The Barnacle Problem
Martin’s explanation for why strategies fail in organisations that did everything right: management systems accumulate like barnacles on a hull.
Three mechanisms do the damage. Many designers, narrow designs. Finance builds a budgeting system, HR a review cycle, IT a security protocol, each sensible alone; together they signal conflicting priorities to the same people, who resolve the conflict by ignoring the newest signal. Layering, never uninstalling. New strategies add new systems; old systems are almost never removed, and they keep doing their job, motivating the old behaviour, beneath the new direction. Invisible drift. Each deviation is small; the alarm only sounds when the company is far off course, when gentle correction is no longer possible.
The crucial insight
Systems motivate behaviour whether or not they were designed to. Nobody decided the expense policy should undermine the innovation strategy; it just does, quietly, every time someone declines to risk a rejected claim. A strategy laid over barnacled systems produces beautiful documents nobody acts on, and the organisation is then diagnosed with "execution problems", which is like diagnosing a ship with laziness.
The remedies
Two, and they are ordered. Audit before cascade: examine what the current systems actually reward before writing the new strategy, floor before house. And appoint a single accountable integrator, which is the argument of CEO as Chief System Designer: coherence across systems cannot be delegated because nobody else spans them. Between audits, the maintenance discipline is honing, not transformation, the hull cleaned continuously or dry-docked catastrophically. And when system signals conflict in the meantime, leader behaviour is the signal that wins.
How this connects to OKRs
OKR rollouts meet the barnacles head-on: install a goal system that rewards outcome thinking on top of a bonus plan that pays for activity, and the bonus plan wins without a fight. This is why our implementations begin with a systems audit rather than a template, and why the question we ask first is not "what are your goals" but "what does your machinery currently pay people to do". The answer is usually the strategy the company has been executing all along.
One line to keep: Before you ask what your goals are, ask what your machinery currently pays people to do.
Our synthesis of Roger Martin’s published work, sources credited. Read the originals: they’re excellent.
What is the barnacle problem? +
Old systems, policies and incentives accrete like barnacles on a hull and quietly motivate the behaviour the new strategy is trying to change. Systems motivate whether or not they were designed to, so a strategy laid over barnacled machinery produces documents nobody acts on.
Why do OKR rollouts hit the barnacles first? +
Install a goal system that rewards outcomes on top of a bonus plan that pays for activity, and the bonus plan wins without a fight. That is why our implementations begin with a systems audit, not a template: what the machinery pays for is the real strategy.
Reading about method is not the same as running it. We install this system and build the capability that stays.
Turn strategy into executed strategy →