David J Anderson - Individually Smart, Collectively Stupid - the great 21st Century Management Challenge

Organizations making some attempt to improve their technology development or IT operations processes are capable of gathering at least basic data on work, customer demand, and capability to supply. One of the simplest and most basic pieces of information is "how long does it take to deliver a piece of work?" and compare that to "How long did we actually spend working on it?" Typical organizations take 10 to 20 times longer to deliver something than they actually spent working on it. Customers right place a high value on time-to-delivery. And yet when it comes to improving process and customer service, most organizations focus on how the work is done and the skills of the workers doing it. This means they are focusing on improving 5-10% of the delivery time and ignoring 90%+, the delays in the flow of work and the root causes of those delays. It seems we are blind to these organizational dysfunctions.
In this key note address, David J. Anderson will look at what causes these delays, why we blind to them, and what we can do about addressing these problems. It's not about the people, it's about the system their operating within. The great 21st Century Management Challenge is to turn the attention of managers to the system they are responsible for operating. David will identify some high leverage ways to switch management focus to the real problem and in doing so make our organizations collectively smart!

Transcript

1. Lean Agile Scotland Edinburgh September 2012 David J. Anderson David J. Anderson & Associates, Inc. dja@djaa.com Twitter @agilemanager Individually Smart, Collectively Stupid
2. Lean Agile Scotland @agilemanager 2001 Agile Manifesto Division 3 is attracting a lot more interest this season!
3. Lean Agile Scotland @agilemanager 2001 Agile Manifesto Division 3 is attracting a lot more interest this season! Adam Smith
4. Lean Agile Scotland @agilemanager 2001 Agile Manifesto Division 3 is attracting a lot more interest this season! It’s not rational, man! Adam Smith
5. Lean Agile Scotland @agilemanager So was sending Rangers to Division 3, an act of collective stupidity? An act of spiteful schadenfreude (taking pleasure from the misfortunes of others). Is it irrational economically, serving no other purpose than providing that feel good factor from gloating over the football results every weekend for the next 3 years? I don't think so!
6. Lean Agile Scotland @agilemanager What’s happening to Rangers is tribal & emotional!
7. Lean Agile Scotland @agilemanager Stack …  To understand how to be collectively smart, we must understand tribal, social behavior. Tribal behavior in the workplace can affect performance and make us look collectively stupid.  but more about tribes later...
8. Lean Agile Scotland @agilemanager Delays
9. Lean Agile Scotland @agilemanager Collective stupidity manifests in poor flow efficiency The first Kanban implementation at Microsoft exhibited 8% flow efficiency initially. Hakan Forss, a consultant from Sweden reports clients typically exhibit < 5% flow efficiency My personal experience is 5% - 15% before any improvement initiative is started time work delay
10. Lean Agile Scotland @agilemanager Flow Efficiency in manufacturing industries can be >70% often >90% Manufacturing workers appear to be collectively smart!
11. Lean Agile Scotland @agilemanager Why is it important to improve Flow Efficiency and eliminate delay?
12. Lean Agile Scotland @agilemanager Little’s Law
13. Lean Agile Scotland @agilemanager Little’s Law Completion Rate Lead Time WIP =
14. Lean Agile Scotland @agilemanager Easily validated by visual inspection of a cumulative flow diagram 0 20 40 60 80 100 120 140 160 180 200 220 240 2/10/04 2/17/04 2/19/04 2/20/04 2/23/04 2/24/04 2/25/04 2/26/04 2/27/04 3/1/04 3/2/04 3/3/04 3/4/04 3/5/04 3/8/04 3/9/04 3/10/04 3/11/04 3/12/04 3/15/04 3/16/04 3/17/04 3/18/04 3/19/04 3/22/04 Device Management Ike II Cumulative Flow Features Time Complete Coded Designed Started Inventory
15. Lean Agile Scotland @agilemanager Easily validated by visual inspection of a cumulative flow diagram 0 20 40 60 80 100 120 140 160 180 200 220 240 2/10/04 2/17/04 2/19/04 2/20/04 2/23/04 2/24/04 2/25/04 2/26/04 2/27/04 3/1/04 3/2/04 3/3/04 3/4/04 3/5/04 3/8/04 3/9/04 3/10/04 3/11/04 3/12/04 3/15/04 3/16/04 3/17/04 3/18/04 3/19/04 3/22/04 Device Management Ike II Cumulative Flow Features Time Complete Coded Designed Started Inventory WIP
16. Lean Agile Scotland @agilemanager Easily validated by visual inspection of a cumulative flow diagram 0 20 40 60 80 100 120 140 160 180 200 220 240 2/10/04 2/17/04 2/19/04 2/20/04 2/23/04 2/24/04 2/25/04 2/26/04 2/27/04 3/1/04 3/2/04 3/3/04 3/4/04 3/5/04 3/8/04 3/9/04 3/10/04 3/11/04 3/12/04 3/15/04 3/16/04 3/17/04 3/18/04 3/19/04 3/22/04 Device Management Ike II Cumulative Flow Features Time Complete Coded Designed Started Inventory Avg. Lead Time
17. Lean Agile Scotland @agilemanager Easily validated by visual inspection of a cumulative flow diagram 0 20 40 60 80 100 120 140 160 180 200 220 240 2/10/04 2/17/04 2/19/04 2/20/04 2/23/04 2/24/04 2/25/04 2/26/04 2/27/04 3/1/04 3/2/04 3/3/04 3/4/04 3/5/04 3/8/04 3/9/04 3/10/04 3/11/04 3/12/04 3/15/04 3/16/04 3/17/04 3/18/04 3/19/04 3/22/04 Device Management Ike II Cumulative Flow Features Time Complete Coded Designed Started Inventory Avg. Comp Rate
18. Lean Agile Scotland @agilemanager Little’s Law Completion Rate Lead Time WIP = We observe that as lead times increases due to delay, we tend to increase WIP in order to have a steady flow of delivery. We keep starting stuff! More WIP tends to cause longer leads times due to multi-tasking and quality issues. We start more stuff! Vicious cycle can spiral out of control! …we increase this …to maintain this,… As this increases,…
19. Lean Agile Scotland @agilemanager In 2005, HP printer firmware division in Boise, Idaho discovered they had 4.5 years of WIP!!! This was invisible until they plotted a CFD based on data from their development tracking system
20. Lean Agile Scotland @agilemanager Induced Work the longer the delay to fix a defect, the longer it takes to fix The longer the delays in designing something the more likelihood of defects. This has been observed to be non-linear. time Defect rate
21. Lean Agile Scotland @agilemanager Eventually, these effects combine to slow the delivery rate to a trickle while large quantities of work remain in progress And finally we stop starting stuff! But we still can't finish our work-in-progress
22. Lean Agile Scotland @agilemanager Cost of Delay
23. Lean Agile Scotland @agilemanager Long lead times add risk and cost. The future is uncertain and faster delivery gives us greater certainty of the utility of what we delivering. Note the use of the word "utility" from economics and not "value". Utility isn't always a tangible monetary value. It merely implies usefulness that didn't previously exist. The longer it takes to deliver something the less certain we can be about its utility. There is an opportunity cost of delay - the risk adjusted utility of what we are building. So faster delivery gives us more certainty and is likely to deliver greater utility.
24. Lean Agile Scotland @agilemanager Stack…  1. Tribal Behavior isn't rational (as Adam Smith defined it)  2. If we are to be collectively smart we must address causes of delay
25. Lean Agile Scotland @agilemanager Lean
26. Lean Agile Scotland @agilemanager How does Lean improve Flow Efficiency?
27. Lean Agile Scotland @agilemanager How does Lean improve Flow Efficiency?  Optimize the whole!
28. Lean Agile Scotland @agilemanager How does Lean improve Flow Efficiency?  Optimize the whole!
29. Lean Agile Scotland @agilemanager How does Lean improve Flow Efficiency?  Optimize the whole!
30. Lean Agile Scotland @agilemanager How does Lean improve Flow Efficiency?  Optimize the whole!  Use a Systems Thinking approach!
31. Lean Agile Scotland @agilemanager How does Lean improve Flow Efficiency?  Optimize the whole!  Use a Systems Thinking approach!
32. Lean Agile Scotland @agilemanager How does Lean improve Flow Efficiency?  Optimize the whole!  Use a Systems Thinking approach!
33. Lean Agile Scotland @agilemanager How does Lean improve Flow Efficiency?  Optimize the whole!  Use a Systems Thinking approach!  To understand this, let’s meet some well known Scottish systems thinkers…
34. Lean Agile Scotland @agilemanager Graeme Obree
35. Lean Agile Scotland @agilemanager Alex Ferguson
36. Lean Agile Scotland @agilemanager So Lean typically employs a system thinking designer like Graeme Obree or Alex Ferguson. The designer designs a new system to optimize the whole and eliminate the waste. But how well does this scale? Obree is a craftsman working empirically on a small scale with few or any collaborators Ferguson is arguably a (benevolent) dictator of a small to medium sized enterprise
37. Lean Agile Scotland @agilemanager Lean text books will tell you to "map the value- stream." In knowledge work, we tend to use an older term "workflow" as the metaphor of a stream implies a single direction and knowledge work flows tend to be complex and involve loops. So we should "Map the Workflow". This gives us a definition of our current process.
38. Lean Agile Scotland @agilemanager How does Lean improve Flow Efficiency?
39. Lean Agile Scotland @agilemanager How does Lean improve Flow Efficiency?  Optimize the whole!
40. Lean Agile Scotland @agilemanager How does Lean improve Flow Efficiency?  Optimize the whole!
41. Lean Agile Scotland @agilemanager How does Lean improve Flow Efficiency?  Optimize the whole!
42. Lean Agile Scotland @agilemanager How does Lean improve Flow Efficiency?  Optimize the whole!  Use a Systems Thinking approach!
43. Lean Agile Scotland @agilemanager How does Lean improve Flow Efficiency?  Optimize the whole!  Use a Systems Thinking approach!
44. Lean Agile Scotland @agilemanager How does Lean improve Flow Efficiency?  Optimize the whole!  Use a Systems Thinking approach!
45. Lean Agile Scotland @agilemanager How does Lean improve Flow Efficiency?  Optimize the whole!  Use a Systems Thinking approach!  Identify wasteful, delaying activities and eliminate them
46. Lean Agile Scotland @agilemanager For knowledge work, this means identifying all the delays and in turn the causes of delay, and then designing them out of our future process. So Lean employs a system thinking designer to design out causes of delay and define our new process. The Lean consultant will then plan and manage the transition from the old process to the new process.
47. Lean Agile Scotland @agilemanager So let's take a look at how an experienced and successful coach introduced a number of famous Scots to their new future state process…
48. Lean Agile Scotland @agilemanager [Video clip from Damned United]
49. Lean Agile Scotland @agilemanager The team may not always accept the proposed new process definition (or the designers opinion)!
50. Lean Agile Scotland @agilemanager Stack…  1. Tribal Behavior isn't rational (as Adam Smith defined it)  2. If we are to be collectively smart we must address causes of delay  3. People (often) resist defined & managed change for emotional reasons
51. Lean Agile Scotland @agilemanager Self-organization
52. Lean Agile Scotland @agilemanager So, if people will resist being told what to do, or perhaps our grand system designer can't get it right all of the time, … the Agile community argues that we should allow individuals to self-organize to produce the best outcome - it's written into the Principles behind the Agile Manifesto. The father of self-organization is, of course, a Scotsman...
53. Lean Agile Scotland @agilemanager Adam Smith’s Invisible Hand
54. Lean Agile Scotland @agilemanager Smith believed that all we required was to construct a set of rational incentives to guide individual behavior and markets would self- organize around optimal, efficient solutions. Loosely regulated markets would produce the optimal system outcome as each individual would act in his/her own best interests.
55. Lean Agile Scotland @agilemanager Couple this with Cartesian decomposition and it was inevitable that the Scottish- French version of the Renaissance would deliver us to Taylorism and the invention of another great Scotsman, Donaldson Brown, CFO at GM, cost accounting. If we were to optimize the whole, all we needed to do was optimize the parts by asking each to act rationally.
56. Lean Agile Scotland @agilemanager Couple this with Cartesian decomposition and it was inevitable that the Scottish- French version of the Renaissance would deliver us to Taylorism and the invention of another great Scotsman, Renee Descartes Donaldson Brown, CFO at GM, cost accounting. If we were to optimize the whole, all we needed to do was optimize the parts by asking each to act rationally.
57. Lean Agile Scotland @agilemanager Couple this with Cartesian decomposition and it was inevitable that the Scottish- French version of the Renaissance would deliver us to Taylorism and the invention of another great Scotsman, Renee Descartes Fredrick Taylor Donaldson Brown, CFO at GM, cost accounting. If we were to optimize the whole, all we needed to do was optimize the parts by asking each to act rationally.
58. Lean Agile Scotland @agilemanager Couple this with Cartesian decomposition and it was inevitable that the Scottish- French version of the Renaissance would deliver us to Taylorism and the invention of another great Scotsman, Renee Descartes Fredrick Taylor Donaldson Brown, CFO at GM, cost accounting. If we were to optimize the whole, all we needed to do was optimize the parts by asking each to act rationally. Donaldson Brown
59. Lean Agile Scotland @agilemanager Unfortunately, it is this line of thinking that caused the need for a Lean revolution.
60. Lean Agile Scotland @agilemanager In recent times, the Internet has enabled and encouraged self-organization on a massive scale. But the best of these, such as the open source software development communities like Linux, do not survive without a benevolent dictator. The Invisible Hand is indeed very visible and attached to a human leader.
61. Lean Agile Scotland @agilemanager We might conclude from this that self-organization of large systems of humans is challenging.
62. Lean Agile Scotland @agilemanager The alternative to Smith's Invisible Hand was constructed by an Englishman - Thomas Hobbes and his Leviathan The concept is simple -
63. Lean Agile Scotland @agilemanager It was an English version of the Renaissance that gave us socialist 5 year plans and command and control structures in corporations - albeit a Frenchman, Henri Fayol, who documented it as an approach to management around the turn of the 20th Century.
64. Lean Agile Scotland @agilemanager It was an English version of the Renaissance that gave us socialist 5 year plans and command and control structures in corporations - albeit a Frenchman, Henri Fayol, who documented it as an approach to management around the turn of the 20th Century. Henri Fayol
65. Lean Agile Scotland @agilemanager Now in the 21st Century, we are aware that both of these extremes don't work. The command and control Leviathan with its designed outcomes and 5 year master plans have been shown to fail us. Equally, we suffer at the Invisible Hand that supposedly guided the markets to act rationally and optimize the outcome.
66. Lean Agile Scotland @agilemanager Men (and women) don't act rationally. Modern knowledge workers will not be commanded and controlled. And those that would command and control are often working with flawed plans and designs built on incomplete models and false assumptions.
67. Lean Agile Scotland @agilemanager Stack…  1. Tribal Behavior isn't rational (as Adam Smith defined it)  2. If we are to be collectively smart we must address causes of delay  3. People (often) resist defined & managed change for emotional reasons  4. After 250 years of the Leviathan & the Invisible Hand, we need a new philosophy to guide the organization of human activity in the 21st Century
68. Lean Agile Scotland @agilemanager Complexity
69. Lean Agile Scotland @agilemanager So why didn’t Mark Cavendish win Gold at the Olympics?...
70. Lean Agile Scotland @agilemanager So why didn’t Mark Cavendish win Gold at the Olympics?... Mark Cavendish
71. Lean Agile Scotland @agilemanager [video clip of Mark Cavendish]
72. Lean Agile Scotland @agilemanager The point of this story is that the obvious script based on observed team capabilities didn't play out because of the psychology of the people involved and pre-race events that changed perception. Had Cavendish not won the final stage of the Tour De France so convincingly, he may indeed be the Olympic champion today.
73. Lean Agile Scotland @agilemanager The British, the Germans and the Australians were "all-in" with their strategy of a sprint finish with their top sprinter taking the Gold. They lacked adaptability and despite this lack of adaptability they also lacked the ability to collaborate in order for their strategy to have a chance. They were collectively stupid and sprinted for 17th place to save face.
74. Lean Agile Scotland @agilemanager Risk (or, the future is uncertain)
75. Lean Agile Scotland @agilemanager Risk is the idea that things may not work out exactly as we would like them to. If we wanted to measure risk it is the chance and the magnitude of an actual outcome being different from our intended outcome. Managing risk is the science of minimizing the difference between a desired outcome and an actual outcome
76. Lean Agile Scotland @agilemanager The British, Australian and German cycling teams failed to manage their risk appropriately. The complexity of the situation meant that effective risk management required collaboration that did not emerge. If we ran the race over again, the outcome would be different, as the starting psychological starting conditions would be different.
77. Lean Agile Scotland @agilemanager Stack…  1. Tribal Behavior isn't rational (as Adam Smith defined it)  2. If we are to be collectively smart we must address causes of delay  3. People (often) resist defined & managed change for emotional reasons  4. After 250 years of the Leviathan & the Invisible Hand, we need a new philosophy to guide the organization of human activity in the 21st Century  5. Human collaboration creates complexity from psychological and sociological influences. Results will change based on the starting conditions, all other things being equal
78. Lean Agile Scotland @agilemanager Risk in Knowledge Work
79. Lean Agile Scotland @agilemanager Unlike manufacturing industry, we often don't know exactly what benefit will be derived from a piece of software. It is almost impossible to determine what a feature or function might be worth. Individually, one might be worthless, but combined with others it will have some utility but precisely how much? This can generally only be determined retrospectively.
80. Lean Agile Scotland @agilemanager As a result, we can't know with certainty which ideas for features and functions will produce the greatest utility. Our reaction to this is to hedge our bets by working on more than one at a time. Hence, due to the nature of knowledge work, multi-tasking is inevitable. It is the economically smart thing to do. It turns out, some multi-tasking isn't stupid at all, it's smart. Multi-tasking helps us mitigate risk. Multi-tasking buys us options. Options have value in the face of uncertainty.
81. Lean Agile Scotland @agilemanager But multi-tasking means things take longer, and longer is likely to affect quality, and incur a cost of delay. So there is a conflict!
82. Lean Agile Scotland @agilemanager This should immediately make you think there is an optimization problem here. What is the optimum amount of multi-tasking (or additional work-in-progress) to mitigate the uncertainty of the future while providing the shortest possible lead times to maximize the utility of the work-in- progress?
83. Lean Agile Scotland @agilemanager The truth is there is unlikely to be a recipe for this. The answer will be situational and context specific. As we can only have perfect knowledge from hindsight, the nature of this problem, means it is unlikely we can design a process to produce an optimal outcome. The GB Cycling team showed us this dramatically during the Olympics.
84. Lean Agile Scotland @agilemanager Instead we want our system to be adaptive and for it to evolve to produce the fittest possible process given current circumstances. And should circumstances change, for that evolutionary capability to be inherent to our system so that it can adapt again and again, optimizing the whole around the current economic challenges of our business and the risks we are managing.
85. Lean Agile Scotland @agilemanager Stack…  1. Tribal Behavior isn't rational (as Adam Smith defined it)  2. If we are to be collectively smart we must address causes of delay  3. People (often) resist defined & managed change for emotional reasons  4. After 250 years of the Leviathan & the Invisible Hand, we need a new philosophy to guide the organization of human activity in the 21st Century  5. Human collaboration creates complexity from psychological and sociological influences. Results will change based on the starting conditions, all other things being equal  6. To cope adequately with complexity and risk we need a system that is adaptive and has an inherent evolutionary capability
86. Lean Agile Scotland @agilemanager Introduction to Kanban
87. Lean Agile Scotland @agilemanager Kanban systems model workflow and limit WIP at each step, signaling available capacity, effectively creating a pull system for work orders. First used at Microsoft in 2005 White boards were introduced in 2007 to visualize workflow, signal capacity and show work orders flowing through the system
88. Lean Agile Scotland @agilemanager Boards typically visualize a workflow as columns (input) left to (output) right
89. Lean Agile Scotland @agilemanager Boards typically visualize a workflow as columns (input) left to (output) right Pull Flow – from Engineering Ready to Release Ready WIP Limit – regulates work at each stage in the process
90. Lean Agile Scotland @agilemanager Tickets on the board are “committed” Backlog items remain “options” Kanban systems defer commitment & enable dynamic prioritization when “pull” is signaled 5 4 4 3 2 2 Input Queue Dev Ready In Prog Done Build Ready Test Release Ready Stage Prod. Done In Prog Development Analysis Commitment point
91. Lean Agile Scotland @agilemanager Tickets on the board are “committed” Backlog items remain “options” Kanban systems defer commitment & enable dynamic prioritization when “pull” is signaled 5 4 4 3 2 2 Input Queue Dev Ready In Prog Done Build Ready Test Release Ready Stage Prod. Done In Prog Development Analysis Commitment point
92. Lean Agile Scotland @agilemanager Tickets on the board are “committed” Backlog items remain “options” Kanban systems defer commitment & enable dynamic prioritization when “pull” is signaled 5 4 4 3 2 2 Input Queue Dev Ready In Prog Done Build Ready Test Release Ready Stage Prod. Done In Prog Development Analysis Commitment point
93. Lean Agile Scotland @agilemanager Tickets on the board are “committed” Backlog items remain “options” Kanban systems defer commitment & enable dynamic prioritization when “pull” is signaled 5 4 4 3 2 2 Input Queue Dev Ready In Prog Done Build Ready Test Release Ready Stage Prod. Done In Prog Development Analysis Commitment point
94. Lean Agile Scotland @agilemanager Tickets on the board are “committed” Backlog items remain “options” Kanban systems defer commitment & enable dynamic prioritization when “pull” is signaled 5 4 4 3 2 2 Input Queue Dev Ready In Prog Done Build Ready Test Release Ready Stage Prod. Done In Prog Development Analysis Commitment point
95. Lean Agile Scotland @agilemanager Tickets on the board are “committed” Backlog items remain “options” Kanban systems defer commitment & enable dynamic prioritization when “pull” is signaled 5 4 4 3 2 2 Input Queue Dev Ready In Prog Done Build Ready Test Release Ready Stage Prod. Done In Prog Development Analysis Commitment point
96. Lean Agile Scotland @agilemanager Tickets on the board are “committed” Backlog items remain “options” Kanban systems defer commitment & enable dynamic prioritization when “pull” is signaled 5 4 4 3 2 2 Input Queue Dev Ready In Prog Done Build Ready Test Release Ready Stage Prod. Done In Prog Development Analysis Commitment point
97. Lean Agile Scotland @agilemanager Tickets on the board are “committed” Backlog items remain “options” Kanban systems defer commitment & enable dynamic prioritization when “pull” is signaled 5 4 4 3 2 2 Input Queue Dev Ready In Prog Done Build Ready Test Release Ready Stage Prod. Done In Prog Development Analysis Commitment point
98. Lean Agile Scotland @agilemanager Tickets on the board are “committed” Backlog items remain “options” Kanban systems defer commitment & enable dynamic prioritization when “pull” is signaled 5 4 4 3 2 2 Flow Input Queue Dev Ready In Prog Done Build Ready Test Release Ready Stage Prod. Done In Prog Development Analysis Commitment point
99. Lean Agile Scotland @agilemanager Tickets on the board are committed. Items in the backlog are merely options 5 4 4 3 2 2 Input Queue Dev Ready In Prog Done Build Ready Test Release Ready Stage Prod. Done In Prog Development Analysis Pink tickets show blocking issues
100. Lean Agile Scotland @agilemanager Pull criteria policies encourage a focus on quality & progress with imperfect information 5 4 4 3 2 2 Input Queue Dev Ready In Prog Done Build Ready Test Release Ready Stage Prod. Done In Prog Development Analysis Policies ~~~~~~ ~ ~ ~ ~ Policies ~~~~~~ ~ ~ ~ ~ Policies ~~~~~~ ~ ~ ~ ~ Policies ~~~~~~ ~ ~ ~ ~
101. Lean Agile Scotland @agilemanager Cost of delay function sketches provide a qualitative taxonomy to delineate classes of risk Expedite – white; critical and immediate cost of delay; can exceed other kanban limit (bumps other work); 1st priority - limit 1 Fixed date – orange; cost of delay goes up significantly after deadline; Start early enough & dynamically prioritize to insure on-time delivery Standard - yellow; cost of delay is shallow but accelerates before leveling out; provide a reasonable lead-time expectation Intangible – blue; cost of delay is not incurred until significantly later; important but lowest priority time impact time time time impact impact impact Colour Function
102. Lean Agile Scotland @agilemanager Allocate capacity across classes of service mapped against demand 5 4 4 3 2 2= 20 total Allocation 10 = 50% +1 = +5% 4 = 20% 6 = 30%
103. Lean Agile Scotland @agilemanager Allocate capacity across classes of service mapped against demand 5 4 4 3 2 2= 20 total Allocation 10 = 50% +1 = +5% 4 = 20% 6 = 30% Input Queue Dev Ready In Prog Done Done In Prog Development Analysis Build Ready Test Release Ready
104. Lean Agile Scotland @agilemanager Benefits of a Kanban System  There are predictable benefits from the merely complicated, mechanical nature of the system...  Deferred Commitment  Reduced & more predictable lead times  Improved quality from removing overburdening  Improved quality due to focus  Controlled quality due to explicit policies
105. Lean Agile Scotland @agilemanager The Kanban Method  What I didn’t predict was that kanban systems, coupled with visualization would create a positive tension catalyzing process evolution  Seeing (lack of) flow, understanding system level effects, enabled improvements in small evolutionary steps  While kanban systems dealt with merely complicated system dynamics, they enabled an evolutionary response to complex problems
106. Lean Agile Scotland @agilemanager Definition of The Kanban Method
107. Lean Agile Scotland @agilemanager Foundational Principles 1. Start with what you do now 2. Agree to pursue incremental, evolutionary change 3. Initially, respect current roles, responsibilities & job titles 4. Encourage acts of leadership at all levels in your organization
108. Lean Agile Scotland @agilemanager Kanban in Action tried to visually capture the principles
109. Lean Agile Scotland @agilemanager Core practices to enable evolutionary capability 1. Visualize 2. Limit Work-in-Progress 3. Manage Flow 4. Make Policies Explicit 5. Implement Feedback Loops 6. Improve Collaboratively (using models & scientific method)
110. Lean Agile Scotland @agilemanager 1. Understand Sources of Dissatisfaction  From viewpoint of internal & external stakeholders  Source of variability that cause dissatisfaction 2. Demand and Capability Analysis  (Ideally) By work item type & class of service 3. Risk Analysis  Identify dimensions to manage & taxonomies 4. Model Workflow  Understand the knowledge discovery process by type 5. Kanban System Design 6. Visualization 7. Roll out Plan  Negotiation & shuttle diplomacy System Thinking Approach to Introducing Kanban This process tends to be iterative
111. Lean Agile Scotland @agilemanager Philosophy Behind The Kanban Method
112. Lean Agile Scotland @agilemanager People resist change for emotional reasons. If resistance wasn't emotional, the logical, reasoning part of the brain could be persuaded with rational argument showing improved economics or risk management.
113. Lean Agile Scotland @agilemanager Emotional reasoning, in this sense, means the older parts of our brain that process sensory information and match patterns rather than the more recent, in evolutionary terms, logical reasoning part of our brains. The mistake is to assume humans behave based on logical reasoning. They don't when there is a dissonance between the newer logical, thinking part of our brain and our older, sensory, emotional brain, the emotional part always wins.
114. Lean Agile Scotland @agilemanager There is a psychological and sociological component to resistance to change: The psychology of the individual ego, self-image and how they derive self-esteem. The sociology of the workplace tribe, an individuals position in the tribal hierarchy, how they are respected and valued within the tribe, and the tribes own self-image and how it derives its tribal pride and sense of worth.
115. Lean Agile Scotland @agilemanager When a change is perceived to threaten the individual’s sense of self or the tribe then it will be resisted.
116. Lean Agile Scotland @agilemanager Kanban embraces the Zen influenced philosophy of Bruce Lee. Kanban should be like water! It should flow around the rock! The rock is emotional resistance to change.
117. Lean Agile Scotland @agilemanager Initial Kanban implementations should be designed to avoid emotional resistance by anticipating the self-image, and tribal memberships of the individuals involved. Kanban implementers predict how self- esteem and tribal worth are derived and avoid changing mechanisms that are core to the psychology and sociology of the individuals involved in the change.
118. Lean Agile Scotland @agilemanager The systems thinking approach to introducing Kanban is designed to identify what can be improved. The implementation should then be designed to introduce only the changes that will not meet with initial resistance while highlighting and raising awareness of possible future improvements.
119. Lean Agile Scotland @agilemanager How many psychologists does it take to change a lightbulb?
120. Lean Agile Scotland @agilemanager Kanban's visual, tactile and collaborative nature acts to move the participants to conclude for themselves that changes need to be made! The lightbulb decides to change itself!
121. Lean Agile Scotland @agilemanager There are no teams or organizational structures defined in Kanban. Keeping existing organizational structures avoids emotional resistance. Changes to roles, team and organization structures must be self-motivated. Roles and organizational structures must evolve.
122. Lean Agile Scotland @agilemanager It turns out that the initial implementation of the kanban (pull) system can often be argued logically because it does not invoke an emotional reaction. Kanban does not threaten the self-image of individuals or the tribes they are members of.
123. Lean Agile Scotland @agilemanager As a result many of the predictable benefits from the mechanical side of implementing a kanban system are easily realized. Predictability improves Lead times are reduced Quality improves But often it is not enough. To cope with the uncertainties of demand and future risks, the organization needs a capability to evolve.
124. Lean Agile Scotland @agilemanager Kanban is not a software development lifecycle or project management process. It is a method for creating institutional evolutionary capability. WIP limits and visualization and a focus on flow provide the tension to catalyze discussion of improvement. Hence the cartoon on the cover of the book!
125. Lean Agile Scotland @agilemanager Conclusions
126. Lean Agile Scotland @agilemanager And so, to pop some things from our stack…  Kanban is an approach to institutional evolutionary capability. This enables an organization to better cope with complexity.
127. Lean Agile Scotland @agilemanager And so, to pop some things from our stack…  Kanban is an approach to institutional evolutionary capability. This enables an organization to better cope with complexity  Whether a new method or process will be embraced and implemented successfully depends a lot on the starting conditions, and the psychological and sociological elements in play - just like planning to win the Gold medal in the Men's Road Race! As a result, every implementation has to be unique and the results from one implementation cannot fully predict the outcome elsewhere
128. Lean Agile Scotland @agilemanager And so, to pop some things from our stack…  Whether a new method or process will be embraced and implemented successfully depends a lot on the starting conditions, and the psychological and sociological elements in play - just like planning to win the Gold medal in the Men's Road Race! As a result, every implementation has to be unique and the results from one implementation cannot fully predict the outcome elsewhere  The field of Behavioral Economics is emerging to incorporate human psychology (neuropsychology) and sociology into our models for economic behavior
129. Lean Agile Scotland @agilemanager And so, to pop some things from our stack…  The field of Behavioral Economics is emerging to incorporate human psychology (neuropsychology) and sociology into our models for economic behavior  Defined and managed change on a large scale doesn't work because the people involved resist the changes. Change must be self- motivated and the organization must be willing to experiment, mutate and evolve to cope with changes in the market and the economic environment.
130. Lean Agile Scotland @agilemanager And so, to pop some things from our stack…  Defined and managed change on a large scale doesn't work because the people involved resist the changes. Change must be self-motivated and the organization must be willing to experiment, mutate and evolve to cope with changes in the market and the economic environment.  We need a new philosophy to guide how we organize for the 21st Century economy. The dichotomy of the Invisible Hand versus the Leviathan no longer serves us well. Self- organization requires leadership and a guiding light by which to steer it. Controls need to define clear boundaries and be neatly aligned to the risks we are managing. Both the True North that guides self-organization and the rules that define the boundaries of empowerment must be capable of evolving as conditions change.
131. Lean Agile Scotland @agilemanager How to be Collectively Smart?
132. Lean Agile Scotland @agilemanager Waste has to be eliminated by a self-aware system that mutates and evolves using models to guide changes gradually improving shortening lead times improving flow efficiency minimizing cost of delay and managing business risk better.
133. Lean Agile Scotland @agilemanager The guiding light is faster delivery of work to produce the better economic results at a sustainable pace. We must visualize key risks such as cost of delay, customers to be served and the quantity and nature of their demand
134. Lean Agile Scotland @agilemanager An institutional capability to reflect and adapt needs to be present and we need leadership that is tolerant to failure and encourages experimentation. Our organization should be constantly adapting and evolving. In doing so waste will be eliminated – not through grand design & managed change - but incrementally through evolutionary adaptation - self-motivated change, led from within and implemented without resistance.
135. Lean Agile Scotland @agilemanager A Decade of Influences Published June 2012 115,000 words of anecdotes explaining my approach to leadership, management & change
136. Lean Agile Scotland @agilemanager Kanban Method Explained Published April 2010 A 72,000 word intro to the topic
137. Lean Agile Scotland @agilemanager Thank you! dja@djaa.com http://djaa.com/
138. http:// www.limitedwipsociety.org Yahoo! Groups: kanbandev Yahoo! Groups: kanbanops http://leankanbanuniversity.com
139. Lean Agile Scotland @agilemanager About… David Anderson is a thought leader in managing effective software teams. He leads a consulting, training, publishing and event planning business dedicated to developing, promoting and implementing sustainable evolutionary approaches for management of knowledge workers. He has 30 years experience in the high technology industry starting with computer games in the early 1980’s. He has led software teams delivering superior productivity and quality using innovative methods at large companies such as Sprint and Motorola. David is the author of three books. The latest is Lessons in Agile Management – on the Road to Kanban. 2010 saw publication of the best selling Kanban – Successful Evolutionary Change for your Technology Business. David is a founder of the Lean Kanban University, a business dedicated to assuring quality training in Lean and Kanban for knowledge work throughout the world. http://leankanbanuniversity.com Email: dja@djaa.com Twitter: agilemanager
140. Lean Agile Scotland @agilemanager German published January, 2011 Translation by Arne Roock & Henning Wolf of IT-Agile
141. Lean Agile Scotland @agilemanager Spanish published May 2011 Translated by Masa K Maeda, PhD
142. Lean Agile Scotland @agilemanager Israel Amdocs Answers.com TypeMock Australia Lonely Planet Telstra New Zealand Ministry of Social Development Brazil Petrobras CESAR Phidelils O Globo Argentina Huddle Thomson- Reuters Kanban System Adoption Examples Globally USA McKesson Vanguard GoDaddy Xbox Motley Fool CityGrid Media Ultimate Software Constant Contact SEP REI Robert Bosch UK BBC IPC Media Financial Times Microsoft Scandinavia Unibet Volvo Skania Spotify Ericsson Mainland EU Ubuntu Xing BWin ASR BBVA China & HK Thomson-Reuters Nike Chile LAN
143. Lean Agile Scotland @agilemanager Kanban System Adoption by Industry  Media  Includes BBC, Sky, Lonely Planet, Time/Life, IPC, Mobile.de, O Globo, Financial Times, NBC Universal, Thomson-Reuters  Games  Mostly small studios includes video arcade thru mobile games to online gambling such as Unibet & Bwin  Manufacturing  Includes Robert Bosch, Volvo, Skania, Petrobras, Nike  Finance & Insurance  Vanguard, Motley Fool, Chase, ASR  Software & Telecoms  Amdocs, Ultimate, Constant Contact, Phidelis, SEP, Huddle, CESAR, Ubuntu  Public Sector  Ministry of Defence (Denmark), Ministry of Social Development (New Zealand)