A friction statement names where execution may slow and the proposed mechanism, so the call can be tested rather than merely believed.
Why it matters when the plan changes
A red rating earns attention and says nothing about what to do. Without a stated mechanism, the natural explanation for slow execution is the person nearest to it, and research on the correspondence bias shows how readily that happens: Daniel Gilbert and Patrick Malone describe inferences about a person drawn from behaviour the situation fully explains, and set out 4 mechanisms that produce them.
The tension is that a mechanism is a commitment. Writing down why execution may slow exposes the call to being wrong, which is uncomfortable for the author and useful for everyone else. In a 2-year forecasting tournament, Barbara Mellers and colleagues found that training, teaming and tracking forecasters improved both calibration and resolution. A stated mechanism is what makes that kind of scoring possible for an organisational call. Without it, the review can only argue about impressions.
In practice
A product launch now depends on a data platform team that reports to a different executive. The friction statement reads: the launch may slip because release decisions need sign-off from a team whose own targets do not include the launch, and no forum resolves conflicts between the two roadmaps. Counter-evidence: the two leads already meet weekly.
Evidence
People tend to infer a person's lasting dispositions from behaviour that the situation fully explains, a pattern produced by several distinct mechanisms.
Daniel T. Gilbert and Patrick S. Malone, The Correspondence Bias, Psychological Bulletin (1995)In a 2-year forecasting tournament, training, teaming and tracking forecasters improved both calibration and resolution.
Barbara Mellers and colleagues, Psychological Strategies for Winning a Geopolitical Forecasting Tournament, Psychological Science (2014)
What it cannot tell you
A friction statement is a hypothesis about a mechanism, not a finding. It can name the wrong interface, and a plausible mechanism can crowd out a better one, which is why it travels with counter-evidence and a named test rather than standing alone.
How Atlas reads it
Friction, where and why, is one part of every forecast and reforecast, alongside the strategic shift, the evidence for and against, what is unknown, the confidence, the owner and what will test the call. The evidence always includes what argues against the call. Every rating is a forecast with a deadline, a confidence and an owner, and the friction statement is the part that says what the rating is about.
Questions
It names a specific interface and a mechanism that could be wrong. A claim that the team is struggling cannot be tested. A claim that release decisions stall because sign-off sits with a team whose targets exclude the launch can be: change the sign-off and see whether decisions speed up.
Because without one, the explanation defaults to the person nearest the symptom. Daniel Gilbert and Patrick Malone's 1995 review of the correspondence bias describes 4 mechanisms by which people attribute to character what the situation explains. A stated mechanism about the arrangement gives the leader an alternative to test first.
They travel together. A friction statement without the observations that point the other way invites false certainty. Recording both lets the reader weigh the call, and lets the later review show whether the mechanism or the counter-evidence turned out to describe what actually happened.
It can involve a person, but it is written about the arrangement: what the role asks, which conditions change the judgement, where authority or support falls short. The rating belongs to the execution area, not to the leader, and the person remains more than the evidence used in the call.
Against the outcome the forecast named, on the date it named. Barbara Mellers and colleagues found in a 2-year forecasting tournament that tracking forecasters improved calibration and resolution, and a friction statement has to be written so that it can be scored in the same way, or it cannot improve.