What your practices actually say
Your values statement is a wish. Your collaboration practices are a diagnosis. Score nine practices against five axes and two organizational cultures reveal themselves.
What your practices actually say
Your values statement is a wish. Your collaboration practices are a diagnosis.
Every organization publishes values. Every organization has practices. The values say what the org aspires to. The practices say what the org actually does when nobody’s coaching.
If you want to know whether your culture supports real agility, stop reading the wiki. Start reading the practices.
Nine practices, five axes
Take nine common collaboration practices. Some come as spectra — you’re already choosing between poles. Score each one against five shared axes:
- Predictability assumed. Does the practice require knowing the answer in advance?
- Feedback speed. How quickly does the practice surface whether it’s working?
- Local discretion. Can people at the edge adjust without asking permission?
- Variety tolerance. Does the practice accept different approaches, or flatten them?
- Evolution capacity. Can the practice change when conditions change?
Rate each practice high, medium, or low. Don’t explain the scores. Let the pattern speak.
| Practice | Predictability | Feedback | Local discretion | Variety tolerance | Evolution |
|---|---|---|---|---|---|
| Fixed roles & responsibilities | High | Low | Low | Low | Low |
| Experimentation | Low | High | High | High | High |
| Evolving practices | Low | High | High | Medium | High |
| Standardization | High | Medium | Medium | Low | Low |
| Centralization | High | Low | Low | Low | Low |
| Avoiding duplication | High | Medium | Medium | Low | Low |
| Accepting duplication | Low | High | High | High | High |
| Sharing | Low | High | Medium | Medium | Medium |
| Formal committees | High | Low | Low | Low | Low |
| Informal communities | Low | High | High | High | High |
| Fixed methods | High | Low | Low | Low | Low |
Look at the table. Two clusters emerge without any prompting.
Practices cluster into two patterns before you name them. If you can score them honestly, the culture reveals itself.
Two clusters, two orgs
Cluster A: control. Fixed roles. Standardization. Centralization. Avoiding duplication. Formal committees. Fixed methods.
This org assumes predictability. It moves slowly. It grants low discretion at the edge. It doesn’t tolerate variation. Its practices don’t evolve.
It’s optimized for compliance. Not adaptation.
Cluster B: adaptive. Experimentation. Evolving practices. Accepting duplication. Sharing. Informal communities.
This org doesn’t assume predictability. It generates fast feedback. It grants high discretion. It tolerates variation. Its practices evolve.
It’s optimized for learning. Not control.
This isn’t a moral judgement. Control patterns are rational when the domain is simple or complicated. When you can predict outcomes, standardization is efficient. When you can’t, it’s a tax.
The failure is mismatch. An adaptive pattern in a simple domain wastes energy. A control pattern in a complex domain kills adaptation.
The question isn’t “should we standardize?” It’s “where should we standardize?” Standardize values and let practices evolve — that’s adaptive. Standardize practices and pretend values don’t need work — that’s control.
The real test
Most organizations claim Cluster B. Almost all default to Cluster A when a deadline looms or an incident hits.
The diagnostic isn’t what’s on your wiki. It’s what happens at 2am.
Do people experiment or escalate? Do communities self-organize or do committees convene? Does someone try something, or does someone ask for permission?
When something goes wrong, does your organization’s first move open up options or close them down?
That’s your culture. Not the one you wrote down. The one you reach for under stress.
Malcolm Bastien
Enterprise Agile Coach
Enterprise Agile Coach helping leaders in Fintech, Banking, and HealthTech build fast flow through systems thinking and aligned autonomy, empowering teams to deliver high-performance results at scale.