The Consensus Trap

Unprincipled consensus-culture is a sign of organizational decay

Mike Cvet
3 min readMay 2, 2024
Photo by Javier Allegue Barros on Unsplash

“We’re big on consensus-based decision-making”

Is often followed by something like

“One of our biggest challenges is with ownership and accountability”

Alignment is the process by which we agree on outcomes, understand the implications of decisions, and organize. Consensus is the process by which we ensure outcomes, decisions, and paths forward are unobjectionable to all parties.

Alignment can organize people around contentious ideas or proposals, embracing healthy tension. Consensus modifies or moderates proposals, and convinces detractors — but it can take time, especially when the group is large.

In teams with weak accountability culture, consensus-based culture can lead to slow decision-making, ambiguous responsibility for outcomes, poor ownership and external attribution for failure. It’s common when accountability principles are not sufficiently communicated and upheld by the leadership team.

It can be a kind of well-intentioned cultural degradation caused by many problems; but particularly from a decay of team trust, lack of clarity and accountability, and roles with ambiguous authority. At its worst, it’s a cynical strategy to diffuse responsibility.

We can, and should, broadly solicit input from stakeholders, experts or teammates when deciding on a course of action or solution. But it’s a mistake to believe that the only correct solution is the one everyone is happy with. We can think about risk many ways, and one way risk manifests is by simply through a lack of expediency, ambition or data-gathering.

Within a team, and within an organization, a few steps can be taken to ensure that progress can be steadily made, without getting trapped in either a consensus cycle or analysis paralysis:

  • Role Clarity and Credibility: Define DRIs for solving problems, empower them to do so, and identify escalation paths for making decisions when outside of the DRI’s scope.
  • Connect Issues to Strategy or the Bigger Picture: Make any given proposal or plan feel less arbitrary by articulating how it supports the longer-term goals of the team or organization.
  • Proportionality: In the software world, many mistakes or failures can be rolled back, re-implemented, or re-designed without too much risk or effort. The amount of alignment or consensus sought should be proportional to the relative risk (see this guidance applied to technical design documents as well).
  • Disagree and Commit: Normalize the expectation that people can, and should, commit to a given solution, even if they disagree with the approach. This requires trust in that the owner will be held accountable for outcomes, positive or negative.
  • Expect Learning as an Outcome: Even if the business outcome isn’t reached, there’s value in learning about why an approach failed in the context in which it was executed. This better informs the next attempt — just make sure the lessons are documented, shared, and internalized, rather than buried and forgotten.
Modeling these concepts across a couple of axes illustrates some of the tradeoffs associated with different combinations of cultural properties. Comparing bottom-left and top-right quadrants: for the same risk profile, faster decision-making is probably preferable, but it depends on the context

Reaching consensus with a set of experts or leaders can be critical when facing a decision which is either irreversible or extremely risky.

For example, when defining a public terms of service, it probably makes sense to get to a place where Finance, Legal, Product, Engineering, and Comms agree on the responsibilities and expectations of service providers and customers. Remember the Zoom ToS fiasco?

The amount of effort put into reaching consensus towards should generally follow similar best practices around development of design documents or RFCs:

The amount of time and effort involved should be proportional to the likelihood of improved outcomes [or mitigation of negative ones]

--

--

Mike Cvet

I’m a former Distinguished Engineer at LinkedIn and Twitter, was an early engineer at a couple startups with successful exits, and hacked around at Red Hat