Something wrong with starting late at a Scrum Gathering!

Did this talk before at SD conference at Santa Clara

Think self-organizing teams are fundamental to all agile methods
– people claim: Unified Process is an agile process – but doesn’t rely on self-organizing teams
– Wicked problems: get the right people together, throw them at the problem

TOPICS:

Self Organization and subtle Control – not all forms of control are evil

Containers, Differences and Exchanges Model

Influencing how the team evolves

Premise: self-organization isn’t just locking a team in a room and saying “just do it” – we influence

What is a Self Organizing Team?
– does not mean:
– – the team gets to decide what goal they pursue
– – or even necessarily who is on the team
– – – some self-org teams are given this responsibility

Complex Adaptive Systems
– a dynamic network of many agents
– – acting in parallel
– – acting and reacting to what other agents are doing
– control is highly dispersed and decentralized
– overall system behavior is the result of a huge number of decisions made constantly by many agents
– e.g. QA group
– in a project:
– – sounds like a software project!

Some Examples
– ant colony
– flock of geese
– us right now – self-organized into room, some at front, some near power outlets
– a crowd batched up to get into a concert or sporting event
– – Jimmy Buffet concert – queueing, at the bar, at the beach etc.
– a family preparing, eating, and cleaning up after a meal
– cars and drivers on the heighway
– a software team

Control is not Evil
– was command and control leader before scrum!
– Simple rules or incentives are used to guide or direct behavior
– – “Drive this direction and on this side on the highway.”
– for bioteams, these are provided by nature
– for our teams rules and incentives can be added by managers or leaders… or in some cases by team members
– Generative rules – rules that generate behavior
– Some control is okay

Quote from Philip Anderson, The Biology of Business about Self-organization

Popular to criticize Taylorism – don’t specify exact steps, instead put in place things that guide behavior

(NOTE: slides on website)

Takeuchi & Nonaka “Although project teams are largely on their own, they are not uncontrolled… ”

What this is not
– we’re not talking about
– – being deceptive or sneaky
– – manipulating people
– – IS subtle rules and guidelines
– nothing I’m going to advocate needs to be secret
– – but there may be reasons why you don’t broadcast your reasons
– – e.g. if you have to fire someone

Containers Differences and Exchanges

Glenda Eoyang: Conditions for Self-Organizing in Human Systems
– Container
– – a boundary within which self-organization occurs e.g. project, team, role, nationality
– Differences
– – there must be differences among the agents acting in our system
– – e.g. technical knowledge, domain knowledge, education, experience, power, gender
– – e.g. individual sub-goals
– Transforming Exchanges
– – agents in the system interact and exchange resources
– – information, money, energy (vision)
– how can we use these to influence the way the team behaves?
– – amplify or dampen the differences
– – re-frame the problem
– – change the communication environment

Comment from audience about “Wisdom of Crowds”
– groupthink!

Using the CDE model
– Adjusting the Containers:
– – formal teams, informal teams, clarify (or not) expectations
– – e.g. the AI programmers thought they could not talk with each other, only the people on their teams
– – introduced a Community of Practices
– Differences:
– – dampen or amplify them within or between containers
– – e.g. if people are having a hard time making decisions because they are all too different, maybe adding people to increase similarity
– Exchanges:
– – insert new exchanges, new people, new techniques or tools
– – e.g. team that needed to get outside help re: architecture
– – cross-training

Containers
– enlarge or shrink team

Differences
– don’t require consensus
– – creativity comes from tension
– – quiet disagreement is not as good as fierce debate that leads to behavior change
– _do_ require consensus
– – e.g. if one person is dominating the discussion
– ask hard questions
– – then expect teams to find solutions

Transforming Exchanges:
– remove a document
– create a document!
– encourage communication between teams and groups
– – who isn’t talking
– add or remove people
– – change reporting relationships
– encourage learning

Exercise:

You are the ScrumMaster or PM:
– situations
– ID one thing to change
– use CDE model

Good Group Discussion around Scenarios

Mike Cohn is really good at creating discussion exercises.  I’ve always been impressed.  The discussion excercise asked us to apply the CDE model to the various scenarios.  In our group we only looked at two out of the five scenarios.  Each time the discussion was great – lots of good ideas from people about how to solve the problem in the scenario.  What wasn’t so good at first was using the CDE model.  It’s easy to just look at the scenario and come up with solutions.  What isn’t so easy is to use the model to generate solutions or to map solutions into the model.  At a personal level, I also found that folks in my group were emphasizing imposing solutions rather than using the Scrum model to have solutions emerge from the team’s own efforts.  For example, the retrospective is a Scrum practice that really should be the first line of defense.

Aother Philip Anderson quote:

“Self-organization proceeds from the premise that effective organization is evolved, not designed.  It aims to create an environment in which successful divisions of labour…”

Variation, selection and retention
– evolution is result of these three elements
– consider a giraffe:
– – variation: longer neck
– – selection: helps it reach food
– – retention: more food means more progeny

Seven levers for influencing team evolution:
1.  Select the external environment
– more than the physical environment
– business, industry
– approach to innovation
– approach to mistakes
– types of projects
– expectations about multi-tasking and focus
2. Define performance
– selection – traits that help us survive
– short vs. long-term performance
– providing training
– support sustainable pace
– explore wild ideas
– not exchanging deadlines for unmaintainable code
– e.g “Up or Out” culture – burn out or be promoted!
3. Manage meaning
– stories from leaders
– keeping messages out
– “we will become profitable this quarter”
– rituals
– Story about Mike’s background (1994)
– – valley of death
– – product did not have a long life
– – no new features
– – decided to create new product
– – “valley” in decline of revenue from old product, vs. increase in revenue for new product
– – as part of this, replaced two-ply with one-ply toilet paper to remind everyone of the need to save costs!
– Another story
– – “our GM counts the cars in the lot every day at 5pm” – not a good culture for Scrum!

4) Choose People
– who is on the team
– adjust:
– – team size, decision making style, location, gender, background, motivation

5) Self-selecting members?
– should a delivery team be allowed full control over who is on the team?
– under all circumstances or only some?  which?
– what are the advantages and disadvantages?
– – people often will choose to work with similar people
– doing this is giving up some control
– “you can self-organize unless I disagree” is not a good message!
6) Evolve vicarious selection systems
– variation-selection-retention
– – selection was determining which variations will be retained – can take a long time
– so we often use vicarious selection systems
– – this is an animal that can smell that a food is poisonous, rather than eating it
– using only the marketplace as our selection mechanism takes too long
– Organizations can have vicarious selection systems:
– – retrospectives, Google’s 20% policy which attracts people to projects, compensation
7) Energize the system
– unless energy is pumped into the system, entropy will set in
– make sure the group has a “clear, elevating goal” or an “igniting purpose”
– motivation
– opportunity
– information
– Teamwork by Larson and LaFasco or Hot Spots by Lynda Gratton
– example: Bill Gates and “Internet Tidal Wave” memo

New book by Mike Cohn:

“Succeeding With Agile”

Affiliated Promotions: