In Fortune on 19 May 2026, Julia Dhar (managing director and partner at Boston Consulting Group, founder of BCG's Behavioral Science Lab) put a long-running puzzle on a behavioural footing. Around 70% of large reorganisations fail, and the figure has not improved in decades. What Dhar and her co-authors found, after surveying 6,000 executives and employees across 15 countries, interviewing more than 50 leaders and behavioural scientists, and reviewing 50 years of evidence, is that the failure is not about strategy at all. It is about a quiet, predictable misreading of the room.
The headline number is the unsettling one. Around 70% of executives report feeling positive about a change they know nothing about. The same change, presented to the people who will have to live with it, lands as anxiety, fatigue and frustration. Dhar's team calls it by its proper name: the false consensus effect.
Why It Is A Behaviour Number, Not A Comms Number
The familiar response to a wobbling transformation is to call it a communication failure. More town halls, more decks, a clearer narrative. Dhar's data points the other way. Communication is downstream of the false consensus, not upstream. Leaders communicate confidently because they feel confident. They feel confident because the people closest to them are nodding. The people nodding are nodding because they are also senior, also informed, also part of the design. The further the design travels from that room, the more the certainty curdles into anxiety in the receivers.
McKinsey's transformation research, repeated across two decades, sits next to this neatly. Around 70% of large change programmes miss their stated goals. It is the same 70%. The new contribution from BCG is naming the mechanism rather than the symptom. A senior layer reads its own positivity as a forecast for everyone else, designs the rollout on that assumption, then learns from the dashboard that the assumption was wrong.
The Dunning-Kruger pattern in self-rated communication ability tells the same story from another angle: confident self-report from leaders, mediocre observed performance under stress, and the gap visible only to the people on the receiving end. In Sidestream's own academic behaviour-change work, building on research from UCL, Cambridge and Bocconi, immersive role-play was roughly 20% more effective than passive modalities at building communication skill, and self-rated confidence did not predict observed performance. We design the self-report out by measuring the behaviour.
What Most Organisations Try
The default playbook is two parts: communicate harder, and survey more. A refreshed town-hall slide, a values poster, a fortnightly pulse with sentiment dashboards. Sometimes a 360-degree feedback round for the senior team, with anonymous comments tagged to behaviours like "listens", "challenges constructively", "follows through". The feedback is shared, often skimmed, mostly forgotten by the time the next decision lands.
The mechanism by which this fails is well established. Roediger and Karpicke, in Psychological Science (2006), showed that being tested on material lifts long-term retention by roughly 50% compared with re-reading. Most senior-team development is the re-reading kind. The reading happens in a calm room. The decisions happen in stressed ones, twenty minutes between meetings, and the brain falls back to the move it already knew. Anders Ericsson's Peak (2016) is even sharper on this: the first-reaction behaviours that close a false-consensus gap, asking before telling, naming the concern nobody else will name, holding silence long enough for the room to fill it, change only with deliberate, feedback-rich practice against a clear behavioural target. They do not change with insight.
What Actually Works
Dhar's prescription is to treat employees as the customers of the change. To do that, the senior team has to first know what the customers actually feel, and second, behave differently when they are talking to them. Both of those are rehearsal questions, not deck questions.
The format that closes the false-consensus gap is a small, mixed group. One or two professional actors playing a stretched director, a sceptical engineer, a quietly furious shift lead. Three or four full conversations, replayed with feedback against named behavioural anchors. By the third rehearsal the leader's first move is the right one, not the rescued one. The transformation pulse looks different six weeks later because the rooms looked different in week one.
What We Do About It
Our immersive simulations and senior-leader workshops are designed for exactly the rooms where the false consensus quietly sets in: the all-hands rehearsals, the layer-two briefings, the conversations a senior team thinks went well and the receiving teams remember as something else. Read also our piece on the HBR six-week drop for the people-skills gap angle, or our piece on the 70% transformation pattern for the broader McKinsey frame.
Transformations fail when the room at the top mistakes its own certainty for everyone else's. Rehearse the conversation the day before, not the week after.
The transformations that hold their numbers past week six are not the ones with the cleverest reorg. They are the ones where the senior team has rehearsed the awkward briefing often enough that the default move is to ask, not announce. Book a call to look at what that rehearsal would look like for your top team.
Book a free 30-minute diagnostic call → or read about our research-backed approach.