Anchoring Change
Anchoring Change is the practice of embedding new behaviors and structures so deeply into an organization that they survive leadership changes, market shifts, and time. Kotter's 8-step model identifies anchoring (step 8) as where most changes secretly fail — not at launch, but in the 12-24 months after launch when the original change leaders move on, attention shifts, and the org reverts to old patterns. Real anchoring requires four mechanisms: (1) Cultural alignment — the change is reflected in stories, language, and what's celebrated. (2) Hiring and promotion — new hires are selected for fit with the change, and promotions reward change-aligned behaviors. (3) Systems and structure — org charts, processes, and incentives are aligned. (4) Leadership succession — the next generation of leaders has been groomed in the new norms. Without all four, the change rents the org for 18-24 months and then leaves.
The Trap
The trap is declaring victory at go-live. Project teams, fundraising, and executive attention all peak around launch and then dissipate. The org assumes 'the change is done' when in fact the change has barely started — anchoring takes 18-36 months of sustained reinforcement. The second trap: anchoring through enforcement instead of embedding. Forcing the new behavior through KPIs and compliance mandates produces compliance during the enforcement period, but the moment enforcement relaxes, the org reverts. Real anchoring shifts identity ('this is who we are') not just behavior ('this is what we do').
What to Do
Build a 24-month anchoring plan BEFORE go-live. Define: (1) Cultural — what stories, rituals, and language will reinforce the change? (2) Talent — what hiring profile, promotion criteria, and onboarding curriculum will reflect the change? (3) Systems — what processes, incentives, and tools embed the change as the default? (4) Succession — who are the next-generation leaders being groomed? Assign explicit owners for each mechanism, with quarterly reviews. Run 'change drift' surveys at 6, 12, 18, and 24 months post-launch to detect reversion early. KnowMBA POV: most transformations spend 80% of effort on getting to launch and 20% on anchoring; the inverse ratio produces durable change.
Formula
In Practice
Lou Gerstner's transformation of IBM (1993-2002) is a master class in anchoring change. The visible change — pivoting from a hardware company to a services and integration company — was launched in 1993-94. But Gerstner spent the next 8 years anchoring it. He rewrote performance management to reward cross-divisional collaboration (anchoring teamwork over silos). He redesigned compensation so executives were paid on IBM stock performance, not divisional results (anchoring 'one IBM'). He systematically promoted leaders who embodied the new identity and managed out those who didn't. He invested in a multi-year cultural reset documented in his book Who Says Elephants Can't Dance? By the time Gerstner retired in 2002, the new IBM was self-sustaining — successor Sam Palmisano continued the trajectory rather than reversing it. IBM stock went from $13 in 1993 to $80+ by 2002. The change anchored because Gerstner spent more years anchoring than launching.
Pro Tips
- 01
The 18-month rule: if the change isn't visible in your hiring profile, performance reviews, and promotion decisions by month 18 post-launch, you're not anchoring — you're enforcing. Enforcement decays; anchoring persists.
- 02
Story matters as much as systems. The folk-tales an org tells about itself ('that time we shipped the impossible deadline,' 'how we beat the competitor') anchor identity. Deliberately cultivate stories that reinforce the new identity, and gracefully retire the old folk-tales that anchor the old identity.
- 03
Watch the language. When people stop using the old terminology unprompted (without being corrected), and when the new terminology shows up in places you didn't put it (org chart titles, casual conversations, candidate interview language), you're anchoring. When the old language persists 18 months post-launch, you're not.
Myth vs Reality
Myth
“Anchoring is the easy part — the hard work is the change itself”
Reality
Anchoring is HARDER than launching change because attention has moved on, urgency has faded, and the original change leaders are typically gone or distracted. Most changes that make it to launch fail in anchoring. Plan for it as a multi-year discipline.
Myth
“Once a behavior is mandatory, it will become habitual”
Reality
Mandates produce compliance, not habit. Habit requires repeated practice in conditions where the new behavior is the path of least resistance. If the new behavior is mandatory but harder than the old behavior, the moment compliance pressure relaxes, the org reverts.
Myth
“Cultural anchoring takes 6-12 months”
Reality
Cultural anchoring takes 24-60 months for major changes. The shorter timelines apply to small process changes. Major identity shifts (Microsoft's culture reset, IBM's services pivot, Adobe's subscription transition) all took 5+ years to fully anchor — and these were the well-executed cases.
Try it
Run the numbers.
Pressure-test the concept against your own knowledge — answer the challenge or try the live scenario.
Knowledge Check
Your org launched a major change 14 months ago. Adoption hit 75% at launch. Now: behaviors are drifting back (down to 52%), the original change team has been disbanded, and the CEO is focused on the next strategic priority. What does Anchoring research suggest is the highest-leverage intervention?
Industry benchmarks
Is your number good?
Calibrate against real-world tiers. Use these ranges as targets — not absolutes.
Time Required for Different Anchoring Mechanisms
Major enterprise transformationsProcess / system anchoring
6-12 months
Incentive / compensation anchoring
12-18 months (one full cycle)
Talent profile anchoring (new hires)
18-36 months (cohort turnover)
Cultural identity anchoring
36-60 months
Source: Kotter, Leading Change (1996) / Gerstner, Who Says Elephants Can't Dance? (2002)
Real-world cases
Companies that lived this.
Verified narratives with the numbers that prove (or break) the concept.
IBM (Gerstner / Palmisano)
1993-2010
Lou Gerstner inherited IBM in 1993 facing imminent collapse — the company was about to be broken up. Gerstner's strategic insight was that IBM's value lay in its ability to integrate technology for customers, not in selling individual hardware boxes. The strategic pivot was announced in 1993-94. But Gerstner spent the next 8 years anchoring it. He rewrote performance reviews to reward cross-divisional collaboration. He restructured executive compensation to be IBM-stock-based, ending the era of divisional bonus optimization. He systematically promoted leaders who embodied 'one IBM' and managed out those who couldn't. He invested in a cultural reset (chronicled in Who Says Elephants Can't Dance?) that touched language, rituals, and stories. Crucially, he groomed Sam Palmisano as a successor who would continue rather than reverse the trajectory. Palmisano took over in 2002 and accelerated the services pivot through 2010. IBM stock went from $13 (1993) to $80+ (2002) to $200+ (2012). The change anchored because Gerstner spent more time anchoring than launching.
Strategic pivot announced
1993-94
Anchoring period under Gerstner
1994-2002 (8 years)
Successor (Palmisano) tenure
2002-2012
IBM stock 1993 → 2012
$13 → $200+
Gerstner's IBM is the canonical example of anchoring done right. The visible change took 12 months. The anchoring took 8 years. He paid attention to all four mechanisms: cultural (language, rituals), talent (hiring + promotion), systems (compensation, performance reviews), succession (Palmisano). Most CEOs declare victory at year 2; Gerstner stayed until year 9. The successor continued the trajectory because the change was anchored, not just enforced.
Hypothetical: Mid-Size Tech — Failed Anchoring
2022-2024
A 1,500-person SaaS company launched an 'engineering excellence' initiative in 2022, focused on code review discipline, test coverage, and reduced tech debt. Adoption hit 78% within 6 months — a clear win at launch. The CTO, the architect of the change, was promoted to a different role at month 12. The change initiative was 'declared complete' and the program team disbanded. By month 18, code review SLAs had drifted from 18 hours back to 4 days. By month 24, test coverage was below pre-initiative levels. Why? The hiring profile hadn't been updated (new engineers weren't selected for engineering excellence values), promotions still rewarded shipping speed over quality, and no successor leadership had been groomed. The systems and talent anchors were missing. A 12-month rebuild attempt at month 30 failed because by then the org had concluded 'engineering excellence is performative.'
Adoption at month 6
78%
Adoption at month 18
~50% drift
Adoption at month 24
Below baseline
Anchoring mechanisms in place
1 of 4 (cultural only)
Strong launch + weak anchoring = expensive failure. The CTO's promotion was the death-knell — without someone owning anchoring, the change had no chance. Anchoring requires explicit succession planning and structural mechanisms, not just initial momentum.
Related concepts
Keep connecting.
The concepts that orbit this one — each one sharpens the others.
Beyond the concept
Turn Anchoring Change into a live operating decision.
Use this concept as the framing layer, then move into a diagnostic if it maps directly to a current bottleneck.
Typical response time: 24h · No retainer required
Turn Anchoring Change into a live operating decision.
Use Anchoring Change as the framing layer, then move into diagnostics or advisory if this maps directly to a current business bottleneck.