture, project GF: So on to the history, tools and mechanisms
management is right up there in terms of of project management. Since projects are
another vehicle mechanism of culture by which temporary and have start and end dates, the first
we kind of create the future. tool we have to have is scheduling. Scheduling
GF: Very well said. has a lot of military history. Sun Tsu discusses it
MM: So when most people talk about projects, in the Art of War and Xenophon discusses it in
they are talking about project tools or project The Persian Expedition. The complexity of many
management tools, correct? projects of the ancient world implies that they
GF: Probably so. But to really understand used diagrammatic tools and graphics. Julius
project management I think we should probably Caesar built a bridge across the Rhine in
start with defining what a project really is and 14 days, while outnumbered and in hostile
what it isn’t. Then we might touch on history territory. To accomplish that, there had to be
and work our way up to tools. some guy in a tent with ink, parchment and
So first, as you know, projects are temporary abacus doing some pretty sophisticated
and always produce unique results. Those results scheduling – and he probably wasn’t just adding
are generally a new product, a new service or columns of Roman numerals. But graphics in
some type of cost reduction. Projects are not general don’t survive the copying process.
ongoing operations. If something repeats and/or Scribes learned to copy characters but
creates non-unique results it is not a project. It archeologists believe that very few learned to
is an operation. draw, so the history of those ancient tools is
Secondly, projects are always progressively pretty much lost.
elaborated. At the beginning of a project we So we have to wait until modern times to
really don’t understand the scope or the details talk about graphical or diagrammatic scheduling
of how the thing is going to be executed. That tools.
understanding develops as we move through The first important one was the Gantt chart.
analysis, design and implementation. Also, as this Frederick Taylor developed scheduling boards in
understanding develops, the outside world is not the last half of the 1800s. You still see these in
standing still. So, scope changes are normal and a lot of factories. Henry Gantt worked with
part of a project that must be managed and Frederick Taylor in the steel industry and then
integrated effectively. So progressively elaborated further developed these ideas while he was
means that as project details become increasingly working in ship-building during World War I.
specific and as scope changes are requested and After the Gantt chart was instrumental in the
approved, the project becomes increasingly success of the Hoover Dam in the ’30s and the
detailed and complex. Interstate Highway Project in the ’50s, it
The third point is that projects are always became a standard part of the project
cost-constrained. I have never seen one that management toolkit.
isn’t. Also in the mid-’50s, we see CPM (the
And finally, although there are single-person Critical path Method) and PERT (Program
projects, when we talk about project management Evaluation and Review Technique) charts being
we are talking about managing multiple developed simultaneously. Dupont was looking
stakeholders. By stakeholders, I mean the project for something useful to do with a Univac and
team, customers, vendors, contractors, executive they developed CPM in conjunction with a
management and so on. So, since the projects naval special projects office. This improved on
we’re concerned with have multiple participants the Gantt chart by allowing activities to be

84 © 2010 Macmillan Publishers Ltd. 1743–6540 Journal of Digital Asset Management Vol. 6, 2, 83–96
 An interview with Glenn Ferrell

organized in a network so that the project progressively elaborated then, as we move left to
manager could focus on the critical path within right down the timeline, complexity is
that network. Despite the fact that CPM increasing and there are always unknowns to the
reduced Dupont’s project shutdowns by about right of wherever we are. Complexity and
25 per cent, they abandoned it, probably unknowns imply a risk of failure.
because of a management change around 1959. Risk management is fundamentally based on
Somehow CPM as a tool survived, but this probability theory. The beginnings of probability
story always makes me wonder if there is a theory start in the renaissance and then are
graveyard of good tools out there that were further developed by Blaise Pascal when he
abandoned because of management changes or starts to think about gambling. Christian
just applying the wrong tool to the wrong Huygens, a friend of Pascal, wrote the first book
problem. on probability theory in the early 1600s – again,
MM: Right. Glenn, what is contemporaneous on the subject of gambling.
with this is that we have a whole branch of Huygans was Dutch and during this period
mathematics that began to support these modern the Dutch and English are involved in a lot of
advanced notions of logistics and operations trading, commerce, shipping etc, not to mention
research. Am I right? a war or two. About 30 years after Huygans’
GF: Right. Statistics and probability book there is this coffee shop in London where
mathematics are probably the most important. It captains get together to exchange information
may be a good time to talk about the probability on hazards, routes and weather. That shop is
when we get into risk management – one of ‘Lloyds of London’, which gradually evolves
the elements of project management that we into the great insurance company. Insurance
want to discuss. companies largely become the green houses
MM: So, in the context of the history of where probability theory develops up to the
projects and project management, looking at point where we see it today.
tools and methods, we have identified two MM: Which is to say, when lots of money
traditions. First, we have operations research and became involved, the practice really accelerated
logistics, how to ensure when you put 50 000 in terms of both the mathematics in the
soldiers into a battle theater, you got enough underlying theory as well as various applications
food, tents and bandages. And, second, from a of that mathematics.
branch of applied mathematics and rigorous GF: Exactly. Another Dutch innovation is
ways of synchronizing scarce resource with ‘options’. Risk was already being hedged with
better schedules and deployment configurations. options in Amsterdam by the 1700s. Project
GF: We are in the 1950s, and the Navy has the managers still have to hedge their risks. For
challenge of building the first nuclear submarine, instance, if you are dealing with a foreign
the Polaris. They bring in Booz Allen and project, it involves some kind of currency
together they develop the PERT technique. exchange. Your job is not to speculate in the
The PERT technique provides both currency market. Your job is to hedge your risk
estimating and scheduling capabilities. So, it adds so that the fluctuations in the currency market
this estimation capability to the scheduling do not affect the outcome of your project’s
capabilities that CPM provides. Today, people financials. So the concept of options plays a part
use these terms interchangeably, but PERT has in risk management.
estimating methods that are used by many MM: Before we get in to the application of
project management professionals. these historical terms and developments to
Critical Chain Project Management (CCPM) business, I wanted to circle back on one area,
adds resource flexibility between tasks, which because it is such a rich conversation that
can sometimes accelerate the overall schedule. I wanted to give it more than just a passing
This technique tends to be more dynamic than notice. That is the notion of scheduling.
PERT/CPM. It is relatively new and evolved Scheduling, I imagine, has many different
out of Goldratt’s Theory of Constraints. implications if it is related to a project. For
That covers the most significant scheduling example, let’s look at the difference in
tools. Let’s move on to risk. If projects are scheduling for a project, which you indicated as

© 2010 Macmillan Publishers Ltd. 1743–6540 Journal of Digital Asset Management Vol. 6, 2, 83–96 85
 Moon

something has a beginning, middle, and ending Processes at the same time, in order to make
and a unique result – versus scheduling for economic tradeoffs.
operations, which you described as structured, MM: Now, this has profound implications as
ongoing and repeatable. Can you elaborate a we look at digital supply chains and digital
little more on scheduling and how it relates to systems where we have business process
project management infrastructure? management platforms that facilitate dynamic
GF: Okay, first off, projects deal with scheduling – as well as practically self-organizing
uncertainty so, for example, every task in a projects.
PERT chart has an optimistic, pessimistic and GF: Right. Individual project scheduling is
realistic duration, and there is specific always dynamic within the constraints of end
mathematical process by which you estimate the dates and milestones. But programs become
overall duration of the critical path. Operations more dynamic when the external environment
schedules have less uncertainty and generally changes. So, rapid changes in the economy,
don’t do anything this complex. new technologies, mergers or acquisitions cause
Secondly, a good project management projects to be dropped or reshuffled within
infrastructure not only supports improvement of a program. With all that change it would be
project management but also supports almost impossible to make informed executive
abstraction so that executive decisions are easier decisions without the recent developments in
to make. This abstraction includes schedules. So, online project analytics. In a company with a
ideally, subproject schedules are ‘nested’ in solid project management infrastructure, these
project schedules, projects in programs and so project analytics systems allow management to
on, so that you can look at things at higher drill down through these layers and focus
levels of abstraction as you go up to the big attention on the most critical areas.
picture. A project manager will create a MM: So that would be the third large vector?
schedule that is made up of activities which are GF: Exactly. We think about this ‘big picture’
connected in a network fashion. However, activity as being performed by executives but, in
a program manager will deal with multiple keeping with our discussion on history, it used
projects and will probably want to relate all to be performed by monarchs. There is a chart
of those projects, perhaps in some kind of in one of Tufte’s books from the 1780s showing
a high level Gantt chart. Then if he wants the balance of trade in England. On the facing
to drill down on those projects, he probably page there is a horizontal bar chart which drills
wants to look at elements of the work down specifically on imports and exports to
breakdown structure (WBS) instead of actual Scotland. So Monarchs could drill down from a
activities. For the program manager, these WBS big picture to detail just like executives can
elements are containers for all the underlying today. The chief differences are speed, accuracy
activities. and flexibility.
So, initiatives can be organized in several MM: And perhaps, as you have just said, it
abstraction layers: Program, Project, Sub-project, represents someone who has responsibility for
for example. A large corporation may also have the realm, someone who has an overall sense of
some type of construct on top which contains accountability for making sure that that things
Programs and Processes. In this context, a Process is move through his realm in ways that are most
a specific manufacturing process like material advantageous to the crown.
handling, pre-assembly, finishing and so on. So you could say that this is the digital lineage
These processes are operations-related and a of project management, the first lineage being
process manager may manage both a single centered on organizing the logistics of warfare
manufacturing process and a number of projects and then later, on the industrial operations.
which are individually intended to make some Second, we have the mathematics of the
change or improvement to that process. This is underlying theory that enables one to calculate
usually where the Six Sigma and Lean projects schedules, probabilities and so on.
go. For companies that organize their projects Then the third, originally designed, as you
this way, the goal would be for executive said, from the point of view of the king looking
management to be able to see both Programs and at his realm, but then increasingly senior

86 © 2010 Macmillan Publishers Ltd. 1743–6540 Journal of Digital Asset Management Vol. 6, 2, 83–96
 An interview with Glenn Ferrell

executives who are concerned with overall For this to work, a project must go through
performance of an organization, not necessarily some kind of a ‘stage-gate’ process which
the status of a project or program. allows you to decide at each individual stage
GF: Right. That is a fair summarization, whether to kill it or continue to fund it.
although we might want to just mention two That stage-gate process has to be managed by
more components of the project management someone – a sponsor, a program manager or
process. a project management office. The stage-gate
First, the fact that a project is cost review is kind of like a submarine surfacing.
constrained leads us to issues of cost accounting – Between reviews, projects tend to ‘run under
that is, cost tracking and estimating and the water’ and to be much less visible to
budgeting because there is a whole raft of tools management.
underneath those areas which have very rich And there has to be some kind of
history. For the most part these tools have been infrastructure that allows the training of project
appropriated by project management from other managers, the training of sponsors and the
disciplines. training of program managers, so that this
MM: So, if you will, that would be the fourth process is smooth and works effectively. The
pillar of project management infrastructure is the group generally manages both stage-gate reviews
cost and the time, cost, and the money – which and this project-process training is the PMO.
you could say in this construct money really But there are many different ways to do this
represents discrete units of time. and, in smaller companies, there are ways to do
GF: Right, and saving time, or getting to this without a full-time organizational construct.
market earlier with your product or MM: In the last 15 years or so, we have seen
improvement could mean either increased the flattening of organizations. Some people
revenue or increased savings or both. Then the have called it the hollowing out of middle
fifth pillar would be related to managing and management, which from a macroeconomic
communicating with all of the stakeholders perspective has the job of filtering, refining and
including project staff, contractors, customers, transforming data and the information and
vendors, operations management, executive information and insights that then become the
management etc. basis for management and leadership to lead and
MM: Fabulous. So there we have the five pillars manage the organizations.
of projects, project management and therefore GF: Right.
the foundation of a project management MM: So the flattening of organizations is now
infrastructure. accelerated by recent economic troubles. What
GF: Right. role does project management have in helping
MM: How about if we take that then into a companies adapt more quickly to the changes,
business context? Specifically, how has project many of the external and macroeconomic in
management become especially relevant in the nature?
business context of today? GF: Okay. When markets shrink, as they are
GF: Well, one of the things that we have seen shrinking now, a company needs a big picture
within the last couple of decades is the growth perspective in order to make intelligent
of PMOs around the country because managers tradeoffs. We talked about a high-level view of
have begun to recognize what I mentioned both programs and processes before. In a shrinking
before – that only about 40 per cent of projects market, it may make more sense for a company
actually achieve their objectives. Ideally, a well- to cut a project or program that is developing a
run Project Management Office (or Program new product or service – rather than make cuts
Management Office) (PMO) should increase the to projects which are supposed to cut cost from
probability that an individual project will an existing ‘Process’. So this should be an
succeed and increase the visibility into the risk executive’s first focus – can the market still
that an individual project will fail. This is absorb the product or service this project is
important so that failing projects can be shut producing. However, as the situation worsens,
down early and resources can be reallocated executives are faced with cutting process
somewhere else. improvement projects and finally (or

© 2010 Macmillan Publishers Ltd. 1743–6540 Journal of Digital Asset Management Vol. 6, 2, 83–96 87
 Moon

simultaneously) with thinning middle MM: So, if we start to develop the business
management ranks in order to survive. context of project management, we see:
This creates challenges. I have talked to
several directors who now have a 100 and 1. Companies are still doing projects and in
more direct reports whereas before they had many cases they are doing more projects,
anywhere from 10 to 30. With this many generally with a shorter duration, by larger
reports, a director has no time to mentor and numbers of people.
coach project managers and managers. But the 2. Companies still structure repeatable programs.
more critical problem is that they have a great
deal of difficulty tracking what is going on. GF: Right, although the current economic
MM: Glenn, it is even exacerbated by the fact situation has undoubtedly resulted in a fairly
that we have e-mail and remote offices, with massive project culling, that is the long-term
telecommuters and so on. That makes the trend. And you are also right that companies are
whole notion of management even more many times embedding repeatable operations
challenging because you know you are not activity under the name of ‘projects’. That’s
physically sharing the same space. very cost inefficient and it tends to unnecessarily
GF: Right. And one of the problems that is at inflate the overall cost of a project management
least partially solved by good project infrastructure.
management is communication overload. Another effect of the current economic
Because as you are aware, communication tends situation is that there are certain categories of
to increase exponentially as the number of projects that are shrinking and types that are
individuals in a particular communications growing.
context grows. A good project manager controls MM: Such as?
communications by putting processes into place GF: Well, until markets start to grow again, the
that structure those communications so work first category of project I think we are seeing
can proceed effectively. Actually, stage-gate culled in this current crisis is new product
reviews contribute a lot to this goal because development. Then as the economics continue
they help minimize ad hoc project reviews that to get more challenging we will see (or in many
drain project resource. cases, have already seen) the culling of process
MM: So, we want more than to just control it. improvement projects, Six Sigma, Lean etc.
Rather we want to add information flows, so Then, as we discussed, comes the organizational
that the remaining flows are of the highest flattening, elimination of middle management etc.
quality and/or applicability in terms of the MM: And why is that?
recipient? Right? GF: In my experience, the product
GF: Right. Communication control is necessary development projects are culled first because
in order to aggregate and summarize those markets are collapsing, either due to the
flows, and now I think we are beginning to economy or disruptive technologies. Let’s look
come to the next level. If projects are still being at the print market. Web and cable TV were
managed by project managers in a sort of an disruptive technologies to the print industry.
eclectic way, where project tracking, estimating When those technologies started to absorb
and scheduling are all done in fairly unique advertising dollars the print markets had not
ways by each individual project manager, then really begun to collapse yet, so the printing
summarizing progress for a higher-level view industry responded by initiating a raft of new
becomes very complicated. Does that make technology projects. These projects included
sense? everything from bar-coding cell phones for
MM: Makes complete sense. catalogue ordering to printing circuits on
GF: Because of this complexity, the lack of paper.
middle management layers and because of the Lots of interesting technologies were either
sheer labor or aggregating and summarizing being discussed, were in an R&D stage or in
progress, the information passed to management real development. Then the market truly
is generally of very poor quality. This makes it began to shrink and those technology projects
more difficult to see major risks and problems. (new product/new process projects) were culled

88 © 2010 Macmillan Publishers Ltd. 1743–6540 Journal of Digital Asset Management Vol. 6, 2, 83–96
 An interview with Glenn Ferrell

and that capital and cash was then routed to things are the way they are have either left or
continuous improvement projects that were are leaving the company. This is a huge loss of
intended to cut cost – implemented through Six knowledge to the company.
Sigma and Lean type methodologies. This made With regards to the overall IT project
sense but required a new process management portfolio, another important issue arises. The
infrastructure, complete with training, process loss of CEOs, executives, middle management
managers, Black Belts etc. The problem was the means a huge loss of knowledge to the
market collapse was accelerating and the company and, more specifically, a loss of insight
continuous improvement projects couldn’t save into the project portfolio. It is quite likely that
enough cost to compensate. Any industry that the new management will want to reshuffle this
services rapidly shrinking markets must portfolio. If the acquired company had a PMO
consolidate. So the money for continuous (Project or Program Management Office) and if
improvement had to be rerouted to mergers and it still exists, you may have the information
acquisitions. necessary to make intelligent decisions about
So in this case, disruptive technologies that portfolio. However, if no ‘knowledge
resulted in new product programs. Those were repository’ like this still exists, there is very little
abandoned when markets began to shrink and information on which to base these decisions
funds were rerouted to cost reduction projects. and they tend to be driven by the internal
When the market collapse accelerated – forcing politics of the acquiring company. Additionally,
industry consolidation – cost reduction projects if there are now two PMOs there will be
were abandoned to fund mergers and significant and probably appropriate pressure to
acquisitions, which also further flattened merge them. This adds another variable to the
management layers. I would expect this same problem.
pattern has occurred in dozens of industries. To make intelligent decisions you have to be
MM: Makes complete sense – say you have able to see projects in a way that evaluates both
a particular project management infrastructure the financial and risk characteristics for every
that you must have in place in order to speed single project. Then you manage the whole as
post-merger of newly acquired divisions or you might an investment portfolio. You can’t
subsidiaries of the firm. A big project with do this without the appropriate infrastructure.
a lot of risk and upside! These are the basic problems with corporate
GF: Okay. ‘marriages’. There are many interesting strategies
MM: And ultimately where many mergers and for dealing with them. Sarbanes–Oxley is
acquisitions run into problems is around the IT obviously one of the greatest motivators for
infrastructures, the ability for IT infrastructure to getting this right.
accommodate that level of systemic change. MM: Right. Well, this sounds like a great place
GF: Absolutely. to then really build out in a more formal way
MM: Can you speak to that? what we call a project management infrastructure.
GF: Sure. Three things. First, at least in a So, what are some of the basic components and
typical large manufacturing company, I don’t frameworks for such an infrastructure?
think anyone truly understands the diversity of GF: From a process standpoint, you first need
an IT infrastructure before they buy the to have a project or program management office
company. So integration costs, schedules and which maintains the process infrastructure for
risks are almost always underestimated. project and program management. This office
Secondly, those systems were implemented to has to drive enough standardization so that
replace human processes that may have operated executives can have much more summarized
very differently from those of the acquiring view of projects throughout the company.
company. There can be profound differences in Remember that standardization is necessary if
data schemas between the two companies. It we want to aggregate and summarize
may not be practical or even possible to information across projects and programs.
translate all of this into a common data MM: So, a project management office then
architecture. And third, many of the only really does two things: one is to provide a
people who understand all of the reasons that common set of tools and frameworks for

© 2010 Macmillan Publishers Ltd. 1743–6540 Journal of Digital Asset Management Vol. 6, 2, 83–96 89
 Moon

defining and running projects and, perhaps just and-so has to be part of the conversation and if
as importantly, to provide a set of reporting he or she is not part of the conversation that
tools that produce uniform data information that will become a show stopper or at least stand in
can be aggregated and summarized up through the gears of progress’.
activities and tasks all the way up to business GF: Right. Good example. All our financial
process. data may be data-warehoused and searchable but
GF: Right. Additionally, they provide a great deal of important knowledge (including
educational services both to sponsors, project what you call stories) is only warehoused in and
managers and sometimes to executives. Also, the searchable by, specific people.
PMO is responsible for the project history or MM: Well, it also neatly syncs up with what
knowledge repository. Even though it is hard to you just said. It is like you have a data architect.
quantify, this repository has huge value to a We might even have corollary in digital asset
company. Unfortunately, these repositories management as it relates to metadata and,
generally either don’t exist or are not searchable. specifically, metadata associated not just with,
Financial data and many other types of data you know, rich media in a massive repository,
generally have a home in a data warehouse but other forms of metadata relate to
somewhere. This is not usually the case for information, IT service policies, configurations
historical project management data. Without of IT resources and so on.
having an organization like a PMO it is difficult And in particular, I remember having a
to retain project history. Without history, you conversation with one of the ‘wild duck’ at IBM
can’t learn from your mistakes or even make in the research group. We had a conversation
good decisions on how to proceed with new around creating a database for no other purpose
projects. than warehousing metadata and database schemas;
MM: Right. Glenn, that also gets to an idea I where you might treat metadata as simply
would like you to develop. I had introduced the another data type within the database.
idea of an organizational historian, you know the We really got going with the idea that
one of two people in a company who simply a metadata repository might enable one to
made it kind of their personal job to study, build a new application or repository through
observe and classify how we, in this company, do a set of queries to the metadata database,
things. These organizational historians know the thereby assembling new schemas so you could
ins and outs of the firm’s culture norms, lessons then understand or model of a new workflow
learned and how things get done. or business processes, as well as understand
In many respects, organizational historians how the new application or service would sink
become the softer, more feminine side to the up to existing applications and IT infrastructure.
harder, most masculine aspects of project Did I make sense, and if so could do go
management office that structure and codify a on that?
lot of those cultural norms and learnings. GF: Yes, the concept that comes to mind is a
GF: Yes, when you mentioned this concept cultural schema that could help a project manager
I immediately thought of data architects. You understand where to find relevant information
know, people who oversee an organization’s on the culture and history of a company. The
data and who know how they all fit together. Project Management Institute recommends that
These are very unique characters. They have an a project manager performs a historical and
understanding of how all of the data schemas cultural investigation of the company at the very
within all the databases in the company fit beginning of a project to improve his chances of
together and also how they map to the business success. Wouldn’t it be nice if there were
and to the business needs. I think it may be someone analogous to a data architect that he
almost as important to have people who consult to figure out where to start?
understand the how a company’s business MM: Right, and again, organizational historians
processes map together. tend to be the informal repositories of that
MM: Well, that is just a process, but the cultural knowledge.
unspoken norms as conveyed by stories right? GF: That is a very interesting concept
You know, ‘If you want to do that, then so- Michael, and I would love to get more insight

90 © 2010 Macmillan Publishers Ltd. 1743–6540 Journal of Digital Asset Management Vol. 6, 2, 83–96
 An interview with Glenn Ferrell

in what kind of processes a cultural historian importance of creating conversational microsites

might use. or portals where an organization has taken upon
MM: Well, I had it related to me by Dr Garry itself to talk to itself about how it does things.
Hare, now the professor of the School of Media Thus, while individuals will move on the ways
Psychology at Fielding Institute. He shared with of understanding and doing things remains set in
me that organizational historians are the a social, conversational context.
individuals who have worked in that particular GF: Good. So you institutionalize a little bit of
company you know 5, 10, 20 and 25 years. knowledge management to be able to absorb it.
Many do not have some meaningful executive MM: Right. And the mental maps by which we
position; in fact almost all were not in a collectively think and get things done – you
position of leadership or authority. could say that those mental maps are another
He went on to say that organizational form of unstructured metadata, as it relates to
historians position themselves in or around knowledge that lives in the human beings that
interesting things going on, like the deployment work there.
of the major new ERP system or a GF: Right. We have a challenge when we talk
transformation of the business process or a about collapsing organizations and collapsing
wholesale redesign of the factory floor or the middle management. If you think of people in a
institution of the new work real policy that is company as analogous to routers in a network,
enterprise-wide. they have access to many resources on a lateral
Some of them will make meaningful basis, who have organizational knowledge and
contribution as a member of staff to these who know how to go to those people and
projects, but unlike people who simply just got retrieve that knowledge. With the flattening of
stuff done, organizational historians kept detailed management hierarchies, many of the lateral
mental notes, in some cases physical notes of links among middle managers tend to break and
who did what and how and if not why not. become more narrowly vertically. As a result
GF: Right. It sounds like a fascinating role and, communications tend to only move up and
to play devil’s advocate a bit, it sounds like a down between management and their direct
role that is at great risk in an organization. For reports. The value of all of that horizontal
example, I managed a competitive intelligence interaction tends to get lost.
project for a Fortune 250 company. I took a MM: Right. It seems to me that it also then
team of cross-market marketing directors gets redistributed to your network of suppliers
(subject matter experts) and facilitated them and partners.
through the construction of a business/ GF: Yes, that is a good point.
competitive intelligence process that would be MM: And so it seems to me that another
generally effective across all of their markets. dimension of project management infrastructure
We documented the process, developed a is really how to access subject matter experts
database and applications and hired a manager to that may have worked for the time, but now
run the process. This worked very effectively have found gainful employment in the industry
for a couple of years. Of anyone in the and specifically among your vendors and
company, this manager had the best handle on partners.
all of the competition within the industry. GF: Right. That becomes an important goal.
However, when the industry was downsized MM: Glenn, in the spirit of really developing
and consolidated, he was one of the first to go. out a project management infrastructure, could
I think a cultural historian faces the same risk, you just bring us up to speed in terms of some
especially if there is not some other mission- major frameworks from models for management
critical role to which he or she is also attached. process-control frameworks such as Lean, Six
The value of cultural history is difficult to Sigma and things like that.
quantify. GF: Sure, I managed a group of project
MM: Well, I think that is true in one respect managers and Black Belts running both Lean
and yet I think collectively we have responded and Six Sigma projects when those efforts were
in a way that offsets many of the downsides of first initiated at Donnelley. Lean is an interesting
what you just described. In particular, the animal because although you can have lean

© 2010 Macmillan Publishers Ltd. 1743–6540 Journal of Digital Asset Management Vol. 6, 2, 83–96 91
 Moon

projects, lean also implies an overall GF: Right.

transformation of the organization. In this MM: And some of the ideas of Lean are simple,
transformation, lean is ultimately pushed down which is first to start off with an end-to-end
to the shop floor and becomes part of the depiction of a business process or set of
operations process. Generally this ‘goal’ of workflows, identifying who does what and how
achieving a lean organization takes many years it, and you know, what are the critical decision
to accomplish. A lean project may be formulated points or handouts.
to move an organization towards this goal, or a GF: Right.
lean project may be a project that simply uses MM: And then in the course of doing that and
lean tools to perform some specific process trying to eliminate as many steps as possible as a
improvement. function of looking at more intuitive or direct
So there may be a threshold to cross at some ways of getting the results produced in sequence.
point in time. When you initiate Lean programs GF: Right. One of the key Lean tools to do
in the organization, you usually initiate them this process-mapping and time-compression
with a series of pilots. These pilots are projects. activity is called Value Stream Mapping.
However, as your shop floor employees begin MM: And then in addition to that, once we
to become Lean practitioners on an everyday have a framework, we then start pushing the
basis, we have to make case-by-case decisions responsibility for controlling the flow down to
on what efforts should be projects and what the individual, such that you make quality his
things should simply be folded into operations. or her personal job; so that you really start
MM: Yes. optimizing with the flow from the bottom up,
GF: So lean is something which generally starts as opposed to simply by management fiat, top
as projects. To push it down into operations down; where the management forces people to
requires shop floor training programs and some make all manner of tradeoffs and, ultimately,
strategy on how to gradually operationalize the induces workers to produce more defects and
expertise and redirect the project management other negative unintended consequences.
organization’s attention to another area. GF: Right. That is an excellent summary. With
MM: So, what are the basic operating principles the capability of stopping the line, the individual
of Lean as you understand it, in this context? can ensure that the problem and perhaps even
GF: Well, the primary operating principle of the process are fixed before the product
Lean is to shrink inventory to nothing. To give becomes part of finished goods – and therefore
you an example: ‘print rolls’ in the print require more extensive rework. This is
industry. Traditional ‘logs’ (or large horizontal fundamentally the same relationship software
strapped stacks of signatures) take a great deal of quality assurance (QA) has to software testing.
labor to handle and feed into bindery hoppers. QA focuses on preventing the problem, while
A ‘print roll’ can be loaded with signatures testing typically catches it after the fact. Both
directly from the end of the press, and then it is Lean and software QA have a strong focus on
picked up by one fork lift driver who can move reducing cycle time. Both are based on Deming
it to a bindery hopper and, by herself, attach it. and the Japanese tools that implemented his
However, print rolls take much more space than principles.
logs and, being expensive, you can afford less of MM: Right. And we know that by reducing
them. Because of these two facts, once print the steps in any one process, you also reduce
rolls were installed at this plant they were forced cycle time. A reduction of cost can lead to
to implement Lean because there was greater outputs given the right infrastructure,
insufficient storage to store work in process. incentives and processes. Did I understand right?
The technology acted as a Trojan horse for a GF: Right, very good.
Lean transformation that stretched all the way MM: Then, Six Sigma is, if you will, a different
back through the supply chain. complementary approach?
MM: Lean, as I understand, it originally GF: Yes. Both Lean and Six Sigma are ‘process
originated in the application of Deming’s improvement’ approaches. Both have similar
principles around total quality management by waste reduction goals. However, Six Sigma is
Toyota. never completely pushed down to the shop

92 © 2010 Macmillan Publishers Ltd. 1743–6540 Journal of Digital Asset Management Vol. 6, 2, 83–96
 An interview with Glenn Ferrell

floor operator. It tends to remain a project should be very comfortable with managing
approach because the skill sets, including a a set of Six Sigma projects. The chief challenge
strong grasp of statistics, are difficult to master. would be that the project management skill set
So although the results of a Six Sigma project – of a certified PMP tends to be higher than that
the specific process improvement – will be of a certified Black Belt. The Black Belt project
continued by the shop floor crew after the management toolset tends to be a simplified
project is completed, it is unlikely that the shop subset of what is normally expected of a certified
floor crew has the ability to start a new Six project manager. Generally, this simplification
Sigma project. works fairly well for Six Sigma type projects.
Another difference is that the Six Sigma On the other hand, PMOs should probably have
DMAIC model (Define, Measure, Analyze, little to do with classical Lean initiatives. I think
Improve and Control) is, in practice, very these should probably be run entirely out of
analogous to the scientific method. So the operations.
model tends to produce a real cause and effect MM: Right. We have also seen a convergence
understanding of the problem. This means that of Lean and Six Sigma. Are there any other
the quality of the knowledge gained during the extemporaneous management process control
Six Sigma project tends to be greater than the frameworks that you find especially relevant to
knowledge gained during a Lean project. IT infrastructure readiness and/or project
The more technologically intensive and complex management infrastructure, in particular as it
the process to be improved is, the more I relates to IT service management and IT service
would lean towards Six Sigma. (However, there delivery?
sometimes comes a point – especially in IT GF: Yes. Are you familiar with Analytical
projects – where a Black Belt just has the Hierarchy Process (AHP)? AHP is used by the
wrong set of tools to tackle the problem.) Pentagon and by companies and institutions
Both Lean and Six Sigma processes require around the country at the very front end of
a training infrastructure. Both may require projects and programs and processes. AHP
various levels of certification. Lean training has significantly improves complex decision making,
to be implemented from shop floor through whether the decisions have to do with overall
management. Six Sigma has to be implemented company strategy or portfolios of projects. AHP
from Green Belt level through management. can be used at any level within the company.
MM: Yes. I’ve used this with teams of Business Unit
GF: So the training infrastructure for either Presidents and Senior Vice President (SVPs) on
Six Sigma or Lean may be organizationally a multi-million dollar company-wide program,
analogous to that implemented by the PMO. and also with engineers and shop floor managers
Likewise, the historical or knowledge on much smaller projects.
infrastructure of a good Six Sigma or Lean With AHP, a group creates hierarchies of
organization should be analogous to that of their goals and then, generally moving from
the PMO. It is just as critical to retain the parent to child layers, quantitatively weights the
history and results of all Six Sigma projects relative importance of those goals. Goals are
as it is to save the history and results of all IT then related to programs or projects; relative
projects. weightings are developed to determine the
By the way, I use ‘PMO’ as convenient degree to which specific projects drive specific
shorthand. In smaller companies, many of the goals etc. Ideally this starts at an executive level
functions of a PMO might be performed by and trickles down to lower levels where goals
other mechanisms and organizational structures. may be progressively elaborated. The
Additionally, I don’t think PMOs necessarily implementation of this has to be customized to
have to be run by the IT organization. the business, so it requires a trained facilitator.
To compare Six Sigma to other forms of Since Quality Function Deployment (QFD)
Project Management, the DMAIC model of is a critical component of AHP, it helps if
Six Sigma is very similar to the ‘stage-gate’ the facilitator is certified or at least trained in
approach that a PMO would implement with QFD. You need real QFD as described by
IT projects. A sufficiently sophisticated PMO Mizuno and Akao. Unfortunately, there are

© 2010 Macmillan Publishers Ltd. 1743–6540 Journal of Digital Asset Management Vol. 6, 2, 83–96 93
 Moon

‘misappropriations’ of QFD that don’t work shareholders, a best estimate as far as future free
well with this process. cash flows of one firm versus another, right?
There are also some specialized software tools GF: Right, right.
to help with this that easily save enough time to MM: Okay. So let us take that same concept
pay for the investment. and apply it to product optimization and media
It is hard to really summarize this. You have mix optimization. One of the problems that
to see it in action to get a real sense of its many organizations have today is where best to
power. Because the process is done in groups, spend my marketing dollars.
there is a kind of ‘honesty’ to it that tends to GF: Right.
override individual political agendas. MM: And as you can well imagine that
Oh! and one more thing about managing represents a rather significant and growing
project portfolios: at least one of the tools that conundrum for many companies. So we could
support AHP has add-ons that support the then create an options market around very
‘efficient frontier’ concept for a project specific media placements and/or points of
portfolio. This makes it easier for management engagement and then have a company, a large
to evaluate entire sub-portfolios of projects company and perhaps its trusted partners then
against one another. So let’s say you have a make bets in terms of which media mix,
fixed budget for your overall portfolio and too product mix, optimization will get the best
many projects, this helps to arrange projects in results: lift in sales with lowest number of
groups and position them on a risk/reward returns and therefore the most profitable.
graph, so that management can decide which So, if you will, we are talking about creating
groups of projects to fund. a prediction market where we ask stakeholders
MM: Let’s expand on the notion of decision across the organization to make their best guess
markets or prediction markets – the application in terms of which media strategy will produce
of game theory and a system using incentives to the best results.
predict strategic outcomes. GF: So the market participants would be
In particular, would you speak to how stakeholders within the organization. Then,
prediction markets might relate to your I would think, for a small organization with
discussion of options analysis and value stream stakeholders who are ‘experts’ you might get an
analysis? And, perhaps, you can address the need output similar to that of the Delphi technique.
to get a sufficiently large group with a necessary MM: Let’s get to the underlying idea of
level of cognitive diversity within the group to prospectively discounting future free cash flows
make their best guess as to the probability of a as a way to calculate the economic value of an
particular event occuring. otherwise intangible asset. We might use a set
GF: Okay. I haven’t seen prediction markets of discount factors derived from a prediction
in action, although many current techniques market; where you have asked people to
use a subject matter expert-driven ‘voting’ estimate what would be the impact of this
technique to weigh relative risk, importance particular discount factor on our ability to
or potential duration. We mentioned AHP, achieve those revenue goals. A discount factor
QFD, but Failure Mode and Effects Analysis might include taking a fully operational DAM
and the Delphi estimating technique are other offline: how much revenue would the firm lose?
examples. GF: It is a fascinating idea. All you need are
MM: So, usually it is a point in time with a mechanisms to manage the prediction market
stated probability of that particular thing and a critical mass of knowledgeable ‘traders’
occurring. Then you have people use funny with a motivation to participate. I know that
money or micropayments of 10 cents, to make there have been some efforts to provide a
bets or buy warrants; that particular thing will ‘prediction market’ for project milestones, where
happen, right? the project stakeholders place bets on when
GF: Yes. these milestones will occur. I would guess that
MM: And so the stock market is one big at the project portfolio level, something like this
prediction market in terms of one company’s could be used to look at the risk of particular
best estimate among all of the stakeholders and external events that might affect the overall

94 © 2010 Macmillan Publishers Ltd. 1743–6540 Journal of Digital Asset Management Vol. 6, 2, 83–96
 An interview with Glenn Ferrell

portfolio. This might already be happening but have the need to streamline the process but
I’m not aware of it. Then there is the question don’t have the training or time to work on it.
on the degree to which the prediction market Business Model innovation is harder to talk
itself might affect the outcome. There’s about because an economic downturn can
definitely a lot to think about. disrupt current business models and force new
MM: As we are beginning to wrap up our thinking in this area. So we are still seeing
interview here, would you speak to the idea of innovation in social media, mobile devices etc.
innovation leadership and specifically what you and there are probably many other innovations
see emerging as the next practices of we haven’t seen. Also, some of these Business
innovation? Particularly in these constrained Model Innovations can drive new Product and
economic times, with the emergence of these Process Innovation.
project management infrastructure? Now, do project management infrastructures
GF: Although there are always bright spots, support or hinder innovation? I don’t think that
Apple comes to mind, I think currently we are ‘fostering innovation’ is an important part of the
seeing a decline in innovation, you know, not mission of project management infrastructures. I
just a lack of innovation, because of these think they are more focused on accomplishing
collapsing management layers and because the the goals of management as effectively,
communication lines are getting more vertical efficiently and economically as possible. That
between reports and management. Too much being said, there is a lot that enlightened leaders
workload and too little time to think out of the in this area can do to create a context for
box. innovation. Providing training that helps people
MM: Could you compare product innovation, ‘think out of the box’ really helps in this area.
which largely is a function of exploiting new Help people see relationships – bring in
OEM technologies from your upstream speakers, establish cross-fertilization relationships
suppliers, and process innovation, which often with companies outside of your industry,
times remain internal to the business. And if introduce folks to lateral thinking, mind-
you can, address business model innovations and mapping, influence diagrams, data flow diagrams
social innovations – all within the context of etc. Introduce your best engineers to Triz. Get
projects and how project management some of your best analysts certified in QFD.
infrastructure supports or hinders innovation. Make sure things like strategy maps, if they
GF: Which makes perfect sense because exist, are getting shared down the right level. If
I believe innovation does not happen unless a they don’t exist do some strategy mapping to
context either exists or can be created for it. get a sense of the big picture. In my experience
Let’s start with product innovation. There are the budget for all of this tends to be in R&D
many examples which indicate that innovation organizations but I think, in a sufficiently
only occurs when a need exists. For example, enlightened organization, it could possibly exist
water power was not extensively used to drive in a PMO. In the future, something like this
mills, lathes etc. until the Plague had wiped might be budgeted in an ‘Innovation Office’
out a significant percentage of the manual run by a ‘Chief Innovation Officer’.
laborers in Europe. In the current economy, In general, consumers have reigned in their
consumers have pulled back spending and the spending because the financial crisis has reduced
number of Stock Keeping Units (SKUs) is both jobs and personal wealth.
being cut back at retail and commercial levels. And so, companies have become overly
This tends to indicate that there is less need for survival-oriented. You might look at Maslow’s
new products and that the context for product needs hierarchy and position a company in that
innovation is just not there. With Process model. Many if not most companies have been
Innovation, I think the context may be that down in the survival and basic needs area.
(1) a need exists and (2) tools or methods to Getting to the social needs area, which is next
explore that need are accessible and (3) enough level up, is a challenge. Classically the
‘space’ has been created to allow time for that organization who is the protector of the social
exploration. After layoffs, when two or three needs in a company is the HR organization.
jobs have been folded into one, employees After all of this organizational flattening the HR

© 2010 Macmillan Publishers Ltd. 1743–6540 Journal of Digital Asset Management Vol. 6, 2, 83–96 95
 Moon

organization is more ‘webified’ than it’s ever functions, but I think that we are going to
been. So, employees interact with their HR see a lot more activity in the monitoring and
organization through a screen instead of through controlling functions, as a direct result of the
a human being. That is a big concern for me. flattening of management layers. That means
So, I think before you can get to innovation, more or better PMOs – or the organizational
you have to be able to move up through that equivalent – to standardize tracking and
social layer up into the self-actualization layers reporting processes and the implementation of
and you cannot do that if you are focused on web-enabled project and program analytic
survival and on basic needs. systems to aggregate information to higher
As job losses bottom out and the economy levels of abstraction and allow drill downs to
begins to grow, then we should begin to see specific areas of concern.
some opportunities for putting more innovation We discussed decision support. In general,
process in place. However, I’m guessing that our decision-making processes are fairly poor
will take the form of new or transformed and emotionally driven. I know this is an
organizational structures. Praveen Gupta has unpopular view. There are tools and processes
done some very nice thinking in this area. that make a huge difference in the quality of
Perhaps we build part of that structure around executive decision making and in the buy-in on
the new role of a ‘Chief Innovation Officer’. those decisions after they are made. I think we
MM: It kind of goes to something that I are just on the threshold of the opportunities
remember reading, you know, one of the great these tools provide. This is a good function to
books by Edward Deming in which he says the be driven out of a corporate Strategy Office or
level of innovation in a company and virtually by a Chief Innovation Officer. IT is not the
correlates to the amount of fear in the company. place for this.
‘The more fearful a company, the less I think new product development has to
innovative it is’. come from a stronger economy and stronger
GF: I agree completely. balance sheets. I’m hoping unemployment has
MM: And it really kind of goes to the notion that bottomed out so that economic growth over the
innovation inherently has some risks to it, but next few years will allow us to collectively
the risk is largely related to change in the status climb out of ‘Maslow’s basement’ (survival
quo and therefore a lot scrutiny. I wholeheartedly mode.) As we get more revenue growth I
concur that some companies are in a better would expect we will see some new
situation to be able to innovate as a function of organizational structures and processes put in
being simply less fearful or higher on the high place to better support innovation. I would like
act of actualization. to see more Chief Innovation Officers. I think
GF: I agree. Although a good deal of the it takes this level of professional to help drive a
innovation investment may not have a more nation-wide dialog on this subject. We
measurable return, companies who don’t invest need to re-examine our educational system here
in innovation can’t adapt to change. too. In a global marketplace innovation is a
MM: Any summarizing comments? national issue.
GF: Well, we talked about the history and MM: Wonderful. Great place to conclude.
some of the tools of project management. Most Thank you so much.
of that history has been focused on the planning GF: Thank you, Michael. It was fun.

96 © 2010 Macmillan Publishers Ltd. 1743–6540 Journal of Digital Asset Management Vol. 6, 2, 83–96
Copyright of Journal of Digital Asset Management is the property of Palgrave Macmillan Ltd. and its content
may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express
written permission. However, users may print, download, or email articles for individual use.