No estimates - a controversial way to improve estimation with results-handouts
-
Upload
vasco-duarte -
Category
Internet
-
view
2.372 -
download
1
Transcript of No estimates - a controversial way to improve estimation with results-handouts
#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_vascohttp://bit.ly/vasco_bloghttp://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
Agility = Chaos!
Just Google
it
#NoEstimates
#Idealist#extremist#chaotic
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...
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 210
1
2
3
4
5
6
7
8
9
10
velocityAverage=LCLUCL
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]
1 3 5 7 9 11 13 15 17 19 210
2
4
6
8
10
12
# of items/stories deliveredLCLUCLaverage
Team: AT
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 160
2
4
6
8
10
12
LCLUCLaverage# of items/stories delivered
Team: RF
1 2 3 4 5 6 7 8 9 10111213141516170
2
4
6
8
10
12
14
16
18
20
LCLUCLaverage# of items/stories delivered
Team: RF2
1 3 5 7 9 11 13 15 17 19 210
5
10
15
20
25
# of items/stories deliveredLCLUCLaverage
Team: SH
1 4 7 10 13 16 19 22 25 28 31 34-1
1
3
5
7
9
11
13
15
17
LCLUCLaverage# of items/stories delivered
Team: K
1 2 3 4 5 6 7 80
2
4
6
8
10
12
14
16
# of items/stories deliveredLCLUCLaverage
Team: MPC
1 3 5 7 9 11 13 15 17 190
2
4
6
8
10
12
14
16
# of items/stories deliveredLCLUCLaverage
Team: AS
1 2 3 4 5 6 7 8 9 100
5
10
15
20
25
30
# of items/stories deliveredLCLUCLaverage
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? Real Data!
A long project
24Sprints
Which metric predicted most accurately the output of the
whole project?
a) After only the first 3 Sprintsb) 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 predicted output: 418 SPs
completed
+20%
The true output: 228 Stories
The predicted output: 220
Stories
-4%!
After just 5 sprints
# of Stories predictive powerStory Points predictive power
The true output: 349,5 SPs
completed
The predicted output: 396 SPs
completed
+13%
The true output: 228 Stories
The predicted output: 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)
1/3
1/10
1/17
1/24
1/31
2/7
2/14
2/21
2/28
3/7
3/14
3/21
3/28
4/4
4/11
4/18
4/25
5/2
5/9
5/16
5/23
5/30
6/6
6/13
6/20
6/27
7/4
7/11
7/18
7/25
8/1
8/8
8/15
8/22
8/29
9/5
9/12
9/19
9/26
10/3
10/10
10/17
10/24
10/31
11/7
0
10
20
30
40
50
60
70
80
6871 71 71 71 71 71 72 72 72 73 73
03
7 7 9 11 12 13
20 20 22 23
Product Backlog Cumulative Flow Diagram
RemainingLinear (Remaining)DoneLinear (Done)
Release on 20th October
2014
Next you will see the forecasted release date when
using Story Points (values 1:2:3)
1/3
1/10
1/17
1/24
1/31
2/7
2/14
2/21
2/28
3/7
3/14
3/21
3/28
4/4
4/11
4/18
4/25
5/2
5/9
5/16
5/23
5/30
6/6
6/13
6/20
6/27
7/4
7/11
7/18
7/25
8/1
8/8
8/15
8/22
8/29
9/5
9/12
9/19
9/26
10/3
10/10
10/17
10/24
10/31
11/7
0
10
20
30
40
50
60
48
51 51 51 51 51 5152 52 52
53 53
02
5 57 8 9 10
15 1517 18
Product Backlog Cumulative Flow Diagram
RemainingLinear (Remaining)DoneLinear (Done)Release on
14th October 2014
Next you will see the forecasted release date when
#NoEstimates (or, all stories = 1 story point)
1/3
1/10
1/17
1/24
1/31
2/7
2/14
2/21
2/28
3/7
3/14
3/21
3/28
4/4
4/11
4/18
4/25
5/2
5/9
5/16
5/23
5/30
6/6
6/13
6/20
6/27
7/4
7/11
7/18
7/25
8/1
8/8
8/15
8/22
8/29
9/5
9/12
9/19
9/26
10/3
10/10
10/17
10/24
10/31
11/7
0
5
10
15
20
25
30
35
28
31 31 31 31 31 3132 32 32
33 33
0 13 3
5 5 6 7
10 1012 13
Product Backlog Cumulative Flow Diagram
RemainingLinear (Remaining)DoneLinear (Done)
Release on 29th September 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, and V.Y. Shen proposed that a good estimation approach should provide estimates that are within 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!