—
Let’s
Redefine
What’s
Possible
—
_
Critical Chain Project
Management (CCPM)_
The Executive Guide
to Winning Results
2
_
Critical Chain Project Management_
Projects, Programs and Productio
HQ
I
n the classic British 1980s TV comedy
Yes, Prime Minister, senior civil servant,
Sir Humphrey, enjoyed running verbal
rings around his boss, PM Jim Hacker.
Although nominally in charge, Hacker’s
pride usually prevented him from
admitting ignorance and truly grasping
a problem. Sir Humphrey, we knew, was
really steering the ship. Meanwhile,
Bernard, the PM’s private secretary,
would point out how actions might be
perceived from different perspectives:
‘I give confidential press briefings; you
leak; he’s been charged under section 2a
of the Official Secrets Act’.
If Hacker is the ‘executive’, then Sir
Humphrey is the ‘engineer’, to use two of
the sub-cultures identified by renowned
organisational psychologist Ed Schein in
his seminal book Organizational Culture &
Leadership. The third is the ‘operator’, or
the frontline worker. Each comes with his
or her own beliefs, values and underlying
assumptions. In a large organisation, the
executive’s implicit desire is: ‘I set the
course, you manage the project, they
deliver the work’. But is this the reality?
It probably feels more like: ‘I expect
results, you promise me a deadline,
they’re late again’.
In the manager’s shoes it might feel
more like: ‘I plan the work, you (my
team) wreck my plan, he (my boss)
promises miracles’. While the operator
feels: ‘I do the work, you (my manager)
get in my way, he (the boss) is clueless
about what it really takes’. From these
different perspectives, everyone is
trying, everyone feels squeezed and it’s
always someone else’s fault. Instead of a
well-orchestrated performance, it’s more
like three groups playing different genres
of music in different keys and time
signatures. No one really sees the whole
picture and the work is often a grind,
wasteful and unjust.
_
In this guide, you’ll gain
insight into the systemic
context and perennial
challenges of project and
portfolio management.
It aims to provide you with
a solid grasp of a wellreasoned
set of principles
for you to intelligently
engage in a conversation
with your PMO and other
key stakeholders about the
relevance of Critical Chain
to your organisation.
_
_
Critical Chain Project Management_ 3
delay
Burn
Revenue
Burn
Revenue
Burn
$+
-
$+
-
$+
Reality
in the Multiproject Environment
due date
late delivery of business benefit
cost overrun
opportunity cost of not being
able to deploy resources to
the next big idea
opportunity cost
Slide 15Part 1
Figure 1
Reality in the Multiproject Environment
of your project as a single entity—with
interconnected moving parts, on an
uncertain journey to accomplish a stated
goal. There are many milestones along
the way, but we only care about one thing:
delivery of what we promised.
Unfortunately, achieving the exacting and
consequential promise of the trifecta of
scope, cost and schedule is not the reality
for most executives. For a given scope,
each project timeline (below) has a due
date with resources burning up cash along
the way. Too often, delays push the due
date into the future, with associated cost
overruns. Worse, the revenue, and hence
business benefit of the project is delayed,
and there’s an associated opportunity cost
as every successor project gets bumped
along the queue.
_
What’s really the problem?_
Why projects are like spaceships
the other organisational elements that
go into supporting an ongoing ability to
successfully manage mission after mission.
And that means you are dealing with a
very complex system. Some things, such as
astronaut training, mission selection and
spaceship design are under your direct
span of control; others, like regulatory
regimes, government subsidies and
launch windows sit within your sphere of
influence and others, like the weather, are
completely ungovernable.
The big question is: when they do, why
are your projects failing? Is it because
you’re not seeing the whole? The parts
aren’t fitting together properly? Or are
you at the mercy of a complex operating
environment?
The answer is probably all three. But
let’s wind back our analogy a bit. While
assembling a spaceship is itself a project,
a project needn’t result in something
so tangible. It could be an app. Or a
conference. Or a piece of regulation.
For our purposes it’s helpful to think
A modern spaceship is a remarkable feat
of engineering. Most of the features would
be known to the NASA engineers on the
early Apollo missions: the structure, the
motor, the controls, the launchpad, the reentry
heat shields must all work together
to get the payload into space—and when
required—safely back again. This analogy
has three vital implications for projects.
One, a spaceship is a whole entity. It
only takes off when all the parts work
in unison; the rocket motor—no matter
how well engineered—goes nowhere on
its own. Two, every part of the rocket
must match the overall design of the
spaceship; you can’t switch a Saturn V
engine for an Ariane—even if the Ariane
engine is the better design. And three, you
can’t run a space program on spaceships
alone. By all means, the assembled parts
which constitute the spaceship create
a complicated machine, but you need a
space centre, an integrated supply chain,
skilled people, a regulatory compliance
regime, a willing buyer, whether
government or private sector, and all
We really only
care about one
thing: delivery
to promise”
4
_
Critical Chain Project Management_
What’s a smart
executive to do?
You’re chairing your regular steering
committee meeting and wondering at
the creeping sense of déjà vu. Costs are
ballooning, deadlines are blown and you
sense that estimates for both are only that—
optimistic guesses based on a rerun of the
last similar project or some just a wild guess.
A disturbing thought looms: What will your
competitors do while you wait?
Behind their earnest expressions of
commitment, your managers don’t really
believe their deadlines. Even the good
ones spend undue time running around
behind the scenes micro-managing, while
preparing variation orders for more time,
more resource or less scope. And on large
projects—the ones designed to realise your
grand future vision—the project end seems
so distant in time from today’s hustle and
bustle that it’s difficult to persuade anyone
of the priority associated with work of such
deep significance. You haven’t a moment
to lose, but they convince themselves they
have all the time in the world. The old joke
rings true: ‘How do projects get late? One
day at a time.’ Because, whether it’s today,
tomorrow, next week, next month or next
year—you’re always on the critical path. A
day lost on the critical path is as useless as
yesterday’s vacant hotel room.
You’re thinking, or at least hoping, ‘There
has to be a better way to manage the
execution risk of my most meaningful work.’
Because, if another way exists that’s better
for you, your managers and the frontline,
aren’t you obliged to look into it?
Understanding the art and science of the
discipline of Project, Portfolio and Program
management in any depth is not usually
the focus of a senior executive. Project
Managers recently achieved Charter
status in the UK and their profession is
now as much a discipline unto itself as is
finance, with their Chartered Accountants.
They have their courses, conferences,
accreditation and ongoing professional
development. As a specialist discipline, you
hope that the people you hire are across
Whether it’s
today, tomorrow,
next week, next
month or next
year—you’re
always on the
critical path.”
The real management
challenges
Once you have decided what the scope of
your strategic portfolio is, at the root of
your operating system, you have two core
challenges to contend with: resourcing and
scheduling.
Start with resources. You can’t just throw
more people at the problem. There’s an
optimum number required to do the work.
Introducing more people creates complexity
and communication issues while diluting
accountability. And you can’t simply throw
more money at it, as you don’t know what
your investment will yield. No matter who
you are, nor what organisation you work for,
the hard truth of limited resources ensures
you must prioritise, figure out the tradeoffs
and make tough decisions about what to
do and, just as importantly, what not to do.
Notwithstanding the odd bit of dumb luck,
superior outcomes derive from the extent
that you apply the power of reason to your
decisions.
On the scheduling side, how do you determine
the sequence you work in, the dependency
of activities, the prioritisation of work on
any given day, the complex choreography of
getting the right people, doing the right work,
in the right place at the right time?
If you’re using a ‘waterfall’ method with a
critical path of sequential steps, your due-date
is baked in. Everyone is confident—until the
project starts, because everyone knows these
plans, brittle as they are, don’t survive first
contact with enemy. With all the moving parts
and complexity, how could they? It doesn’t
help that we’re often our own worst enemy—
adding scope, switching priorities, or dragging
key resources onto competing projects or into
the daily maelstrom. As Mike Tyson famously
put it, ‘Everyone has a plan until they get
punched in the mouth’.
How to avoid that brittleness? One way is to
use the increasingly popular Agile method.
Which works very well for some things. But
the way Agile hits a deadline is to dump scope.
That’s right. You get your project on time, but
it’s not what you were led to expect. And your
customers’ idea of a ‘minimum viable product’
might be very different from that of the
people designing the product or service. And,
despite what its advocates would have you
believe, it doesn’t scale.
You can toughen up KPIs and tie bonuses
to performance, but working harder leads
to burnout and is not sustainable. It’s
reminiscent of asking nineteenth-century
doctors to save more patients by working
faster. The root cause was hygiene; doctors
weren’t washing their hands. A change in
the process led to immediate results and
many lives saved. So, let’s look at a process
of project management that would be
revolutionary if it hadn’t already been in use
for the past 30 years. It’s the biggest kept
secret in managerial leadership. And it’s called
Critical Chain Project Management (CCPM).
their accountabilities and that you can
empower them to get on with it. You want
to be presented with relevant and timely
data to support your sense- and decisionmaking—not
getting stuck in the weeds.
At your level of work, your accountability
is to make the big strategic decisions,
then control the work which will build
the organisation of five years into the
future and beyond. But how can you be a
responsible steward of the business and
an inspiring leader of its people if you and
your organisation rely on flawed means
to determine where to focus your limited
resources? What is your data really telling
you? What questions are you asking it to
answer? What insights do you get from
those questions? And what, then, is the wise
course of action?
_
Critical Chain Project Management_ 5
_
Goldratt’s innovations_
In 1984, physicist-turned-managementconsultant
Eli Goldratt published The Goal
and set off a seismic shock in production
management whose ripples are still being
felt today. Written as a novel, with a hero
who has to save his factory from closure,
the book proposed a new way of looking
at production through the lens of what
Goldratt called the Theory of Constraints
(TOC). It became a bestseller, selling
more than three million copies to date.
But although widely read, few companies
followed through. Many who ‘got it’ felt
the principles were just common sense—or
too simplistic to apply to their situation.
Those who implemented it found TOC to
be the most effective management system
for production since the Toyota Way. And
given that Jeff Bezos lists The Goal in his
top three business books, it has doubtless
influenced his thinking and Amazon’s
gargantuan success.
Resource Task Vendor Customer
Project
The Law of Large Numbers
[Central Limit Theorem]
Figure 2
The Central Limit Theorem: The Law of Large Numbers
The power of
aggregation
Why did it take a physicist to introduce
a genuinely new system of management
for business? One reason is that Goldratt
recognised the importance of mathematics
and statistics. In probability theory,
the law of large numbers states that ‘as
the number of identically distributed,
randomly generated variables increases,
their sample mean (average) approaches
their theoretical mean’ [SOURCE:
Brittanica.com]. Put more simply, as
illustrated, regardless of how skewed the
distribution of any given activity, throw
enough of them together and the overall
distribution will be the famous bell curve
of the normal distribution.
Goldratt didn’t discover this phenomenon,
known as the Central Limit Theorem.
It was proved in 1713 by Swiss
mathematician Jakob Bernoulli (who also
gave his name to the principle that allows
things that are heavier than air to fly, but
that’s a different story). The late Israeli
physicist’s innovation in productivity
was realising that this law held the key to
resourcing and scheduling.
Resource aggregation: stabilising demand
Figure 3
The Power of Resource
Aggregation
The Power of Resource Aggregation
RESOURCE POOL
Time
Demand
Time Time Time
PROJECT 1 PROJECT 2 PROJECT 3 PROJECT 4
Time
Demand
Average
The Power of Resource Aggregation
RESOURCE POOL
Time
Demand
Time Time Time
PROJECT 1 PROJECT 2 PROJECT 3 PROJECT 4
Time
Demand
Average
The Power of Resource Aggregation
RESOURCE POOL
Time
Demand
Time Time Time
PROJECT 1 PROJECT 2 PROJECT 3 PROJECT 4
Time
Demand
Average
deviation of requirement for the
aggregated resource becomes way less
and therefore much more stable (see inset
above with resource pool). Projects usually
contain enough randomly generated
variables to take advantage of Bernoulli’s
maths. So, you can achieve a much more
stable consistency on the demand side—
and have the people you need at the right
time—by factoring aggregate probability of
demand into your resourcing algorithm.
Let’s say you have four projects (as shown)
with demand (y-axis) for a particular role
varying over time (x-axis). If you increased
headcount to serve the maximum demand,
you’d go broke. But if you don’t account for
those spikes, you’ll fall behind.
One of the hard problems of scheduling,
then, is managing the complexity that
arises from variance in demand. When
you aggregate the resource, the standard
6
_
Critical Chain Project Management_
Many people will tell you they have software that looks at their
resources across the whole business so they can make accurate
resource plans. This is a great—and necessary—first step. But it’s
not sufficient. Like the brittleness of the simple waterfall schedule
mentioned above, resource planning goes fine until you actually
need people. In the dynamic midst of multiple projects, and the
cacophony of the daily bind, how could you possibly know who
should turn up three weeks from now? For which task? Are you
sure you’ll need them on Tuesday and not Wednesday? Will they
even be available? Who’s behind? What’s taking longer than we
thought? What’s the implication? And so on.
the project. Crucially, because the information is logically derived
from objective data, everyone knows it is valid, and can thus trust
the need to subordinate their own actions to what is best for the
whole. Everyone knows where to pay attention—and you can take
timely, unified, course correcting action. This is possible because
of another kind of aggregation—that of the tasks themselves.
In the early stage of project planning, everyone gets a bit nervous.
The project manager asks for estimates: How many people? What
tasks? By when? If the manager holds their feet to the fire for a
completion date that’s certain, we all know what happens. People
give themselves a cushion. It’s human nature.
Critical Chain accepts the inherent uncertainty of estimates.
But instead of letting everyone give themselves their own
contingency to safeguard their individual estimates, the method
rids the individual estimates of their local contingencies. It then
identifies the resource levelled critical path and adds an aggregate
contingency to the end of that path—called a project buffer.
The table shows, in simple form, the maths behind this idea. We’ll
explain more on the buffer, as well as the meaning of the P90 and
P50 estimates below.
Figure 4
The Power of Schedule Aggregation
25
Buffer
10
1
10
2
10
3
10
4
10
5
20
1Tasks
20
2
20
3
20
4
20
5
The Power of Schedule Aggregation
$+
due
date
100 days
Resource Task Vendor Customer
Project
BENEFIT
$+
due
date
100 days
due date
75 days
BENEFIT
Task 1 Task 2 Task 3 Task 4 Task 5
Unbuffered
duration
Buffer Duration
Days
Saved
% of P90
P90 20 20 20 20 20 100 100
P50 10 10 10 10 10 50 25 75 25 25%
P90
P50
Schedule aggregation:
the magic 25%
What you need, in addition to a well-articulated resource plan
is a signalling system. Goldratt applied mathematics, logic and
systems thinking to the problem. His genius was in creating a
system of management that signals what resource, by type and
quantity are required where and when. At any given moment,
everyone can see the task causing penetration of the critical path
and thus know the effect that early or late delivery is having on
_
Critical Chain Project Management_ 7
which is the ‘uplift’ step. Critical Chain is really about being
more mindful with the resources you have.
After production, Goldratt turned his attention to managing
projects. As we’ve seen, projects were ripe for renewal as the
failure rate from waterfall management was simply too high.
TOC encountered even more pushback here because project
management was establishing itself with PRINCE2, PMBOK and
other frameworks. But those who tried the TOC Critical Chain
approach to project management—including BHP, Boeing, Delta
Airlines and all branches of the US Department of Defence—
have taken their performance to new heights.
Critical Chain provides a way to assign the resources available
to you to the most profitable opportunities and ensure your
team can respond to, and focus on, conditions in the real world.
It gives you a way to do three things:
•	 Deliver your promised business case—in full, in less
time, every time
•	 Mindfully manage resources to reduce your ‘cash burn’
•	 Free up your people sooner to tackle the next big idea
Let’s start with what’s radically different about Critical Chain
and some of the core principles behind why it works. We’ll
finish with how this helps you keep control of your portfolio and
succeed despite risk inherent in execution.
The 5-Step FOCUS
Process of continuous improvement
F
OO
CC
U
S
FIND
the constraint
OPTIMISE it
COLLABORATE
around it
UPLIFT
system
performance
START AGAIN
Slide 46Part 5
Figure 5
The 5-Step FOCUS
Focus on your goal
Goldratt’s Theory of Constraints is called a theory but it’s
really built on a profoundly simple and pragmatic premise:
every system has a constraint. What’s preventing a production
line from producing infinite throughput in zero time? Or what’s
stopping a single project from being delivered immediately?
We all intuitively understand the truth behind these questions—
somewhere there’s a performance-limiting bottleneck or a
weakest link in the chain.
The conclusion Goldratt drew from this truism is that a gain at
the constraint means a gain for the system as a whole. Organise
your system around the constraint and you can get remarkable
results. He developed five simple steps to help you do this.
Goldratt’s five focusing steps used words like ‘exploit’ and
‘subordinate’, which can be misinterpreted by the people you’re
asking to do the work. So we tweaked the language to create a
memorable acronym. We think Goldratt would have appreciated
it, since he said TOC is ultimately about ‘focus’.
First, of course, you must find your constraint—the ratedetermining
step of your whole system. (Better yet, design
it into your system.) We don’t need to get into the details of
optimisation and collaboration, except to say that this is where
you get resourceful. By rejigging the system you help the
constrained resource—whether a person or a bit of technology—
go faster or do more. A key realisation is how much you can
increase performance here before you need further investment,
8
_
Critical Chain Project Management_
Critical chain, not path
The definition of project success is the
on-time (or sooner) delivery of what you
promised, using no more resources than
anticipated. The first thing, then, is to work
out all the necessary tasks and decide
in what sequence they need to be done.
Once you’ve optimised for things that can
be done concurrently, you’re left with an
apparently minimum timeline. The longest
chain of sequential tasks defines the length
of the project.
In this case, the constraint is this sequence.
It may include many people. But at any
one point in time, only one task is on the
critical path. So isn’t this what the critical
path method already does? Unfortunately,
classic critical path scheduling deals only
with the issue of task precedence. Unless
you’re operating with infinite resources,
your project duration is almost always
extended by resource contention—
different tasks competing for the same
resources at the same time.
A project is like a relay race, with the baton
passed along the chain of tasks and their
participants. What’s far more important
than the date of the handover is the will of
everyone to complete the overall race in
the shortest time reasonable and possible.
A necessary condition of knowing what is
reasonable and possible demands that you
take resource availability into account, on
top of the already mentioned need to define
task durations and dependencies.
Task duration
Frequency
A
P10
B C
P50 P90
BUFFER =
Σ (P90 - P50)
Summing the Parts to Serve the Whole
10% probability
of completion
50% probability
of completion
90% probability
of completion
Figure 6
Project Buffers
In the sidebar (Figure 7.3), you can see how
the critical path lengthens when resolving
resource contention.
The resource levelled critical path becomes
a critical chain by constructing the logical
dependencies, sizing the tasks according to
their ‘touch time’ effort (that is eliminating
any unnecessary queueing), and applying
the aggregate contingency in the form of
buffers, placed at the end of the chain, to
absorb the inevitable risk and uncertainty
that comes once execution begins.
A profound implication of Goldratt’s
Theory of Constraints is that nonconstrained
resources, by definition, have
spare capacity. From the timeline view,
technically, any task or chain of tasks not
on the critical path has ‘float’—it could
start earlier or later and still deliver on
time to the schedule of the critical path.
To get the most out of your system, what
non-constraints must do, supported by the
relevant signal, is support the constraint
in more quickly completing the relay—not
maximising its own production.
Buffers
Ask someone to set their own deadline for
a task and they’ll pad their estimate. They
know they’ll be asked to take on something
more urgent, yet still be expected to do
the project. Or they’ll try to factor in the
fog of other projects they’re being asked
to work on. Critical Chain takes the idea of
contingency and gets it out in the open.
Estimates start from the assumption that
the person is fully kitted for the task; they
know what’s required and have the people,
materials, information and money to do
the job. Instead of asking them to estimate
a due date with 90% certainty (P90), the
method asks for 50% certainty (P50). In
other words, people are expected to meet
their estimated due dates only half the time.
(Which half doesn’t matter because we’ve
aggregated resources and tasks with the
laws of probability.) This takes the pressure
off and allows for more realistic estimates.
The project manager takes into account the
duration of the chain of P50 tasks and adds
a buffer to the end, usually equal to half the
length of the chain it supports. Crucially, we
expect this buffer to be consumed. We just
don’t know at what point or by whom. Yet.
Assuming the project has more than a
handful of tasks and participants, the law
of averages means that some activities will
go faster than anticipated and some will
take longer. Provided you’ve got a way of
signalling which tasks are starting to drag the
chain, the variation will wash out in the end.
The maths here is really calculus, that
is, analysing the rate of change. Our two
variables are the rate at which the chain is
being consumed (ie, how you’re progressing
on the tasks) and the rate at which the
buffer is being consumed (ie, how that tracks
against expectation). This allow us to create
the ‘critical ratio’ which gives us our signal.
That signalling system is the ‘fever chart’
(see later).
_
Critical Chain Project Management_ 9
The Critical Path: Basic Logic
+
- $ total: 84 days
Task 1
Task 2
Task 3
Task 4
48 days
20 days
24 days
36 days
The Critical Path
flexible
start
workstream 1
workstream 2 Task 5
16 days
Task 1
Task 2
Task 3
Task 4
48 days
20 days
24 days
36 days
flexible
start
workstream 1
workstream 2
A
A
B
C
!?
resource
contention
The Critical Path: Resources Allocated
total: 84 days
+
- $
C
Task 5
16 days
The Critical Path
Task 1
Task 2
Task 3
Task 4
48 days
20 days
24 days
36 days
flexible start
workstream 1
workstream 2
A
A
B
C
The Critical Path: Resources Levelled
total: 100 days
total: 84 days
+
- $
Slack C
Task 5
16 days
The Critical Path
workstream 1
workstream 2
The Critical Chain: Beating the Murphy Factor
Original due date: 100 days
+
- $
The Critical Path
Task 1
Task 2
Task 3
Task 4
Task 5
24 days
10 days
12 days
18 days
8 days
A
A
B
C
C
workstream 1
workstream 2
The Critical Chain: Buffer Applied
total: 100 days
total: 75 days
25% reduction
+
- $
The Critical Path
Task 1
Task 2
Task 3
Task 4
Task 5 project buffer
24 days
10 days
12 days
18 days
8 days 25 days
A
A
B
C
C
feeding buffer
feeding
buffer
Task 3
12 days
C
workstream 1
workstream 2
The Critical Chain: Buffer Applied
total: 100 days
total: 75 days
25% reduction
+
- $
The Critical Path
Task 1
Task 2
Task 4
Task 5 project buffer
24 days
10 days
18 days
8 days 25 days
A
A
B
C
Figure 7.1: Identifying the critical path
Figure 7.2: Resource contention
Figure 7.3: Resource levelling
Figure 7.4: Scheduling at p50 — beating the Murphy factor
Figure 7.5: Introducing the project buffer
Figure 7.6: ...and the feeding buffers
Task duration
Frequency
A
P10
B C
10% probability
of completion
P50 50% probability
of completion P90 90% probability
of completion
Buffer = � (P90 - P50)
Summing the Parts to Serve the Whole
Task duration
Frequency
A
P10
B C
10% probability
of completion
P50 50% probability
of completion P90 90% probability
of completion
Buffer = � (P90 - P50)
Summing the Parts to Serve the Whole
Task duration
Frequency
A
P10
B C
10% probability
of completion
P50 50% probability
of completion P90 90% probability
of completion
Buffer = � (P90 - P50)
Summing the Parts to Serve the Whole
In this simplified plan, the red represents
the project burn and the green the
realisation of the project benefit. There are
two workstreams which run concurrently
until the last stage. Tasks 1, 2 and 5 create
the longest chain; workstream 2, with Tasks
3 and 4, has a little wiggle room. As long as
Task 4 is served up in time for Task 5’s start,
the relay race is won.
When we introduce resources, though,
we find a conflict. Resource A (the person in
blue) is assigned to Task 1 and Task 4, which
overlap in the plan. If you have a large team
with proper taxonomy of resources, now’s
the time to find a replacement for one of the
tasks. But Resource A might be a single lab
or machine or aircraft hangar. So you need
to shift Task 4’s start date.
Now the plan has a clear critical path
with resources available for each task. But
the original hoped-for lead time has gone
from 84 days out to 100 days.
If, however, we schedule with
expectation of each task being accurately
estimated only 50% of the time (P50) we
dramatically cut our lead time. Of course,
we have to factor in the probability of half
the tasks taking longer than this.
THE MAGIC 25%
We intelligently take half of the length of
the critical path and add that to the end of
the project. This is now our critical chain and
the project buffer is what will give us a valid
signal for how the project’s travelling.
We can also see that, thanks to our P50
estimates and the project buffer, the original
due date of 100 days has been brought
forward to 75 days—a 25% performance
benefit, confident of being delivered in no
small measure because of the power of the
central limit theorem articulated above.
For tasks that are not on the critical path,
we can also add in feeder buffers. This
allows us to protect their integration points
with the critical chain and gives us the
flexibility to move resources around to keep
the overall delivery promise on track.
Turning this simplified representation of
the critical chain into an ‘executable work
plan’ would include detailing all task names
with grammatically correct task oriented
sentences, defining resource names with a
consistent taxonomy of roles to allow for
resource demand to be measured against
supply, and ensuring that every task has a
successor. It is recommended to have
checklists for more granular work, as there
is no need to clutter the performance
reporting with data that is primarily of use
to the person accountable for completing
the activity.
The Critical Chain method saves you
time, makes more money and builds a
solid reputation for effectively managing
execution risk: more projects, in less time,
on time, every time. It also provides you with
a way to inspire your people to be the best
they can be, in service of the greater goal.
After all, what you’re asking for is
demonstrably both reasonable and possible.
10
_
Critical Chain Project Management_
Burn
Revenue
Burn
Revenue
Burn
Revenue
$+
-
$+
-
$+
Opportunity
Benefit in the Multiproject Environment
due date
early realisation of benefits
cost saving
opportunity benefit
opportunity benefit
early completion
Slide 21Part 1
Figure 8
Opportunity Benefit
in the Multiproject
Environment
WHAT’S A DAY REALLY WORTH?
Two, your project is of the type that doesn’t
appear to benefit from early completion.
But, if you’re in charge of building, say, a new
Olympic stadium, you can’t miss the opening
ceremony. An early-warning system can
save you having to ramp up resources in the
final phase to guarantee the due date. Also,
when you know the cost of every day of your
project’s ‘burn’, and you have a highly reliable
delivery capability, you can evaluate the cost
savings of just-in-time completion.
Three, you might have a ‘necessary
condition’ project, such as complying with
new regulations. In the airline industry, for
example, a day late could ground your airline.
Few people can accurately tell you what
a day is worth—and how you’re tracking
to deliver to it. But expertise in TOC and
CCPM will provide you with a ready and
reliable method for doing so. It’s a way of
using reason to make it possible to navigate
effectively through the complexity of
your strategic portfolio. It contributes to
maximising the delivery of value by looking
through the front window, relying on lead
indicators rather than depending on the false
signals of the rear-view mirror.
‘Time is money’ is a cliché, but few
executives understand the value of a day.
You can calculate value in projects in three
main ways.
One, your project delivers a financial
benefit after completion. Say a mining
company’s new mine will, on completion,
produce $1m of value per hour. Each day
late represents millions in missing revenue.
Knowing what this number is and sharing
it with your team will change how you all
think about due-date promises and the
trade-offs required to keep them.
Eliminate bad habits
We’ve already talked about how the
use of buffers can dramatically reduce
estimate padding. But there are several
other behaviours that the Critical Chain
method can help eliminate.
Parkinson’s Law states that ‘work
expands to fill the time available for its
completion. When you’re working to P50
estimates, the available window for ‘busy’
work, unnecessary meetings and general
faffing is curtailed.
Gold-plating can charitably be seen as
over-delivering. But often the project
doesn’t warrant it; delivering to spec
would be fine. Again, the Critical Chain
approach doesn’t offer the time to
indulge in gilding the lily.
Student Syndrome is the habit of leaving
work till the last minute. The impression is
of working hard but the bulk gets done just
in time. Bring that time forward and you
benefit from the deadline.
Multi-tasking has been proven many times
to be less productive than focused work.
(clearing WIP)
One other benefit. In a normal project
environment, people who finish tasks early
often won’t let on. Why? Because they
think they’ll be expected to finish as quickly
next time, too. But P50 estimates allow
the team to celebrate an early finish while
not expecting it. Of course, the person may
find that avoiding some of the laws and
syndromes above means they can work
better in flow, which is more productive and
at a personal level, more rewarding.
To change
the way you
work, you have
to change the
way you work”
_
Critical Chain Project Management_ 11
COVARIANCE
One reason projects run late is the
statistical principle of covariance. Let’s
say every 10-day task has a 90% chance
of completing on time. When you multiply
the probabilities together, the actual
chance of the overall project being
delivered on time in 50 days is only 59%.
And yet everyone was so nearly sure that
they would each, on their own, with 90%
confidence, be on time.
Input
Days
for task
Outputs
Process
101010 10
90% 90% 90% 90%
10
90%
Control
Probability of delivery:
0.9 x 0.9 x 0.9 x 0.9 x 0.9 = 0.95
= 0.59 = 59%
_
Projecting proactively_
Too often we think of the word ‘project’ as a noun, a thing,
when it’s really a verb, a process. The real innovation of
Critical Chain for the executive is the ability to view accurate
projections at any point in the project. Every project is a
journey from A (an idea) to B (its delivery). That journey
happens in time. You need a way to interpret cause and effect
in the dynamic swirl of real-time events.
% longest chain complete
%projectbufferconsumed
100
90
80
70
60
50
40
30
20
10
0
0 10 20 30 40 50 60 70 80 90 100
Figure 9
The Critical Chain Fever Chart
Manage by exception
While it’s interesting to understand how the internal combustion
engine works, it’s not a necessary condition of knowing how to
drive a car. For that, you need to understand the instrumentation,
controls and how to respond to your environment. The preceding
sections of this guide have shown you a glimpse under the hood
at the maths and algorithms lurking in the principles. Now it’s
time to get into the driver’s seat.
This is a very simplified ‘fever chart’ (sometimes called a trend
chart) which tells the story of a particular project from its
beginning to its completion. The x-axis shows the percentage of
the critical chain that’s been completed, while the y-axis indicates
the percentage of buffer consumed at any given point in time. The
yellow band is where you want to be—the flow zone—using up the
buffer steadily as the project progresses.
In the green zone, everything is going smoothly. Anything in
the red zone means pay attention. At the point in time on the
project when the chain was only 45% complete, the buffer was
60% consumed, and thus caused the spike into the red on the
fever chart. With the right software (in simple cases without any
specialist software at all), you can establish what’s happening at
that red data point, who’s working on it, figure out what help they
need and deploy the appropriate non-constrained resources to
provide support to get back into the race.
If you dial back a click or two on your zoom, you can see your
whole portfolio in flight. Now each dot in Figure 10 (below)
represents a project. Project 3 is doing fine, while Project 2 is
almost complete but needs a little help to finish on time. Project 1
isn’t even halfway complete, but we can already see it needs some
attention before it veers further off course.
This is what management by exception really means. You now
have finely calibrated instrumentation to show you the reality of
your portfolio and provide you with the means to make intelligent
decisions to deliver the portfolio to promise.
% longest chain complete
%projectbufferconsumed
100
90
80
70
60
50
40
30
20
10
0
0 10 20 30 40 50 60 70 80 90 100
project 3
project 1
project 2
Figure 10
Fever Chart of the Progress of the Project Portfolio
12
_
Critical Chain Project Management_
Every fever chart tells a story
Figure 11
Managing Portfolio Performance
Longest Chain Complete
BufferConsumed
140%
120%
100%
80%
60%
40%
20%
0%
0% 20% 40% 60% 80% 100%
Longest Chain Complete
BufferConsumed
120%
100%
80%
60%
40%
20%
0%
0% 20% 40% 60% 80% 100%
Longest Chain Complete
BufferConsumed
110%
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
0% 20% 40% 60% 80% 100%
Longest Chain Complete
BufferConsumed
110%
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
0% 20% 40% 60% 80% 100%
Longest Chain Complete
BufferConsumed
110%
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
0% 20% 40% 60% 80% 100%
Longest Chain Complete
BufferConsumed
140%
120%
100%
80%
60%
40%
20%
0%
0% 20% 40% 60% 80% 100%
Longest Chain Complete
BufferConsumed
120%
100%
80%
60%
40%
20%
0%
0% 20% 40% 60% 80% 100%
Longest Chain Complete
BufferConsumed
160%
140%
120%
100%
80%
60%
40%
20%
0%
0% 20% 40% 60% 80% 100%
Longest Chain Complete
BufferConsumed
110%
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
0% 20% 40% 60% 80% 100%
1 2 3
4 5 6
7 8 9
But our PMO already
does this, doesn’t it?
It may or may not surprise you and your
fellow executives to discover that all
project management doesn’t happen
like this. Surely, some think, every
method has tools to track progress and
enable action? What about earned value
and critical path management? But
it’s simply not the case. Conventional
waterfall project management planning
is simply too brittle and focuses on what
happened yesterday, not what you have
to do today to outperform. That said,
adopting TOC can happily embrace
other methodologies. It creates a potent
combination with Lean and Six Sigma,
and we have created our own take on
Agile which we call Reliable Agile—but
that’s a subject for another guide.
The previous two charts are simple stylised graphics to
make a point. The nine charts above, however, are taken
from real-world projects after completion. Each one tells a
story. In the [2] top-middle, we can see a very well-executed
plan. In [1] something went haywire around the middle.
Perhaps a part didn’t arrive or weather prevented work.
The team focused there and brought it back on track. On the
bottom row, for [7], [8] and [9], we see the ‘caterpillar effect’
of living in the deep green grass, which is usually due to
proper full-kitting enabling quick completion of early tasks.
And [6] was the problem child that had to be cajoled all the
way. But every single project finished on time.
The beauty of these charts is not only how they help you
manage during the project, but what they can tell you
afterwards—you can have a proper wash-up after the
project completion. Why did things drift into the red at that
point? What might we do better next time? How can we take
advantage of the learning? This is the essence of a proper
process of ongoing improvement.
_
Critical Chain Project Management_ 13
_
What’s the catch?_
With all this maths and talk of aggregation, you might expect
people to feel like cogs in a machine. That’s not our experience
at all. Frontline operators, whether developers or mechanics, are
usually the first to see the benefit. Finally, they have a way to see
what really needs to be worked on—and in what sequence. They
can raise a hand and ask for help when the line in the fever chart
starts to head north. And when it feels like what they’re being
asked to do is neither reasonable nor possible, they can prove to
their bosses why. It leads to a more ‘just’ way of working all around.
If you think all this sounds too good to be true, you’re right. You
can’t simply flip a switch and introduce Theory of Constraints or
CCPM. It requires a fundamental change in mindset. Your people
need to understand that, instead of doing their normal work, they
may need to support a colleague who’s carrying the baton on the
critical chain. For such a transformation, you’ll need to consider
both organisational design and organisational learning. But when
they see how their role directly influences business outcomes,
people are more engaged. And when you shift from a reductionist
worldview of local optimisation to a holistic systems approach, the
results are remarkable.
At the outset, this guide compared the executive’s plight to that
of the fictional PM, Jim Hacker. We promised to help you have
the requisite conversation with your PMO about better ways
to manage your initiatives so that you can be in control of the
execution risk of delivering on your strategy and accomplishing
your organisational goals.
The goal of effective portfolio management is to see the whole,
know the constraint and focus your effort where it counts. On his
deathbed, Eli Goldratt was asked what was the ultimate constraint
in business. His answer was simple: ‘Management attention’.
We hope you feel yours has been repaid.
Further reading
The Goal, Eli Goldratt
Critical Chain, Eli Goldratt
More Than Just Work, David Hodes
The Ensemble Consulting Group YouTube
channel includes videos that explain the
‘science of work’ in more detail.
14
_
Critical Chain Project Management_
_
Our Results: Percentage and Dollar Values_
CAPACITY
REALISED
$45m
ADDITIONAL PROFIT
IDENTIFIED
Manufacturing
25%
despite independent auditors giving
a 3% chance of on-time completion
REDUCTION IN
LEAD TIMES
IN COST
SAVINGS$180m
35%Business
Management
Systems
$45m
IN ADDITIONAL
INCOME
Generated capacity for
REDUCTION IN
CREDIT APPROVAL
LEAD TIMES
35%
IMPROVEMENT
IN PRODUCTIVITY
50%
Construction Liquidated
damages in excess of
$100m
AVOIDED
New Product
Development
$20m IN ADDITIONAL
REALISED PROFIT
INCREASE IN PRODUCT
LAUNCHES from 100 to 12525%
Engineering
$100m
IN ADDITIONAL
PROFIT
IMPROVEMENT
IN PRODUCTIVITY
33%
Banking
_
Critical Chain Project Management_ 15
_
We’ve been giving tier-one
Australasian companies this
leverage for over 15 years.
_
Our innovations in productivity give ambitious
executives three ‘value levers’
1. Deliveryour promised business case
2. Mindfully manage resources to reduce your ‘cash burn’;
— in full, on time, every time;
3. Free up your people sooner to tackle the next big idea.
Our innovations in productivity give
ambitious executives three ‘value levers’
—
Let’s
Redefine
What’s
Possible
—
—
Level 1501,
Westfield Tower 2,
101 Grafton Street,
Bondi Junction,
NSW, Australia 2022
—
t +61 2 9387 3955
focus@ensembleconsultinggroup.com
EnsembleConsultingGroup.com
© Ensemble Consulting Group
—
Are you an ambitious
executive looking
to apply the Theory
of Constraints and
other innovations in
productivity to your
organisation? 
_
Get in touch_
_
VP Projects
_
Even if I was allowed to mention
dollar value numbers, this is still the
first thing I would always write:
These outcomes are so valuable they
literally print their own money.”