Your Training Partner
Techniques Toolbox
The shared STAR compass of a group of five. Four axes start from the centre: Separateness at the top, Tuning on the right, Action at the bottom and Reason to work together on the left. Each axis runs from low at the centre to high at the rim. The five Tuning dots sit near the rim. The five Action dots sit near the centre. The Separateness dots spread along the whole axis. The Reason dots cluster in the middle. A box gives the result: "We get along well but accomplish little." A second box names the axis to strengthen: Action.

Generative Relationships STAR

Generative Relationships STAR is a 25-minute meeting structure in which a team assesses its working relationships. Each member places one dot on each of the four axes of the STAR compass: Separateness (the diversity of the group), Tuning (mutual listening), Action (the opportunities to act together) and Reason to work together (the shared purpose). In small groups, the members read the configuration their dots form. From it they infer the type of result their way of working together produces. They then choose actions to strengthen the weak axes. Professor Brenda Zimmerman designed the STAR model. Henri Lipmanowicz and Keith McCandless adapted it into one of their Liberating Structures and describe it in The Surprising Power of Liberating Structures (2013).

Goal

With Generative Relationships STAR, a team diagnoses its own working relationships and then chooses the actions that strengthen them. According to the book, the members decide together on the follow-up "without intermediaries".

In 1996, David Lane and Robert Maxfield used the term "generative relationships" for relationships that produce sources of value nobody could foresee. In 1999, Brenda Zimmerman and Bryan Hayday presented the STAR model. In it, each letter stands for one dimension of a potentially generative relationship. Getting along well is therefore not enough: the book gives "we get along well but accomplish little" as an example of a result.

According to the book, the structure helps a team perform better and manage itself. It sharpens the purpose and identity of the group. It helps people stop looking for culprits and start understanding the group's patterns of interaction. It combines "diagnosis and treatment" without separating the planners from the doers. It reduces the frustration of members who are disappointed by how the team works and by what it achieves. The authors' site adds that it heals damaged relationships. It links the structure to the Liberating Structures principle "Practice Self-Discovery Within a Group".

The session produces a written record: the shared compass of each small group and the result each group reads in its configuration. It also produces the list of first actions with who carries them out and by when.

In our view, the structure serves the business analyst when a project team is stuck with no visible technical cause. The configuration of the dots shows whether the blockage lies in listening, in the opportunities to act or in the shared purpose. The analyst then proposes the structure that strengthens the weak axis instead of writing a diagnosis the team did not make.

You'll find the other structures in the repertoire under Liberating Structures.

Usage

When to use it

  • Team whose results disappoint: the group links its results to its patterns of interaction.
  • Criticism aimed at individuals: the group moves from looking for culprits to its patterns of interaction.
  • Strategy retreat: the group examines how it works together and what it achieves.
  • New team to be formed: the group settles its composition and its purpose.
  • Organisation changing its culture: the structure helps it move towards more inclusion, engagement or grassroots innovation.
  • Damaged relationship between two people: the compass for two helps mend the relationship.
  • Team follow-up: the group repeats the compass later and assesses its progress.
  • Mixed business and IT team: in our view, the Separateness axis shows whether the team draws out its different kinds of expertise.
  • Remote team: the structure works online with no major adjustments.

When not to use it

  • Group where rating the team can backfire on the rater: an announced restructuring or an internal investigation under way. Start with 1-2-4-All on a work question.
  • Blockage caused outside the relationships: a recurring tool failure or a late supplier. Use root cause analysis instead.

Description

The STAR compass

The STAR compass crosses four axes at its centre. Each axis carries one attribute of the relationship. It runs from "low" at the centre to "high" at the rim. In the book and on the site's template, Separateness sits at the top, Tuning on the right, Action at the bottom and Reason to work together on the left.

The four initials spell STAR in English only. Translations therefore keep the English labels.

AxisAttributeDefinition in the bookQuestion to the participant
SSeparateness: diversityThe amount of diversity in perspective, expertise, and background among group membersHow diverse are we as a group? Do we draw out our diverse perspectives among members?
TTuning: mutual listeningThe level of listening deeply, reflecting, and making sense of challenges togetherHow well are we in tune with one another?
AAction: acting togetherThe number of opportunities to act on ideas or innovate with group membersHow much do we act together?
RReason to work together: the shared purposeThe benefits that are gained from working togetherHow important is it that we work together? How clear is our purpose?
The four attributes of the STAR compass. The definitions come from the book. The questions come from the book and the site.

The invitation, the room and the roles

The facilitator invites the group to assess its team on the four attributes and then to work together on actions that make its relationships more generative. The site offers an invitation to read aloud: "Together, we'll delve into the dynamics of our team's relationships using the STAR Compass. By understanding what makes our interactions more or less generative, we can unlock hidden potential and build stronger connections."

Each participant receives a compass template and a pen. Each small group has a compass drawn on a flip-chart sheet. One more compass serves the whole group. The site adds an open wall and space to move around.

The whole team takes part and everyone has the same opportunity to contribute. The site calls the roles host and participants.

In our view, an analyst who belongs to the team places their dots like the other members. A colleague from outside the team then facilitates the session. Otherwise the analyst only keeps time and asks the plenary question.

The sequence

  1. The facilitator reads the invitation and hands out the compass template. 1 min.
  2. Each person assesses the team alone with the help of the questions and places one dot on each axis of their template. 4 min.
  3. Participants form groups of four to seven. Each person copies their dots onto the group's shared compass. The group discusses the placements that look alike and those that differ. 5 min.
  4. Each group reads its way of working together from the dots. It names and records the type of result this way of working produces. 5 min.
  5. Each group looks for actions to strengthen the weak axes and records them. 5 min.
  6. In plenary, the groups compare their assessments and their actions. The facilitator asks a few participants which first actions make sense for the group. 5 min.
Whole group
Alone
Groups of 4 to 7
1the template in handWhole group1 min
2one dot per axisAlone4 min
3the shared compassGroups of 4 to 75 min
4the result namedGroups of 4 to 75 min
5the proposed actionsGroups of 4 to 75 min
6the first actions: who, what, whenWhole group5 min
The sequence of Generative Relationships STAR according to the site. Each person rates the team alone. The small groups pool the dots, name the result and look for actions. The whole group picks the first actions.

The 2013 book follows the same order in 25 minutes.

ItemAuthors' siteBook (2013)
Introduction1 min: the invitation and the templateNone
Individual assessment4 min: one dot per axis on one's own template5 min: each person places the team on the four attributes
Small groupsFour to sevenTables of four
Sharing5 min: the dots on the shared compass, then the similarities and the differences5 min: the dots on the shared compass, then the exchange in pairs and in fours (1-2-4) looking for "consensus and differences"
Type of result5 min5 min
Actions5 min5 min
Plenary5 min: compare the assessments and the actions, then a few participants say which first actions make sense5 min: gather the actions and decide "What first steps can we take right now?"
Calculated total25 min25 min
The sequence of Generative Relationships STAR according to the authors' site and according to the 2013 book. The totals are our calculation.

The book advises moving from individual reflection to pairs, then to table conversations. In our view, a first session follows the site and adds this exchange in pairs to step 3.

Reading the configuration

The group reads the four axes together. The dots show how the team works. From them the group infers the type of result this way of working produces. The sources give three configurations as examples.

S Separateness
T Tuning
A Action
R Reason to work together
none
High Tuning + No Action
"We get along well but accomplish little"
S Separateness
T Tuning
A Action
R Reason to work together
High Action + Low Tuning
"Routine results with no innovation"
S Separateness
T Tuning
A Action
R Reason to work together
High Tuning, Separateness and Action + Low Reason
"Many false starts"
The book and the site give these three configurations and their results. Each dot sums up the position of the group on the axis. A grey axis lies outside the configuration.

The book also shows the compass of typical conference participants: high Separateness, high Tuning (a lot of talk), little Action and a fuzzy Reason. According to its caption, these participants would make their relationships more generative if they clarified their purpose and drew up an action plan.

In our view, the group first discusses any axis where the dots spread from the centre to the rim. On that axis the members do not experience the same team. An average would hide that gap.

From weak axes to actions

In our view, a plenary with several groups sees several configurations. It works first on an axis that is weak in every group. When an axis is weak in a single group, that group discusses it.

Both sources require everyone to know who does what by when. The site asks for at least one action. The book asks for at least one per participant. It encourages the members to research, organise and act on their own remedies.

To strengthen a weak axis, the authors propose following on with other Liberating Structures. The site has revised the book's list.

Weak axisAuthors' siteBook (2013)
SeparatenessConversation Café
What, So What, Now What?
Conversation Café
Shift & Share
What, So What, Now What?
TuningTroika Consulting
Heard, Seen, Respected
Wise Crowds
Troika Consulting
Agreement-Certainty Matrix
Heard, Seen, Respected
Action15% Solutions
Options Place
Min Specs
25/10 Crowd Sourcing
15% Solutions
Open Space
Min Specs
Reason to work together9 Whys
Drawing Together
Nine Whys
What I Need From You
The Liberating Structures the authors propose to strengthen each weak axis, according to the site and according to the 2013 book.

Online

On the site, the small groups work in breakout rooms. Each person draws their own compass. A visual collaboration space holds the shared compasses. A tech host manages the rooms. In plenary, the site proposes a Chatterfall: each person types their answer in the chat and sends it on the facilitator's signal.

In the book, participants place their dots on a shared whiteboard. They discuss them in pairs, then with the whole group. The book suggests giving one person a "synthesizer" role to keep the session moving. A chat version of 1-2-All produces the actions.

Variation: the compass for two

The sources propose the compass for two people. In our view, each person places one dot per axis for the relationship. The two compare the axes where their dots differ, then each names one action.

Variation: the personal compass

The site applies the compass to a single person with revised questions: "What do you bring to the group that might be different? To what degree do you reflect on your personal experience? Are you creating self-development opportunities to act on insights? To what degree are you clear on the purpose of your work?"

Strings

The book proposes a string for a group whose relationships hold back its work. Impromptu Networking starts the exchanges and creates a few connections. Generative Relationships STAR then diagnoses the patterns of interaction. The group identifies what each person and the formal leaders can do. 1-2-4-All finally produces the next steps.

In its description of Heard, Seen, Respected, the book groups Generative Relationships STAR with Troika Consulting, Helping Heuristics, Appreciative Interviews and Conversation Café among the structures that help mend a relationship. In chapter 4, the book names it with Ecocycle and What I Need From You among the structures designed to diagnose relationship issues.

What makes the structure fail

Both sources warn against judging where people place their dots. In our view, a participant who hears that their dot is wrong moves it or falls silent. The facilitator treats each dot as a perception to understand.

The sources present the compass as a conversation aid without a scale. A team that compares averages from one session to the next treats a conversation as a survey. In our view, the team compares the configuration and the gaps between the dots.

A group that blames a weak axis on a colleague falls back into looking for culprits. The facilitator then brings the discussion back to the patterns of interaction the four axes show.

A management team that collects the compasses and picks the remedies on its own separates the planners from the doers again. The members who made the diagnosis choose the actions.

In our view, a line manager in the room pulls the dots towards the rim. The manager places their dots like the other members and speaks last in the plenary.

AI considerations

Neither the book nor the site discusses AI.

The team places its own dots and reads its own configuration. The site links the structure to the principle "Practice Self-Discovery Within a Group": the group discovers its patterns of interaction for itself. A model that delivers the diagnosis brings back the intermediary.

Before the session, a model can translate the invitation and the four questions for a bilingual team. The facilitator checks that the translation keeps the meaning of each axis. German "Abstimmung" for "Tuning" drifts towards the coordination of tasks. "Gegenseitiges Zuhören" keeps the listening.

After the session, a large group can hand a model the photos of the compasses and the action lists. With eight groups of five, the model transcribes the forty dots on each axis and groups similar actions. The facilitator checks the result against the sheets. One misplaced dot changes the configuration.

The action lists name people. They go only to an AI service that the organisation has approved for personal data.

Examples

The shared compass of a group of five shows the configuration the sources associate with the result "we get along well but accomplish little".

S Separatenesshigh
T Tuninghigh
A Actionhigh
R Reason to work togetherhigh
low
What the team gets
We get along well but accomplish little
Axis to strengthen: Action
Constructed case. The shared compass of the team behind the benefits portal of an AHV/AVS compensation office. Each of the five members placed one dot on each axis. In this configuration the group reads the result "We get along well but accomplish little".

The Separateness dots spread from the centre to the rim. The group discussed them before naming the result. It picks the Action axis and follows on with 15% Solutions. Each member notes one action they can carry out with no extra budget or authorisation.

WhoWhatWhen
Analyst 1Show the mock-up of the application screen to two front-desk staffTuesday
Analyst 2Write the acceptance criteria of the first user story with a developerWednesday
Developer 1Deploy the home page to the test environmentThursday
Developer 2Deliver a first screen of the application form to the business ownerThursday
Business ownerSettle the three open questions on the application form aloneFriday
Constructed case. The actions chosen by the team behind the benefits portal of an AHV/AVS compensation office. The book asks for one action per participant.

All the actions target the Action axis. None targets the Tuning axis, which is already strong.

Visualisations

One diagram shows the shared compass of a team and the result it reads in its configuration. A second compares the three configurations the sources give. A third shows who works together at each step of the sequence, for how long and what each step hands on to the next. The attributes, the two versions of the sequence, the strings and the chosen actions fit in rows and columns.

Cost

PhaseLevelJustification
PreparationLowPrint one compass template per participant. Draw one compass per small group and one for the plenary.
ExecutionMedium25 minutes according to both sources. In our view, the facilitator must know how to bring a group that is looking for a culprit back to the configuration of the dots.
DocumentationLowA photo of each shared compass and the list of actions with who, what and when.

Tooling

A printed compass template per participant, pens and flip-chart sheets for the shared compasses are enough. Sticky dots of a single colour replace the dots drawn on the shared compass.

Remotely, Miro or Mural hold the compasses. Zoom or Microsoft Teams provide the breakout rooms. The Chatterfall runs in the video call's chat.

Sources

Brenda Zimmerman designed the STAR model. Henri Lipmanowicz and Keith McCandless adapted it into a structure. The book and the site both say so. The definitions of the four attributes come from the writings of Lipmanowicz and McCandless.

  • Henri Lipmanowicz, Keith McCandless, The Surprising Power of Liberating Structures: Simple Rules to Unleash a Culture of Innovation, 2013, ISBN 978-0-615-89337-2: the description of Generative Relationships STAR with its attributes, steps, purposes, tips, variations, examples and attribution (Field Guide, part four); the string with Impromptu Networking and 1-2-4-All (chapter 6); the mention in chapter 4 among the diagnostic structures; the references from Heard, Seen, Respected.
  • Generative Relationships STAR, Liberating Structures: the site's version with its invitation, timings, roles, online use, personal variation, strings and the principle "Practice Self-Discovery Within a Group".
  • Brenda Zimmerman, Bryan Hayday, "A Board's Journey into Complexity Science: Lessons from (and for) Staff and Board Members", Group Decision and Negotiation 8(4), 1999, pp. 281-303: the presentation of the STAR model, in which each letter stands for one dimension of a potentially generative relationship.
  • David Lane, Robert Maxfield, "Strategy under complexity: Fostering generative relationships", Long Range Planning 29(2), 1996, pp. 215-231, from the Plexus Institute summary: the term "generative relationships" and its definition.
Gemba Walk
All techniques
Glossary