Co-creating Safe-to-Fail Experiments
Use shared stories to identify problems and possible changes, then test several small interventions in parallel. Watch what happens, expand changes that help, and stop or modify those that cause harm.
THE IDEA
Work with stories to co-create small experiments for change
Asking people to share stories and compare their interpretations can reveal what matters to them and where they see room for change. In the Cynefin framework’s Complex domain, these accounts help a group choose probes whose effects it can observe over time. Running several small, safe-to-fail experiments in parallel lets the group explore different possibilities instead of committing to a single change program. The approach also considers the Hawthorne Effect: attention to an intervention may itself affect people’s behavior. (7)
A housing-department probe
A local-authority housing department used story-sharing sessions to discuss what helped or hindered its work. Late rent payments created time-consuming follow-up, including face-to-face encounters with upset, stressed, and sometimes abusive tenants. Staff proposed offering tenants one or two rent-free weeks, perhaps at Christmas, if they had paid on time for a specified number of months.
The proposal was to test the incentive in a small area, expand it if it worked, and scrap it if it did not. The group discussed what success and failure would look like and how it would decide whether to expand or stop the experiment. Its hope was to reduce both late payments and the distress involved in chasing them. The excerpt does not report a pilot result.
Story-based sessions can also use deliberately open instructions to help people explore possibilities beyond the usual agenda. The authors call this a shallow dive into chaos, followed by a return to the Complex domain. Participants compare what they notice in different parts of the organization and develop ideas for probes. They may initially find the lack of a defined answer disconcerting.
The probe–sense–respond approach expands changes that produce desirable effects and dampens those that do not. Rather than reporting a move from a score of 4.5 to 4.9 without understanding it, a group can ask which experiences it wants to make more common and which it wants to reduce. Stories provide context for those judgments; the experiments test whether a proposed change actually removes a hindrance.
When to use it
Improving employee morale by allowing small workflow adjustments and testing their impact.
Reducing late payments in a housing department through a targeted incentive probe.
Encouraging innovation by giving ambiguous instructions and observing emergent ideas.
Evolving organizational culture by amplifying positive stories and dampening negative patterns.
FROM THEORY TO EVERYDAY LIFE
In practice
Housing Department Rent Payment Probe
A local authority housing department faced chronic late rent payments and time-consuming follow-up. After story-sharing sessions, they designed a small experiment: offering one or two weeks rent-free (e.g., at Christmas) to tenants who paid on time for a set number of months. The group proposed a small pilot, with expansion to other areas if it succeeded and abandonment if it failed. This safe-to-fail probe aimed to reduce stress for employees and improve payment behavior.
A MOMENT TO REFLECT
Put it to work.
What stories are circulating in your organization that reveal the current disposition for change?
How can you design a small safe-to-fail experiment to test a potential improvement in your team?
What would success or failure look like for your proposed experiment, and how would you amplify or dampen it?
CONTINUE THE CONVERSATION
Source texts
- Cynefin - Weaving Sense-Making into the Fabric of Our World ED02 (Dave Snowden Sonja Blignaut Zhen Goh)