Upload
vasco-duarte
View
755
Download
0
Embed Size (px)
DESCRIPTION
Often we hear that estimating a project is a must. "We can't make decisions without them" we hear often. In this session I'll present examples of how we can predict a release date of a project without any estimates, only relying on easily available data. I'll show how we can follow progress on a project at all times without having to rely on guesswork, and we will review how large, very large and small projects have already benefited from this in the past. At the end of the session you will be ready to start your own #NoEstimates journey.
Citation preview
#NoEstimates
Vasco Duarte@duarte_vasco
A way to improve estimates that gives you results!
Learn more about NoEstimates:
• How it can help you turn around a failing project
• How it can help you show what is possible and stick to that
• How it can help you find very early if you are late (and get your manager, or customer, to believe you)
• How to apply #NoEstimates without threatening anyone
Become a Beta Reader and get the book for free!
http://NoEstimatesBook.com
Vasco Duarte@duarte_vasco
http://bit.ly/vasco_blog
http://bit.ly/vasco_slideshare
http://NoEstimatesBook.com
#NoEstimates
pictoquotes
Kent Beck – Extreme Programming
Ken Schwaber - Scrum
Taiichi Ohno – Toyota Production System
Edwards W. Deming – Everything above...
“If I have seen further it is by standing on the shoulders of giants” - Isaac Newton
Just Google
it
Customer Collaboration over Contract NegotiationResponding to Change over Following a Plan
#NoEstimates is easy!
1.Select the most important piece of work you need to do
2.Break that work down into risk-neutral chunks of work
3.Develop each piece of work4.Iterate and refactor
#NoEstimates How-to
Is the system of development stable?
(ref: SPC)
I AM GOING TO GO AHEAD AND
ASK YOU TO DELIVER 10
STORIES NEXT SPRINT...
0
1
2
3
4
5
6
7
8
9
10
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
velocity
Average=
LCL
UCL
Target
Actual, measured throughput over 21 sprints
WTF!!!!!!#%&!
Can we use the data we observe to predict the system throughput and detect changes that affect system
stability?
1.Velocity outside limits 3 times in a row (“outside limits”)
2.There are 5 or more points in sequence (“run test”)
System stability rules
More in the 1-day #NoEstimates WorkshopInformation by email: [email protected]
0
2
4
6
8
10
12
1 3 5 7 9 11 13 15 17 19 21
# of items/storiesdelivered
LCL
UCL
average
Team: AT
0
2
4
6
8
10
12
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
LCL
UCL
average
# of items/storiesdelivered
Team: RF
0
2
4
6
8
10
12
14
16
18
20
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
LCL
UCL
average
# of items/storiesdelivered
Team: RF2
0
5
10
15
20
25
1 3 5 7 9 11 13 15 17 19 21
# of items/storiesdelivered
LCL
UCL
average
Team: SH
-1
1
3
5
7
9
11
13
15
17
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35
LCL
UCL
average
# of items/storiesdelivered
Team: K
0
2
4
6
8
10
12
14
16
1 2 3 4 5 6 7 8
# of items/storiesdelivered
LCL
UCL
average
Team: MPC
0
2
4
6
8
10
12
14
16
1 2 3 4 5 6 7 8 9 1011121314151617181920
# of items/storiesdelivered
LCL
UCL
average
Team: AS
0
5
10
15
20
25
30
1 2 3 4 5 6 7 8 9 10
# of items/storiesdelivered
LCL
UCL
average
Team: FC
#NoEstimates delivers!
Counting Stories vs. Estimated Story Points
Q: Which ”metric” is more accurate when compared to
what actually happened in the project?
A long project
24Sprints
Which metric predicted most accurately the output of the
whole project?
a) After only the first 3 Sprints
b) After only the first 5 Sprints
Disclaimer...This is only one project!
Find 21 more at: http://bit.ly/NoEstimatesProjectsDB
After just 3 sprints
# of Stories predictive powerStory Points predictive power
The true output: 349,5 SPs
completed
The predictedoutput: 418 SPs
completed
+20%
The true output: 228 Stories
The predictedoutput: 220
Stories
-4%!
After just 5 sprints
# of Stories predictive powerStory Points predictive power
The true output: 349,5 SPs
completed
The predictedoutput: 396 SPs
completed
+13%
The true output: 228 Stories
The predictedoutput: 220
Stories
-4%!
Q: Which ”metric” is more accurate when compared to
what actually happened in the project?
But there is more...
#NoEstimates
RegularEstimates
“The chart is a snapshot of one team of 20+ teams over a 2 year period.” – Cory Foy
Which is more
predictable?
What difference does a Story Point make in a project that used both Story Points and
#NoEstimates?
Next you will see the forecasted release date when
using Story Points (values 1:3:5)
6871 71 71 71 71 71 72 72 72 73 73
0 3 7 7 9 11 12 13
20 20 22 23
0
10
20
30
40
50
60
70
80
90
100
1/3 1/17 1/31 2/14 2/28 3/14 3/28 4/11 4/25 5/9 5/23 6/6 6/20 7/4 7/18 8/1 8/15 8/29 9/12 9/26 10/10 10/24 11/7
Product Backlog Cumulative Flow Diagram
Remaining
Done
Linear (Remaining)
Linear (Done)
Release on 20th October
2014
Next you will see the forecasted release date when
using Story Points (values 1:2:3)
48
51 51 51 51 51 51 52 52 52 53 53
0 2 5 5 7 8 9 1015 15 17 18
0
10
20
30
40
50
60
70
80
1/3 1/17 1/31 2/14 2/28 3/14 3/28 4/11 4/25 5/9 5/23 6/6 6/20 7/4 7/18 8/1 8/15 8/29 9/12 9/26 10/10 10/24 11/7
Product Backlog Cumulative Flow Diagram
Remaining
Done
Linear (Remaining)
Linear (Done)Release on
14th October 2014
Next you will see the forecasted release date when
#NoEstimates (or, all stories = 1 story point)
28
31 31 31 31 31 3132 32 32
33 33
0 1 3 35 5 6 7
10 1012 13
0
10
20
30
40
50
60
1/3 1/17 1/31 2/14 2/28 3/14 3/28 4/11 4/25 5/9 5/23 6/6 6/20 7/4 7/18 8/1 8/15 8/29 9/12 9/26 10/10 10/24 11/7
Product Backlog Cumulative Flow Diagram
Remaining
Done
Linear (Remaining)
Linear (Done)
Release on 29thSeptember 2014
Conclusion...
All dates within 3 weeks of each other in a 38 to 42 week
project (a margin of ~8%)
Data used with permission from Bill Hanlon at Microsoft
”At that point, I stopped thinking that estimating
was important.”
Bill Hanlon: http://bit.ly/BHanlon
In 1986, Profs. S.D. Conte, H.E. Dunsmoir, andV.Y. Shen proposed that a good estimationapproach should provide estimates that arewithin 25% of the actual results 75% of the time
--Steve McConnel, Software Estimation: Demystifying the Black Art
In this presentation you have seen examples that far outperform what estimation specialists consider a ”good estimation”. In common they have one way to look at work: #NoEstimates
#NoEstimates testimonial
The deviation between estimated and actual velocity would have been approximately 15% lower if we would have used #NoEstimates.
We have analyzed data from 50 Sprints…
…at no time the story point based estimation was better than #NoEstimates.
One more thing...
80% Late or Failed
Source: Software Estimation by Steve McConnell
The larger the project, the bigger the problem
Source: Software Estimation by Steve McConnell
Source: Software Estimation by Steve McConnell
Comparison of 17 projects ending between 2001 and 2003. (Average: 62%)
Take #NoEstimates and experiment!
Learn, Be Agile!
Learn more about NoEstimates:
• How it can help you turn around a failing project
• How it can help you show what is possible and stick to that
• How it can help you find very early if you are late (and get your manager, or customer, to believe you)
• How to apply #NoEstimates without threatening anyone
http://NoEstimatesBook.com
Become a Beta Reader and get the book for free!