The call came on a Tuesday afternoon. A CEO I had known for years — someone who ran a mid-market distribution company with genuine operational complexity — wanted a recommendation for a CTO. He had just lost his third in four years.

"They keep leaving," he said. Then a pause that carried more weight than the sentence. "Or we keep letting them go."

I asked what had happened each time. The stories were structurally identical. A crisis — a failed migration, a security scare, a system that buckled under a growth inflection — followed by an urgent hire. The company found someone with the right resume: enterprise experience, calm under pressure, a track record of turning chaos into order. Each time, the new CTO delivered. Stabilized the environment. Reduced risk. Built capability where there had been none. And then, somewhere around month fourteen, friction arrived. Not over performance. Over fit. The organization wanted its pre-crisis rhythm back. The CTO wanted to build something that would prevent the next crisis from arriving in the first place. Within two years, they were gone.

He thought he had a hiring problem. I recognized the pattern because I had lived inside it.

There is a version of this story — the comfortable one — where the problem is the leader. Poor communication skills. Insufficient political awareness. A failure to "read the room." I spent years accepting that version about myself. Each time the cycle completed — crisis hire, stabilization, friction, departure — I added another line to the private ledger of things I needed to improve. I would learn to be more patient with consensus processes. I would calibrate my directness. I would study the politics with the same intensity I studied the architecture.

The ledger grew longer. The pattern did not change.

The Research That Reframed the Pattern

The shift started when I stopped treating each cycle as a personal performance review and started treating it as data. Three cycles is a coincidence. Five is a dataset. And datasets deserve frameworks, not self-help.

Amy Kristof-Brown's 2005 meta-analysis of person-environment fit — PE fit — synthesized 172 studies encompassing 836 independent effect sizes. The finding that stopped me: values misalignment between an individual and their organization is among the strongest predictors of turnover. Stronger than job dissatisfaction. Stronger than compensation mismatch. PE fit measures the congruence between what a person values and what an organization actually rewards. When the gap is large enough, departure is not a choice. It is gravity.

Reading that research was like finding an X-ray of something I had felt but could not name. Every cycle had the same shape — not because I kept making the same mistakes, but because the structural conditions kept producing the same outcome. The leader who thrives in crisis — decisive, systems-oriented, willing to name uncomfortable truths — is exercising capabilities the organization urgently needed. But those same capabilities, once the fire is out, produce friction against the organization's baseline values. Directness becomes "abrasiveness." Architectural ambition becomes "overengineering." Investment in prevention becomes "spending on problems that have not broken yet."

Benjamin Schneider's Attraction-Selection-Attrition framework explained why the baseline was so resistant to permanent change. His research — spanning 13,000 managers across 142 organizations — describes a cycle that operates below conscious intention. Organizations attract people who share their values. They select for cultural similarity. And over time, they grow more homogeneous as misfits attrite. The framework's central prediction — that within-organization personality similarity exceeds within-industry similarity — has been validated repeatedly. The organization is not choosing homogeneity consciously. It is producing it structurally, through thousands of small decisions about whom to hire, promote, and retain.

The crisis creates a temporary rupture in the ASA cycle. The company's normal selection criteria — consensus orientation, operational coordination, cultural harmony — are briefly suspended in favor of raw capability. They need someone who can diagnose failure modes, make architectural decisions under pressure, and build systems that work. For a few months, the organization's stated values and its actual needs align perfectly. The systems thinker walks into what appears to be an environment built for them.

It is not a deception. It is a temporary state. And it is the mechanism that kept the cycle turning through my career and through the careers of every technically deep leader I have compared notes with since.

The Interview and the Reversion

I have been on both sides of this distortion, and understanding the mechanism did not make me immune to it. That is the part I find most worth admitting.

When a company is in pain — systems failing, vendors circling, the board asking pointed questions about technology risk — it presents as an organization that values technical depth and decisive leadership. The language of crisis-mode hiring sounds like the language of permanent commitment. "We need someone who can make hard calls." "Technology is critical to our future." "We are ready to invest." Both parties mean every word. Both are describing a version of reality that has a shelf life measured in quarters, not years.

At one organization, the interview conversations centered entirely on building technology as a competitive capability. By month sixteen, the conversation had shifted to reducing the team I had built to stabilize their infrastructure. Not because the team was underperforming — the metrics were unambiguous. Because the crisis was over, and the structural view of technology as a cost center had quietly reasserted itself. The ASA cycle doing what Schneider's research predicts: restoring the organization's pre-crisis equilibrium.

The post-crisis reversion is predictable, not malicious. That distinction matters. I wasted years being angry at organizations for changing the rules, when the more accurate reading was that the rules had never actually changed — they had just been temporarily suspended. The organization did not betray its commitment to technical depth. It returned to a commitment it never actually held. The crisis had created an impression of alignment that felt real because, in that moment, it was real.

This is where the technology-as-cost-center dynamic compounds the problem. In organizations where technology is not the product — retailers, distributors, services firms, PE portfolio companies — the CTO role exists in persistent structural tension. Preventing crises requires investment, capability, and operational maturity. Governing technology as overhead requires minimization. The CTO who prevents the next incident looks, to the finance function, indistinguishable from someone spending money on problems that do not exist. The return on prevention is invisible by definition. I have built business cases for investments whose primary evidence of success was the absence of the disaster they prevented. That is a difficult argument to win in a quarterly review.

The recruitment signal problem follows directly. At the attraction stage of Schneider's ASA cycle, the organization in crisis temporarily speaks the systems thinker's language. Architecture matters. Depth matters. The org chart says the CTO reports to the CEO. The critical question — "What does this organization value at month fourteen, when nothing is on fire?" — is the one question the crisis environment makes nearly impossible to answer honestly. Not because anyone is hiding the answer. Because the crisis has genuinely, temporarily changed it.

I have asked myself, more than once, whether I should have read the signals more carefully. The honest answer, informed by the research, is that the signals during a crisis are genuinely ambiguous. The organization is not performing a false version of itself. It is experiencing a real, temporary shift in what it needs. The mismatch is temporal — the right leader for the wrong phase — and temporal mismatches are the hardest kind to detect from inside them.

Three Paths Out of the Cycle

After watching this pattern across enough organizations and enough careers — including my own, examined with the kind of honesty that only comes after you stop defending the narrative you preferred — the data points converge on three structural paths that lead somewhere other than the cycle's default ending.

Engineering-first organizations. PE fit research is clear: congruence between personal values and organizational values produces higher satisfaction, stronger performance, and lower turnover. Organizations where technology is the product, or where engineering capability is genuinely structural rather than ornamental, have ASA cycles that select for architectural depth. The evaluation is not the CEO's interview rhetoric. It is where technology sits in the org chart, how investment decisions are governed, what the CTO tenure pattern looks like over the past decade, and whether the board includes anyone with a technical operating background.

Advisory and fractional engagement. The crisis hire cycle contains an asymmetry worth examining: the systems thinker's highest-value contribution is concentrated in the first twelve to eighteen months — the diagnostic phase, the operating model design, the capability build. The advisory or fractional model matches the engagement to the window where PE fit is strongest, then exits before the ASA cycle completes its reversion. This is not a lesser form of leadership. It is an engagement structure that aligns capability to context without requiring the leader to absorb the organizational immune response that follows stabilization. The fractional CTO movement has grown 57% since 2020, which suggests I was not the only one reading the data this way. I took this path. The analytical case was stronger than the emotional reluctance to leave traditional employment — though I will not pretend the reluctance was small.

Independence. For leaders who have cycled through enough organizations to recognize the pattern as structural, building a practice removes the PE fit mismatch entirely. You are no longer inside an ASA cycle that was not built for you. The trade-off is market risk for organizational friction — a different kind of uncertainty, but one where the feedback loops are transparent and the evaluation criteria are yours to define.

What I Told the CEO

Four evaluation signals are accessible before the offer letter. Each tests whether the organization's crisis-mode presentation reflects durable values or temporary need:

Post-crisis narrative. How does the company describe what happened to the last technology leader? If the story centers on the individual's personality rather than the structural conditions, the ASA cycle is intact and unexamined.

Role authority structure. Does the technology leader have genuine budget authority and a direct line to the CEO? Authority that requires permission from a non-technical chain of command is a coordination role, not a leadership role — regardless of the title.

Predecessor tenure patterns. Three technology leaders in five years is not a hiring problem. It is an environmental signal. The data is there if you ask for it.

Investment thesis. Is technology positioned as a capability investment or a cost to be managed? The answer is not in the job description. It is in the capital allocation history and where IT spending sits in the P&L.

That CEO who called me about his third CTO — I did not give him a name. I gave him these questions. How does your board talk about technology when nothing is broken? What happened to the last CTO's initiatives after they left? When you describe wanting a "strategic technology leader," are you describing a permanent role or a crisis response?

He was quiet for a long time. Then he said something I have thought about since: "I think we keep hiring for the company we want to be during the crisis, not the company we actually are."

That sentence is the crisis hire cycle reduced to its mechanism. And the way out — for organizations and for the leaders who keep cycling through them — starts with reading the pattern as structural signal rather than personal failure. I spent years adding entries to that private ledger, convinced the problem was something I had not yet learned. The research says otherwise. The mechanisms are named. The pattern has information in it. And the career that compounds rather than resets begins with the willingness to ask, before you accept the role, what the room looks like when the fire is out.

References

• Kristof-Brown, A.L., Zimmerman, R.D., & Johnson, E.C. (2005). Consequences of individuals' fit at work: A meta-analysis of person-job, person-organization, person-group, and person-supervisor fit. Personnel Psychology, 58(2), 281-342.
• Schneider, B. (1987). The people make the place. Personnel Psychology, 40(3), 437-453.