The Fastest Learner Wins: Continuous Learning Through Rapid Release Cycles

lsoftware development
e a n
The Fastest Learner Wins
Living with Black Swans

mary@poppendieck.com Mary Poppendieck www.poppendieck.com

Agenda

1. Compelling Offer
Observing
2. Immediate Connection
Questioning
3. Adoption Chain
Networking
4. Validated Assumptions
2
Experimenting
August 12 Copyright©2012 Poppendieck.LLC l e a n

Todd Park
2010 Harvard Graduate (Economics)
Booz Allen Hamilton Consultant (Managed Care)
Athenahealth (1997)
 Maternity Clinic
 Health Care Records
 Very successful IPO

Retirement (2007)
US Health and Human
Services CTO (2009)
2011

Artist Regina Holliday

3 August 12 Copyright©2012 Poppendieck.LLC l e a n
US Federal Government CTO (2012)

Agenda

1. Compelling Offer
Observing
Questioning
3. Adoption Chain
Networking
4
Experimenting

Strategic Inflection Point

Business goes on
to new heights

10x change in some
element of the business.
What worked before doesn’t work now.
The executives are the last to know.

5 August 12 Copyright©2010 Poppendieck.LLC l e a n Business
declines

Compelling Offer
10X Improvement in some aspect of the offer
Salesforce.com
 SaaS gave a 10X reduction in installation and operating costs
Skype
 Peer-to-peer IP telephony for a 10X reduction in long distance charges
Wikipedia
 Open source collaboration for a 10X increase in the speed and a 100X
reduction in cost for encyclopedia development and maintenance
Dropbox
 10X improvement in ease of use and safety for those who wanted to keep
multiple devices in sync.
YouTube
 Cell phones gave a 10X improvement in capability to take casual videos,

YouTube gave a 10X improvement in ease of sharing those videos.
See also: http://ecorner.stanford.edu/authorMaterialInfo.html?mid=2732

Working Backward
PRESS RELEASE
Heading –
Working Backward Name the product in a way the reader (i.e. your target
customers) will understand.
Sub-Heading –
1. Write a Press Release Describe who the market for the product is and what
benefit they get. One sentence only underneath the title.
Summary –
Give a summary of the product and the benefit. Assume
the reader will not read anything else so make this good.
Problem -
Describe the problem your product solves.
Solution –
2. Write a list of FAQ’s Describe how your product elegantly solves the problem.
Quote from You –
(and answers) A quote from a spokesperson in your company.
How to Get Started –
3. Describe the Customer Describe how easy it is to get started.
Experience Customer Quote –
Provide a quote from a hypothetical customer that
describes how they experienced the benefit.
4. Write a User Manual. Closing and Call to Action –

7
http://www.allthingsdistributed.com
/2006/11/working_backwards.html

Wrap it up; give pointers where the reader should go next.

http://www.quora.com/What-is-Amazons-approach-to-
product-development-and-product-management

Exercise: Write
a Press Release
Innovation Skill: Observing
Working as a group, write a press
release for a product real or imaginary.
1. Aim the press release at
PRESS RELEASE
consumers of the product.
Title –
2. Create a tag line to quickly
Central theme (Tag line) – summarize the central theme.
Summary (10X effect) – 3. Summarize the offer in
Problem – terms of its10X effect.
Solution – 4. Center your message on a
Getting Started –
single theme that will define
what the product is and is not.

8
Call to Action –

Agenda

1. Compelling Offer
Observing
Questioning
3. Adoption Chain
Networking
9
Experimenting

Release Cycle
6 Months
Quick & Dirty Value Stream Map:
Need a Need a
Feature Feature Design Develop Harden UAT

Release Cycle Release Cycle Release Cycle

Value-Added Time
Total Cycle Time Time
Total Cycle
Start Average Start End

Development Model:
 Releases are very painful
 Avoid releases!
Thanks to Kent Beck for his ideas on cadence.

Release Cycle
Quarterly
 Hardening: 2 – 4 weeks
 Early integration testing becomes essential
 Typically: 2 – 4 week iterations
 Code from each iteration goes to integration testing

Business issues (if software is
sold/delivered to customers):
 How to price and sell releases?
 Which releases to support?
 Supporting multiple branches

can create a support nightmare

Release Cycle
Monthly
Now you need:
Hardening  3 days
Cross Functional Team
Visualization
Short Daily Meetings
SBE/TDD really working!

Works best for:
 Software as a Service (SaaS)
[Any download to local machines is pushed.]

 Internal Software

Release Cycle
Weekly/Daily/Continuous
Kanban works well
Iterations become irrelevant The team
Estimating is not very important is everyone.
No branching – Develop on the trunk
Test & deployment automation is essential
Rapid cycles of learning drive portfolio decisions

Things. Just. Work.
 One or our large web-based customers
has been deploying daily for five years!
 Google: gmail deploys 2X/week

 What is the fastest release cycle here?

Immediate Connection
Bent Victor Our
Inventing on Principle
http://vimeo.com/36579366 Guiding Principle:
People should not be
told what to do.
Combining Principles:
Teams should adjust what
Bent’s Guiding Principle: they are doing based on
Creators need an what team members learn
immediate connection
with what they create. directly from their efforts.
Bret Victor invents tools that enable people to understand and
create. He has designed experimental UI concepts at Apple, interactive
data graphics for Al Gore, and musical instruments at Alesis.

Exercise: Draw a
Feedback Matrix
Innovation Skill: Questioning
I adjust what I do based on:
Immediate
visibility of results

1. Have everyone at the
table plot a point for a
typical development
team members.
2. Add additional points
for members from
info about results

different functions.
3. Plot the ideal point
Delayed

for team members
working on that

15
Indirect information
about consumers

August 12 l e a n
Direct connection
with consumers
Copyright©2012 Poppendieck.LLC
compelling offer.

Agenda

1. Compelling Offer
Observing
Questioning
3. Adoption Chain
Networking
16
Experimenting

The Age of the Platform

Contributor Consumer

Platform
(infrastructure and rules)

Adoption Chain
Innovators often forget Complementors!

3G
Music? Repair? Books?
Networks?

The first MP3 Player
The first 3G Phone Run-flat tires The first e-reader
1998
2002 Honda Odyssey 2005 2006


Cost vs. Relative Benefit
Adopters see value differently.
Innovator’s Perspective Adopter’s Perspective
Innovator
++
Relative ++
Benefit
Product Distributor
+
Cost ++
Product Cost
+ Additional Retailer -
Costs/Risks

Consumer
++
+++

New New Old
This adoption chain
Product Product Product
will not work!
Benefit Benefit Benefit

If any organization on the adoption chain perceives

19
a negative value, the adoption chain is broken.
August 12 Copyright©2012 Poppendieck.LLC l e a n The Wide Lens, by Ron Adner

Exercise: Evaluate
the Adoption Chain
Innovation Skill: Networking
You would like to develop a product with compelling value
and provide immediate connection between the product
creators and consumers. In order to achieve this, you need Adoption Chain
the cooperation of organizations within and outside of you
Party 1
++
company that are not part of your immediate team. ++
1. Make a list of the parties who must change the way they
Party 2 +
do things in order for the improvement to be successful.
2. Look at the benefits to be gained and the cost of change Party 3 -
from the perspective of each party.
Party 4 ++
3. Are there parties where the cost outweighs the benefit? +++

4. What can be done to change the equation so the benefits

20
are greater than the cost for that party?


Agenda

1. Compelling Offer
Observing
Questioning
3. Adoption Chain
Networking
21
Experimenting

Validated Assumptions
Only a few large companies have been able to sustain growth
over time by coming up with successful new disruptive
businesses. These companies share a common practice: They
have systems in place that encourage small, cross-functional
employee teams to conduct frugal experiments. Scott Cook, Intuit
Validated Assumptions:
1. List assumptions that must be true in order for the product to succeed.
a) Which ones are most important?
2. Devise experiments to determine whether critical assumptions are true.
a) Create a measurable hypotheses that will demonstrate clear cause and effect.
b) Run many, quick experiments to test the hypotheses.
Lean Startup:
1. Start with a Business Success Model
a) Success metrics which demonstrate clear cause and effect.
2. Establish a baseline – with a Minimum Viable Product

l e a n
3. Target every initiative at improving a success metric
4. Do not add capabilities without validation (eg. split test)
22 August 12 Copyright©2012 Poppendieck.LLC

Experimental Approach

Site visitors are randomly
A assigned to see version A or B B Don’t Miss this excellent paper:
Online Experimentation at Microsoft
by Kohavi , Crook, & Longbotham
Presented at KDD 2009
(Knowledge Discovery & Data Mining)
http://exp-platform.com/expMicrosoft.aspx

14.5% Conversion B is better than A 18.6% Conversion

Amazon.com Case Study 1. Most of the time the guess about how customers
Amazon’s Greg Linden created a prototype will behave will be wrong – even for experts!
which gave personalized recommendations
based on items in the shopping cart. The feature 2. Test early – don’t waste a lot of time on in-
was opposed by a marketing SVP who told Greg depth analysis or planning the perfect design.
to stop, but Greg created a test anyway. He was
3. Test often – most experiments don’t tell much.
allowed to pushed it live. The feature ‘won’ by
such a wide margin that it was immediately 4. A failed experiment is not a failure – it’s a

l e a n
adopted, increasing sales by an estimated 3%. learning opportunity. The only failures are failure
http://glinden.blogspot.com/2006/
04/early-amazon-shopping-cart.html to learn or failing to conduct a good experiment.
23 August 12 Copyrignt©2011 Poppendieck.LLC

Why?
Impact Maps

Why? Who? Who?
Who?
How? How? How?
What?
What? What?

Connections
=Assumptions
24 August 12 Copyright©2012 Poppendieck.LLC l e a n Images and ideas from:
Impact Mapping Handbook
by Gojko Adzic – Available soon
Check out:www.impactmapping.org

Exercise: Find and
Validate Assumptions
Innovation Skill: Experimenting
Validate assumptions for your compelling offer.
1. Draw an Impact Map.
Impact
a) Start with the goal – make it measurable.
Map
b) Who needs to be involved?
c) How will they contribute? Why?
d) What needs to be done to support each contribution? Who?
2. List the assumptions connect each node on the map. How?
3. Prioritize the assumptions; select the top two. What?
4. How will you measure these two assumptions? Assump-
tions?
5.

25
Devise an experiment to validate each assumption.

lsoftware development
e a n
Thank You!
More Information: www.poppendieck.com

mary@poppendieck.com Mary Poppendieck www.poppendieck.com

The Fastest Learner Wins: Continuous Learning Through Rapid Release Cycles

Recommandé

Recommandé

Contenu connexe

Tendances

Tendances (6)

Similaire à The Fastest Learner Wins: Continuous Learning Through Rapid Release Cycles

Similaire à The Fastest Learner Wins: Continuous Learning Through Rapid Release Cycles (20)

Plus de drewz lin

Plus de drewz lin (20)

The Fastest Learner Wins: Continuous Learning Through Rapid Release Cycles