Post on 14-Apr-2016
description
1
Agile Product Management
Mike CohnSeptember 29, 2005
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
Mike Cohn—background
Programming for 20 years
Author ofUser Stories Applied
Agile Estimating and Planning
Java, C++, database programmingbooks
Founding member and director ofthe Agile Alliance and the ScrumAlliance
Founder of Mountain GoatSoftware
Process and project managementconsulting and training
2
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
The planning onion
Agile ProductManagement is
mostly concernedwith the space fromProduct to Iteration
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
Today’s agenda
Prioritizing
Project chartering
Working with the developers
Establish a financial model
3
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
Prioritizing
Why prioritize?
Percent of features used in Word and PowerPoint:
Users workingalone
Used 12%–16% offeatures
Used 26%–29% offeatures
10-persongroups
Source: International Software EngineeringResearch Network Workshop (2000).
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
Factors in prioritization
Business value
Risk reduction
Change in relative cost
Learning / uncertainty
Business valuetrumps other
factors…
…but there are otherfactors to consider
4
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
Learning / uncertainty
Product knowledgeWhat should we develop?
Project knowledgeHow should we develop it?
The purpose of a projectis to:
Deliver new capabilities
Generate new knowledge
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
One view of reducing uncertainty
Low
High
LowHigh
Means Uncertainty(How)
End U
nce
rtain
ty(W
hat)
A waterfall approach
5
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
An agile approach to uncertainty
Low
High
LowHigh
Means Uncertainty(How)
End U
nce
rtain
ty(W
hat)
An agile approach
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
Sliders
Think of
risk reduction
knowledge generation
change in relative cost
as sliders that move business priority forwardor back
6
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
Example: User interface design
Will this cost more orless letter?
Do sooner Do laterPrioritization basedsolely on business
value
Will designing the UIreduce risk?
Will doing thisgenerate useful
knowledge?
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
Example: a security framework
Do sooner Do later
Will this cost more orless letter?
Prioritization basedsolely on business
value
Will doing thisgenerate useful
knowledge?
Will doing this reducerisk?
7
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
Prioritizing with Kano analysis
Three types of features
Threshold /Baseline
Must be present in orderfor users to be satisfied
Linear The more of it, the better
Exciters /Delighters
Features a user doesn’tknow she wants, until
she sees it
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
Surveying users
To assess whether a feature is baseline,linear, or exciting we can:
Sometimes guess
Or survey a small set of users (20-30)
We ask two questionsA functional question
How do you feel if a feature is present?
And a dysfunctional questionHow do you feel if that feature is absent?
8
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
Functional and dysfunctional forms
I expect it to be that way.
I like it that way.
I am neutral.
I dislike it that way.
I can live with it that way.
If your hotel roomincludes a freebottle of water,
how do you feel?
I expect it to be that way.
I like it that way.
I am neutral.
I dislike it that way.
I can live with it that way.
If your hotel roomdoes not include afree bottle of water,how do you feel?
Functionalform of
question
Dysfunctionalform of
question
X
X
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
Categorizing from a question pair
E
I
I
I
R
E
I
I
I
R
E
I
I
I
R
Q
R
R
R
R
L
M
M
M
Q
Liv
e W
ith
Dis
like
Lik
e
Expect
Neutr
al
Live With
Dislike
Like
Expect
Neutral
Functio
nal Q
uestio
n
DysfunctionalQuestion
MandatoryM
LinearL
ExciterE
QuestionableQ
ReverseR
IndifferentI
9
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
Aggregating results
Apply formatting themes
Automate report execution
Export reports to PowerPoint
Theme Exciter
6
8
42
Lin
ear
21
43
18
Mandato
ry
62
39
27
Indiffe
rent
2
8
9
Revers
e
5
2
2
Questionable
4
0
2
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
What to include
All of the baseline features
By definition, these must be present
Some amount of linear features
But leaving room for at least some amountof exciters
10
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
Today’s agenda
Prioritizing
Project chartering
Working with the developers
Establish a financial model
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
Establishing a shared vision
Teams do best when they have a “clear,elevating goal” and “unified commitment”†
It’s the product manager’s job to focus the teamand find this clear, elevating goal
Tools for establishing a shared vision:
Source: †Teamwork by Carl Larson and Frank LaFasto
Elevator statement‡
Press release
Vision box‡
Magazine review
11
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
Elevator statement
A short statement of the product’s positioning
Explains the product to someone in 2 minutes
Follows this syntax
For (customer)
Who (statement of need or opportunity)
The (product name) is a (product category)
That (key benefit, compelling reason to buy).
Unlike (primary competitor)
Our product (statement of primary differentiation).
Sources: Crossing the Chasm by Geoffrey Moore
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
Example elevator statement
For dentists and their assistants
who need to efficiently schedule appointments
Dental Clinic 2.0 is desktop and web-basedappointment scheduling software
that supports office and remote access.
Unlike competitive products,
Dental Clinic 2.0 is easy to use andaggressively priced.
12
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
Product vision box
Design a box for the software
Even if the software won’t ship ina box
Write 3-4 key bullet points to sellthe software
Easier to come up with 15
The challenge is distilling the listto 3-4 key points
Sources: Agile Project Management by Jim Highsmith.
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
Sample product vision box
Easy integration with mostinsurance systems
Support for multiple chairswith appointment setting bychair
Multi-language support
One click database backup
13
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
Press release to come
Collaboratively write the press releaseyou’d like to see released at the end
What are the key points you’d make about theproduct
What quotes would you have and who wouldthey be from?
CEO? Team members? Customers?
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
Press release templateThe _______company announced today the successful completion of the_________project. This project provides __________________________________________________. The customer for this project,________________, indicated in a recent interview that they selected___________ as their supplier due to the following key benefits:
1. ______________________________________2. ______________________________________3. ______________________________________
________________ also identified several features that they felt wereparticularly useful. These include:
1. ______________________________________2. ______________________________________3. ______________________________________
____________ noted that the single most important benefit of theirsuccessful project was “_______________________________________________________________________.”
14
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
Magazine review
Write the review you’d like to see in PCMagazine
Use whatever magazine you’d like or evenmake one up
What key features would be singled out?
What will be said about your productcompared to the prior version?Competitors?
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
Today’s agenda
Prioritizing
Project chartering
Working with the developers
Establish a financial model
15
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
Communicate the driving factor
Be clear about what’s driving the projectUsually date or features
But NOT both
Feature-driven projectsProject will ship when the desired featureshave been developed
Date-driven projectsProject will ship on the date, regardless ofwhether all desired features are included
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
Example date-driven projects
Tax softwareSoftware to comply with periodic changesin government regulationsSoftware for fantasy footballThe Olympics
16
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
Value-driven projects
Making appropriatetradeoff decisionsbetween date andfeatures
Committing tomaximizing value
Deferring tradeoffdecisions
Acknowledginguncertainty about theproduct and theproject
1.6x
1.25x
1.15x
1.10x
x
0.9x
0.85x
0.8x
0.6x
Project Schedule
InitialProductDef ini tion
Approved Product
Definit ion
Requirements Specification
Product Design
Speci fica tion
Deta iled Design
Specif ication
AcceptedSoftware
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
Tradeoff matrix
Scope
Schedule
Resources
Low Defects
Fixed Firm Flexible Target
300+story points
4-5 months
$400k
1 high bugper month
Tradeoff Matrix
17
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
Set high expectations
You do not control HOW the developmentteam does its work
But you do get to say WHAT they produceand WHEN it’s produced
I don’t tell a mechanic how to fix my car, but Ido ask for an estimate, say what type of partsto use, approve work, and may call for aprogress report if it’s a big job
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
Some expectations you should have
“At least once a month, Iwant my hands on the
keyboard to try out newfeatures.”
Ensures a focus on user-visible features, not justframeworks andarchitecture.Avoids a demo of a snake.
“I want to be able to seeprogress at least once a
month. And I’m only goingto measure progress
through working, testedfeatures.”
Encourages iterations.Stresses importance ofconsistent focus on quality.Provides insight intoprogress.
18
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
Some more expectations
“My computer runsautomated tests when Iturn it on. I expect this
application to have similarautomated tests.”
Ensures team willautomate tests.
Set realistic deadlines.
Too aggressive or too laxare not motivating.Teams work fastest withrealistic deadlines.Be careful of making it allabout the date.
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
Today’s agenda
Prioritizing
Project chartering
Working with the developers
Establish a financial model
19
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
Understand product financials
Business model
Cur
rent
cus
tom
ers
Futur
e cu
stom
ers
Mar
ket a
ccep
tanc
e
Com
petit
ion
Period 1 Period 2 Period 3
Communicate this model to the whole team!
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
Typical sources of return
New revenue
Incrementalrevenue
Retainedrevenue
Operationalefficiency
20
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
An example
36,500$15,000$4,000$10,000$7,50008
36,500$15,000$4,000$10,000$7,50007
29,000$7,500$4,000$10,000$7,50006
29,000$7,500$4,000$10,000$7,50005
20,750$7,500$2,000$7,500$3,75004
$3,750
$2,500
0
New
Revenue
$5,000
$1,600
0
Incr.
Revenue
$2,000
$2,000
$2,000
Retained
Revenue
$7,500
0
0
Oper.
Efficiencies
18,25003
–23,900–30,0002
–88,000–90,0001
Net Cash
Flow
Dev.
Cost
Q
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
Useful financial measures
Net present value
The present value of the money the project willearn
Return on investment / internal rate of return
The rate of return earned by the project per aperiod of time
Discounted payback period
The amount of time until a project pays back itsinitial investment
21
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
Project (theme) comparison matrix
Slides copyright 2000-2004, Michael W. CohnAll slides copyright 2003-2005, Mountain Goat Software
Themes and stories on theplanning onion
Compare financials and desirability (Kanoanalysis) Compare projects to
select a portfolio
Compare themes to layout a multi-release
product plan
Compare themes andstories to create a
release plan