The performance frameworks that actually change behaviour: Clear, Newport, Collins and the evidence behind the habits

Performance research keeps splitting into two unhelpful extremes: motivational content that feels good and doesn't change behaviour, and academic research too abstract to apply on a Tuesday. The writers in this collection built careers bridging that gap, with documented methods specific enough to test. James Clear reframed habit change as a problem of environment design rather than willpower. Cal Newport built an entire counter-discipline around protecting the kind of attention a fragmented workday destroys. Jim Collins ran a multi-year comparative research project to find out what actually separated companies that sustained exceptional results from similar companies that didn't. This guide takes the collection's signature frameworks in turn, with the study or documented mechanism behind each, before addressing where they apply to teams as well as individuals.

The common thread worth naming up front: every framework below trades a comfortable, motivation-first story for a less flattering, structure-first one. None of them promise that trying harder is the answer, which is precisely why they've held up better under scrutiny than most of the genre they get shelved alongside.

Identity-based habits

Clear's framework, developed across Atomic Habits and years of prior writing, argues that durable behaviour change works by changing the identity a person holds about themselves, not by relying on willpower applied to a goal. A person who identifies as "someone who doesn't miss a workout" sustains the behaviour differently than someone pursuing "a goal to get fit," because each missed session becomes evidence against or for a self-concept rather than just a lapsed intention.

A worked example: rather than setting a goal to read more, Clear's framework builds a small, easy-to-repeat action, one page a day, tied to an existing habit, and frames each completion as reinforcing the identity "a reader," so that a day missed feels like an identity inconsistency worth correcting rather than simply a broken streak. The commonly repeated claim that habits form in 21 days doesn't hold up against the research Clear himself cites, which found actual automaticity timelines ranging from 18 to 254 days depending on the behaviour.

Where it fails: identity-based framing works well for habits with a clear, repeatable, low-friction action; it is a weaker tool for genuinely difficult, high-variance efforts (a major career change, a complex creative project) where no small daily action reliably represents the identity being built, and the framework's clean mechanics don't translate cleanly to goals that resist decomposition into a repeatable unit.

Deep work

Newport's framework, developed across his own academic career and subsequent writing, distinguishes deep work, cognitively demanding effort performed without switching or interruption, from shallow work, the logistical, low-cognitive-load tasks (email, routine meetings) that expand to fill available time because they're easier and produce visible responsiveness. His documented argument is that most valuable, differentiated output comes specifically from deep work, and most knowledge workers' calendars protect none of it.

A worked example: a writer who blocks two uninterrupted hours each morning, phone in another room, notifications disabled, produces measurably different output than the same writer working the same two hours with periodic message-checking, even though both sessions might subjectively feel like "working," because the switching cost of re-entering deep focus after each interruption is documented as substantially longer than the interruption itself.

Where it fails: deep work requires a job structure that permits extended uninterrupted blocks, which is a reasonable ask for research, writing or programming roles and a much harder fit for customer-facing or highly collaborative roles where responsiveness is the actual job requirement, not a distraction from it.

Compounding effort

The claim that small, consistent improvements outperform large, inconsistent ones over a long enough horizon runs through both Clear's habit research and Collins's organisational research, and the counterintuitive part, documented in Collins's Good to Great study, is how unglamorous the actual compounding period looks from inside it: years of unremarkable, consistent execution before results become visible enough to attract outside attention.

A worked example: Collins's research on companies that sustained exceptional stock performance found a consistent pattern of years of disciplined, incremental operational improvement preceding the visible transition point, rather than a single dramatic strategic pivot; the companies that looked like overnight successes from outside had documented years of unglamorous compounding that preceded the moment anyone noticed.

Where it fails: compounding effort requires the underlying direction to actually be correct; consistent small improvements applied to a fundamentally wrong strategy compound the wrong result just as reliably as they compound a right one, and neither Clear's nor Collins's framework specifies how to verify the direction is correct before committing years to compounding it.

Level 5 leadership

Collins's research, based on a multi-year comparative study contrasting sustained-performance companies against similar companies that didn't sustain it, found the CEOs at the sustained performers combined intense professional will, a fierce resolve to produce results, with personal humility, a documented tendency to credit others and external factors for success while taking personal responsibility for failure. This combination directly contradicted the louder, more charismatic leadership profile business media has historically celebrated.

A worked example: Collins's research describes leaders who, in interviews conducted for the study, consistently redirected credit for company success toward their teams, board, or luck, while taking personal ownership of setbacks, a documented pattern the research found correlated with sustained rather than short-term performance, unlike the comparison companies whose leaders more often did the reverse.

Where it fails: Level 5 leadership as a description is easier to identify in retrospect from interview transcripts than to install deliberately in a leader who doesn't naturally combine both traits, and the framework offers limited guidance on how to develop genuine humility in a leader whose instincts run the other way, as opposed to simply performing humility for the benefit of the same study.

Start with why

Simon Sinek's framework, developed across his own writing and widely viewed public talks, argues that organisations and leaders who communicate their underlying purpose first, before describing what they do or how they do it, build more durable trust and commitment than those who lead with product features or process detail. His documented claim is that people are moved to action more reliably by a stated reason than by a stated method, and that most communication gets the order backwards.

A worked example: two companies selling comparable products, one leading its marketing with a stated purpose ("we exist to make technology accessible to people who've been excluded from it") and one leading with feature specifications, are documented in Sinek's own case studies as producing measurably different customer loyalty, on the argument that the purpose-led framing gives customers a reason to stay that a features comparison alone doesn't provide.

Where it fails: a genuinely compelling stated purpose only builds durable trust if the organisation's actual behaviour is consistent with it over time; Sinek's own framework doesn't protect against an organisation stating an inspiring purpose it doesn't structurally live up to, a mismatch that tends to produce more cynicism than no stated purpose would have, once customers or employees notice the gap.

Vulnerability and original thinking

Grant's documented research on productive disagreement found that teams which surface and work through conflict on substance, rather than maintaining surface harmony, tend to reach better decisions, because unaddressed disagreement typically resurfaces later at a more costly point rather than actually disappearing. This connects to a broader theme across leadership research in this category: that admitting uncertainty or disagreement openly, rather than performing confidence, tends to build more durable trust and better decisions than the alternative.

A worked example: a team that surfaces a disagreement about a strategic direction directly in a planning meeting, with Grant's framework encouraging structured devil's-advocate roles to make the disagreement explicit rather than avoided, reaches a more stress-tested decision than a team where the same disagreement stays unspoken and resurfaces as quiet resistance during execution.

Where it fails: productive disagreement requires a team culture with enough psychological safety that surfacing conflict doesn't get personally punished, and installing Grant's framework in a team where raising disagreement has previously carried a real career cost requires fixing that underlying safety problem first, something the framework describes as a precondition rather than something it can install on its own.

Extreme ownership and discipline as freedom

Jocko Willink's framework, developed from his military leadership experience and subsequent writing and consulting, argues that a leader should treat every failure within their team as their own responsibility to own and correct, rather than distributing blame downward, on the documented claim that this posture is what actually earns a team's willingness to follow difficult orders under pressure. His companion principle, that discipline creates freedom rather than restricting it, holds that structured routine removes the daily decision fatigue that would otherwise consume the energy needed for harder, less routine decisions.

A worked example: a leader whose team missed a deadline, applying extreme ownership, opens a review by identifying what they personally failed to communicate, resource or check on, before examining any individual team member's specific errors, a sequencing Willink's framework argues changes how honestly the team subsequently surfaces its own mistakes, since blame isn't the first thing being distributed.

Where it fails: extreme ownership applied without limit can obscure genuine cases of individual negligence or bad-faith performance that do warrant direct accountability, and a leader who absorbs blame for every failure regardless of its actual source risks removing a useful signal about which team members need direct, individual correction rather than systemic leadership adjustment.

---

The Performance & Mindset category collects these frameworks, and others including Peter Drucker, Andy Grove and Oliver Burkeman, as downloadable .md files for Claude, ChatGPT or any LLM. Burkeman's own documented work on finitude, treating the limits of a finite lifespan as a premise for prioritisation rather than a problem to optimise away, offers a useful counterweight to the more optimisation-heavy frameworks above; it's worth reading precisely because it questions the productivity genre's own premise rather than adding another technique to it.