IT Demand Management and Capacity Planning: Why Estimation Is Vital to Balancing the Scale

The Intelligence behind
Successful Software Projects
IT DEMAND MANAGEMENT
AND CAPACITY PLANNING:
WHY ESTIMATION IS VITAL TO
BALANCING THE SCALE
-1-

Quantitative Software Management
Executive
Summary
-2-
Agenda
Challenges associated with demand
management & capacity planning:
• “Realistic” demand estimation
• Effective resource optimization
• Detailed resource planning to support
capacity utilization

Executive
Summary
Why Is IT Development Capacity
Planning so Difficult?
-3-

Executive
Summary
-4-
A Difficult Juggling Act in a Complex
Environment
Production CapacityBusiness Demand Technology & Business
Executive Management

Executive
Summary
-5-
Computerworld – How to Develop an
Effective Capacity Planning Process
Requirements/Estimates
Productivity baseline
Estimates Transformed into
Resource Plans
Aggregate demand compared
to actual capacity
Recommended Best Practice
Assessing
Resource Optimization
“How to develop an effective capacity
planning process”, Rich Schiesser,
Computerworld, Mar 31, 2010
Configure skills/roles
Top-Down Estimation

Executive
Summary
-6-
Why is Matching Demand & Capacity
so Difficult?
• Business stakeholder insatiable
appetite for competitive capability
• Poor IT estimation & poor project
stakeholder negotiation
• Ability to predict amount of resources
/skills as required over the course of a
project
• Dynamic nature of the total volume of
projects in the development pipeline

Executive
Summary
“Realistic” Demand Estimation
-7-

Executive
Summary
-8-
“Realistic” Demand Estimation
• What we would like vs what is possible
• What facts can we bring to bear?
 Scope of the work (size)
 Productivity to perform the work
 Availability skilled labor
• How important is accuracy?
• How do we negotiate some realistic demand solutions?

Executive
Summary
9
Terminology
Targets, constraints, estimates, commitments, and plans are not the same
thing:
• Target - A goal, what we would like to do or achieve
• Constraint - Some internal or external limitation on what we are able to do
• Estimate - A technical calculation of what we might be able to do at some
level of scope, cost, schedule, staff, and probability
• Commitment - A business decision made to select one estimate scenario and
assign company resources to meet a target within some constraints
• Plan - A set of project tasks and activities that (we calculate) will give us
some probability of meeting a commitment at a defined level of scope,
budget, schedule, and staff
Organizations sometimes confuse these terms and the business practices they
represent.

Executive
Summary
-10-
Identifying Unrealistic Stakeholder
Expectations
• QSM research has found that the 2 most
common reasons projects fail is:
 Unrealistic cost & schedule expectations
 Unmanaged requirements growth
• Need an effective mechanism to quantify
stakeholder requirements - scope
• Need and effective method to translate
requirements into time and effort
• Need to provide practical alternatives when
expectations don’t meet reality
If we can’t get this
right we will never
have a effective
capacity planning
solution!

Executive
Summary
-11-
How? Top-Down Scope-Based
Estimation
Particularly good at identifying unrealistic expectations:
• Doesn’t require a lot of detailed information
• Relatively quick
• Few hidden assumptions
• Explicitly calibrated from history (local or industry)
• Very flexible for scope, staffing, duration, etc., changes
• Considered industry “best practice”

Executive
Summary
orem ipsum dolor sit
amet, consectetur
adipiscing elit, sed do
eiusmod tempor
incididunt ut labore et
dolore magna aliqua.
Ut enim ad minim
veniam, quis nostrud
exercitation ullamco
laboris nisi ut aliquip ex
ea commodo
consequat.
orem ipsum dolor sit
amet, consectetur
adipiscing elit, sed do
eiusmod tempor
incididunt ut labore et
dolore magna aliqua.
Ut enim ad minim
veniam, quis nostrud
exercitation ullamco
laboris nisi ut aliquip ex
ea commodo
consequat.
12
Two Sizing Approaches
Analogy Sizing
Comparing this system to the known sizes of
similar system(s).
Adjustments can be made by percentage or
by including/excluding functions.
Artifact Sizing
Counting and measuring system artifacts and
work products and scaling to the size of the
final system.
Different artifacts have different knowledge
“densities” adjusted by their gearing factors.
Copyright © 2014 QSM Inc
Historical
System of
known size
System
being
estimated
% difference. Functions
included or excluded
System
being
estimated
Table A
Screen1
Screen2
Model X
Gearing
Factors
Requirements,
Use Cases
Screens
Tables,
DatabasesModels,
workflows
System Work Products
and Artifacts

Executive
Summary
-13-
Gearing Factor
 As long as we work in a single unit
for history, calibration, and
estimation, we do not need a
Gearing Factor
 If we want to compare or sum sizing
in different units or to use industry
databases, we have to normalize to
a common unit
 The Gearing Factor is the
normalizing base
 We can calculate the Gearing Factor
from history

Copyright © 2014 by QSM, Inc.
Successful Software
Projects
14 ®
Gearing Factor
 As long as we work in a single unit for history,
calibration, and estimation, we do not need a
Gearing Factor
 If we want to compare or sum sizing in different
units or to use industry databases, we have to
normalize to a common unit
 The Gearing Factor is the normalizing base
 We can calculate the Gearing Factor from history

Executive
Summary
-15-
Top-Down Estimation Particularly
Effective Early in the Project Lifecycle
EstimationUncertainty
Concept Reqts Design Construction Dev Testing Qualification Testing
Functional Measures:
Business Requirements
Function Points
Agile Epics/Stories
Use Cases
Component Measures:
Modules
Screens/report/forms/ETLs
RICE Objects
Package Business Process Configurations
Story Points
7X 5X 3X .75X .25X .05X

Executive
Summary
-16-
Functional Measures:
Business Requirements
Function Points
Agile Epics/Stories
Use Cases
Component Measures:
Modules
Screens/report/forms/ETLs
RICE Objects
Package Business Process Configurations
Story Points
7X 5X 3X .75X .25X .05X

Executive
Summary
-17-
7X 5X 3X .75X .25X .05X
Historic sizing &
performance data
is the key to
reducing
uncertainty early in
the lifecycle!

Executive
Summary
-18-
Estimates are Uncertain By Their
Nature
 Estimates are always uncertain
 The job of the estimation process is not to
remove the uncertainty, it is to measure it
 SLIM-Estimate simply translates
uncertainty into risk
 We accept and deal with uncertainty in
many aspects of our lives
“Forecasts of the future are
inherently uncertain.”
Larry Putnam Sr.
Measures for Excellence
Yourdon Press 1992 p.207
“The key issue… is documenting
the estimate’s uncertainty…”
Steve McConnell.
Software Estimation
Microsoft Press 2006 p.251
“Prediction is difficult, especially
if it involves the future.”
Niels Bohr.
Physicist and
Nobel Laureate

Executive
Summary
-19-
Accuracy and Early Feasibility
Estimates
• Early estimates have more uncertainty.
 Driven by what we know and what we don’t know.
– It’s good to know what you don’t know
• Good place to focus our attention to reduce uncertainty
• Some contend that without high accuracy estimates are of no use
and provide little value.
 Estimates don’t need to be 10 decimal places accurate to identify cost
and schedule proposals that are patently unreasonable.
 “It is better to be roughly right than precisely wrong.” - John Maynard
Keynes
• At this stage of the project lifecycle estimates should be judged on
their ability to help make better business decisions.
The Accurate Useful Estimate

Executive
Summary
-20-
Trapping High Risk Proposals
• Most projects start off as proposals that need to be prioritized and
justified.
 Priorities are usually aligned with the organizations strategic goals.
 Justification may be based on benefits to the organization.
– New revenue to be realized
– Savings or cost efficiencies
 Many times the benefit is expressed in terms of ROI , IRR or payback
period.
• Most proposals have some high level description of capabilities that
relate to the size & scope to be developed and implemented.
• Most proposals have a target or desired schedule and cost.
• This is enough information to generate a useful feasibility estimate.

Executive
Summary
-21-
Typical Project Proposal
Initially these calculations are
based on what the business
would like to happen. We need
to re-evaluate them based on
the feasibility estimate of what
is likely to happen.

Executive
Summary
-22-
Identify High Risk Projects as They Enter
the Budgeting & Approval Process

Executive
Summary
-17-
Compare Proposed Expectations to
Historical Performance
Desired
Cost &
Schedule
Historical
data and
benchmark
trends
Impossible Zone
Impossible Zone

Executive
Summary
-17-
Compare Proposed Expectations to
Historical Performance
Desired
Cost &
Schedule
Even though
there is
variability in
the estimate it
is useful in
identifying
when desired
costs and
schedules are
not reasonable
for a given
amount of
scope
Impossible Zone
Impossible Zone

Executive
Summary
-25-
Evaluating Alternative Plans
To get alignment between what we would like and what is likely?
• There are a finite number of options to explore:
1. Reduce scope to meet schedule and cost goals
2. Increase staffing to meet schedule
– Decreases probability of meeting cost and lowers reliability
3. Negotiate for more schedule and budget
– Historic data can help backup your case
4. Increase productivity
– Usually not able to influence much in the short term
– Should be backed up by data
5. Combination of the above

Executive
Summary
Effective Resource Optimization
-26-

Executive
Summary
-27-
Effective Resource Optimization
• How do we make sure we are using our scarce resources
so as to optimize the amount of work that can get done?
• What facts can we bring to bear?
 Scope of the work (size)
 Historic staffing data (internal & industry)
• Trading-off cost & schedule
 Recognize what we can influence though staffing/resources
 Unanticipated benefits of effective trade-offs
– Lower cost
– Higher throughput
– Higher reliability

Executive
Summary
-28-
Using Benchmarking to Assess
Resource Utilization – Internal
Trends
Internal Baseline
Schedule Performance
1 10 100
Effective IU (thousands)
0.1
1
10
100
Duration(Months)
Effort Performance
1 10 100
0.01
0.1
1
10
100
1,000
Effort(PHR)(thousands)
Productivity Performance
1 10 100
0
5
10
15
20
25
30
35
Productivity(Index)
Staffing Performance
1 10 100
1
10
100
1,000
PeakStaffing
CompanySample Projects Avg. Line Style 1 Sigma Line Style 2 Sigma Line Style 3 Sigma Line Style
With a modest amount of historic
data, we can understand key
performance trends and variability.

Executive
Summary
-29-
Using Benchmarking to Assess
Resource Utilization – External
Comparison
External Benchmark
Schedule Performance
1 10 100
0.1
1
10
100
C&TDuration(Months)
Effort Performance
1 10 100
0.01
0.1
1
10
100
1,000
C&TEffort(PHR)(thousands)
Productivity Performance
1 10 100
0
5
10
15
20
25
30
35
PI
Staffing Performance
1 10 100
0.1
1
10
100
1,000
C&TPeakStaff(People)
CompanySample Projects QSM Business Avg. Line Style 1 Sigma Line Style 2 Sigma Line Style 3 Sigma Line Style
External comparison to industry
benchmarks can highlight
opportunities for improvements
and optimization.
Close to average
Close to average Higher than average
Higher than average

Executive
Summary
Opportunity to Optimize Staffing
Levels
Staffing Comparison
Peak Staffing
1 10 100
1
10
100
PeakStaff(People)
Comparison of Company Sample Projects to QSM Business
C&T Peak Staff (People) vs. Effective IU
C&T Peak Staff (People) Values
Benchmark Reference Group:
QSM Business
Comparison Data Set:
Company Sample Projects
Difference From Benchmark
at Min
Effective IU:
1200
4.33
5.92
1.59
at 25% Quartile
Effective IU:
4920
6.62
12.22
5.60
at Median
Effective IU:
8840
7.90
16.51
8.61
at 75% Quartile
Effective IU:
11100
8.46
18.55
10.10
at Max
Effective IU:
33420
11.78
32.68
20.90
Comparison breakpoints based on min, max, median and quartile values for the data set: Company Sample Projects
Company Sample Projects QSM Business Avg. Line Style
How industry staffs
How this organization staffs
Opportunity to take
on more work

Executive
Summary
Opportunity to Lower Cost
Effort Comparison
Effort (PHR)
1 10 100
0.1
1
10
100
C&TEffort(PHR)(thousands)
C&T Effort (PHR) vs. Effective IU
C&T Effort (PHR) Values
QSM Business
at Min
Effective IU:
1200
599.37
2301.02
1701.65
at 25% Quartile
Effective IU:
4920
1591.89
5917.57
4325.68
at Median
Effective IU:
8840
2388.30
8760.07
6371.77
at 75% Quartile
Effective IU:
11100
2795.98
10202.24
7406.26
at Max
Effective IU:
33420
5996.75
21337.60
15340.85
Industry effort cost
This organization’s effort cost
Potential savings for
redeployment on additional
demand or if we outsource
to an external vendor this
represents potential savings
to the organization

Executive
Summary
-32-
For Almost No Schedule Penalty
Schedule Comparison
Schedule
1 10 100
1
10
C&TDuration(Months)
C&T Duration (Months) vs. Effective IU
C&T Duration (Months) Values
QSM Business
at Min
Effective IU:
1200
2.71
2.61
-0.10
at 25% Quartile
Effective IU:
4920
3.83
3.62
-0.21
at Median
Effective IU:
8840
4.42
4.15
-0.27
at 75% Quartile
Effective IU:
11100
4.68
4.38
-0.30
at Max
Effective IU:
33420
6.13
5.66
-0.47
Industry durations
This organization’s durations
Very little difference!

Executive
Summary
-33-
Resource Optimization Takeaways
• Use historical data to assess current resource utilization.
• Compare current resource utilization to outside industry benchmarks.
• Identify opportunities for optimization (don’t over staff!).
 Make the time-effort tradeoff relationship work for us.
– You pay almost no schedule penalty.
– Cost goes down and reliability goes up.
 For more information on this - http://www.qsm.com/research
 Document the results and benefits of making better business decisions.

Executive
Summary
Translating Estimates into
Resource Demands
-34-

Executive
Summary
Common Practice vs Best Practice
Labor Hour Estimates
Most Common Industry Practice Today Best Practice Estimates Breakdown
Effort by Skill by Month
This approach doesn’t
help the organization
determine when these
resources are need as
the project progresses.
This approach identifies what skills are needed when and can
easily feed a PPM system where specific people can be allocated
to the project.

Executive
Summary
-36-
How Skills Flow on-off a Project
Release
• Need to understand for
any given development
methodology how the
skilled manpower builds up
and rolls off a project.
 Agile
 Waterfall
 Package Implementation
 Etc.
• Need to be able to
estimate the skill demands
and pass to PPM system.

Executive
Summary
37
Product Development Lifecycles & Skills
• All software development lifecycles include four primary activities
 What, How, Do & Deploy/Fix
 Some SDLCs are more sequential and others include more concurrency in
theses activities
• Types of labor needed changes as we transition across activities
Methodology What How Do Deploy/Fix
Waterfall Concept Rqmts. &
Design
Construct &
Test
Deploy
RUP Initiation Elaboration Construction Transition
Agile Initiation Iteration
Planning
Iteration
Development
Production
SAP ASAP Project
Preparation
Business
Blueprint
Realization &
Final Prep
Go Live
Knowledge Acquisition Implementation Usage
Types of labor needed

Executive
Summary
-38-
Top Down Estimate Skills/Role
Configuration
• You specify the
skills/roles defined in
your organization and
the rates charged for
those labor categories.
• Then you allocate the
skills across the lifecycle
appropriate to your
organization and
development
methodology(s).
These allocations can be
determined by mining
data from corporate
PPM/resource planning
tools.
Import
from
PPM

Executive
Summary
-39-
The QSM PPM Integration Framework Works with
Any Enterprise PPM/Resource Planning Solution
QSM Webinar: From Proposal to Project:
Getting Resource Demand Early

Executive
Summary
-40-
The QSM PPM Integration Framework Works with
Any Enterprise PPM/Resource Planning Solution

Executive
Summary
Determining Aggregate Demand
and Matching Demand to Capacity
-41-

Executive
Summary
Schedule Staffing Effort & Cost
Monthly Avg Staff (L0)
< Ireland, India & US >
3 6 9 12 15 18 21 24 27
Oct
'13
Jan
'14
Apr Jul Oct Jan
'15
Apr Jul Oct Jan
'16
0
50
100
150
200
250
300
350
people
3 6 9 12 15 18 21 24 27
Oct
'13
Jan
'14
Apr Jul Oct Jan
'15
Apr Jul Oct Jan
'16
0
25
50
75
100
125
150
175
people
3 6 9 12 15 18 21 24 27
Oct
'13
Jan
'14
Apr Jul Oct Jan
'15
Apr Jul Oct Jan
'16
0
20
40
60
80
people
Staffing - Demand verses Capacity
IT Demand
Staffing plans
for individual
projects across
3 development
centers (35
projects
approved and
in the pipeline)
Demand at
3 Dev
centers
• India
• US
• Ireland
Aggregate
IT
resources
required =
305 FTE
staff
Capacity Limit
250 People
When the demand exceeds capacity
1. Eliminate projects
2. Slip start dates
3. Selective headcount reductions
Demand exceeds
Capacity for 9 months

Executive
Summary
-43-
Adjust Start Dates on the Future
Projects in Order to Match Demand to
Capacity
Staffing & Schedule
Monthly Gantt Chart (L3)
< Start dates adjusted to not exceed 250 Max capacity>
3 6 9 12 15 18 21 24 27
Oct
'13
Jan
'14
Apr Jul Oct Jan
'15
Apr Jul Oct Jan
'16
IRELAND DEVELOPMENT...
Ireland Project 001
Ireland Project 002
Ireland Project 003
Ireland Project 004
Ireland Project 005
Ireland Project 006
Ireland Project 007
Ireland Project 008
Ireland Project 009
Ireland Project 010
Ireland Project 011
Ireland Project 012
Ireland Project 013
Ireland Project 014
Ireland Project 015
INDIADEVELOPMENT CEN...
India DC Project 001
India DC Project 002
< Start dates adjusted to not exceed 250 Max capacity>
3 6 9 12 15 18 21 24 27
Oct
'13
Jan
'14
Apr Jul Oct Jan
'15
Apr Jul Oct Jan
'16
0
20
40
60
80
people
Projects start dates are moved out in time in
order to drop the total aggregate staffing
under 250 which is the maximum capacity.

Executive
Summary
-44-
Delaying Start Dates on 8 Projects
Matches the Demand to the Capacity
Schedule Staffing Effort & Cost
< Start dates adjusted to not exceed 250 Max capacity >
3 6 9 12 15 18 21 24 27
Oct
'13
Jan
'14
Apr Jul Oct Jan
'15
Apr Jul Oct Jan
'16
0
50
100
150
200
250
300
people
3 6 9 12 15 18 21 24 27
Oct
'13
Jan
'14
Apr Jul Oct Jan
'15
Apr Jul Oct Jan
'16
0
20
40
60
80
100
120
140
people
3 6 9 12 15 18 21 24 27
Oct
'13
Jan
'14
Apr Jul Oct Jan
'15
Apr Jul Oct Jan
'16
0
20
40
60
80
people
8 projects needed
to be delayed to
make demand
match capacity.
The start date
delays averaged 5
months but
ranged from 2 to 8
months in
duration.

Executive
Summary
What Is the Demand for Skills
Across the portfolio?
-45-

Executive
Summary
-46-
What Are the Skill Requirements
across the Portfolio?
0.00
50.00
100.00
150.00
200.00
250.00
300.00
Oct2013
Nov2013
Dec2013
Jan2014
Feb2014
Mar2014
Apr2014
May2014
Jun2014
Jul2014
Aug2014
Sep2014
Oct2014
Nov2014
Dec2014
Jan2015
Feb2015
Mar2015
Apr2015
May2015
Jun2015
Jul2015
Aug2015
Sep2015
Oct2015
Nov2015
Dec2015
Jan2016
Feb2016
Mar2016
Architect
Database Administrator
Quality Assurance
Development
Data Architect
Business Analyst
Project Manager/Lead

Executive
Summary
-47-
Summary
• Capacity Planning and Demand
Management go hand in hand. It requires:
 Sound estimation capability early in the
lifecycle
 Effective stakeholder negotiations of
time/effort/capability
 Ability to forecast effort by skill level by
month and feed a PPM system
 Ability to perform what-if analysis on the
portfolio quickly when IT Demand exceed
Capacity

IT Demand Management and Capacity Planning: Why Estimation Is Vital to Balancing the Scale

More Related Content

What's hot (20)

Similar to IT Demand Management and Capacity Planning: Why Estimation Is Vital to Balancing the Scale (20)

Recently uploaded (20)

IT Demand Management and Capacity Planning: Why Estimation Is Vital to Balancing the Scale

Editor's Notes