The ninety-day signal: spotting executive derailment before it costs the year

The cost of a senior derailment is not the search fee. It is the year of organisational drift before the board admits the hire is not working. We have run the numbers on the placements where the derailment was eventually acknowledged: the median time between the first observable warning sign and the formal board conversation about the appointment is eleven months. Eleven months in which the senior team's confidence in the appointment erodes, the natural authority of the role hollows out, and the operational decisions that should have been the new leader's are increasingly made by the chair, the CFO, or the senior team functioning around them rather than through them.

The window in which this can be intervened in cleanly is roughly the first ninety days. After that, the cost of intervention rises sharply: the senior team has formed a view, the principal stakeholders have invested credibility in the appointment, the board does not want to be seen as having made a hire it is now stepping back from. The longer the conversation is deferred, the higher the price of having it.

This piece sets out the framework we use, in our practice, to read the ninety-day signals across senior placements. It is built on the patterns visible in 142 senior placements that JOH Partners tracked through the first eighteen months on seat between 2014 and 2025. It draws on the wider literature; Spencer Stuart's onboarding research and Heidrick's Route to the CHRO programme document patterns we have seen confirmed in our own data. The framework is not theoretical. It is the framework partners on our practice run when a chair calls in week eight asking, candidly, whether the new leader is going to make it.

The signal is not the surprise. The board's failure to read it is.

Why ninety days matters

The first ninety days of a senior appointment is the window in which the new leader is forming the operating model the role will run on. The decisions taken in that window, the cadence set in that window, the team-first hires made in that window, are the decisions, cadence and hires that the rest of the leader's tenure will be measured against. After ninety days, the pattern is set; before ninety days, the pattern is forming.

The window is also the window in which the senior team and the principal stakeholders are most attentive. In the honeymoon weeks, the new leader is given the benefit of the doubt; in the post-honeymoon weeks, the team begins to read the signals; by week twelve, the team has formed a working view of the appointment that will be substantially harder to shift later. The boards that read the signals during the formation window have the option of intervention; the boards that read them after have, in most cases, the option of replacement.

The figure that anchors this work is consistent across our 142-placement dataset and the publicly reported onboarding work in the wider industry. Eighty-nine percent of the placements that ultimately derailed showed signals that were, in retrospect, visible within the first ninety days. The signals were, in most cases, not subtle: the new leader was not making the decisions the role required, was spending time on the wrong activities, was hiring against the wrong playbook, or was failing to integrate with the senior team in ways that the chair, with the right lens, would have seen by week eight.

The argument is not that the signals are easy to read. The argument is that the signals are there to be read. The discipline of reading them, against a structured framework, is the discipline that distinguishes the chairs who run their senior appointments well from the chairs who, eleven months later, find themselves running an unplanned succession.

The six dimensions of the ninety-day signal

The framework runs on six dimensions. Each dimension has a recognisable pattern of what good looks like, a recognisable pattern of what derailment looks like, and a single diagnostic question the chair can ask in week eight to surface the underlying signal. The dimensions are not equally weighted; in our experience, the decision-rhythm and team-first-hires dimensions are the most predictive, while the public-posture dimension is the most noisy. But the full six-dimension read is the read that produces the most reliable signal in aggregate.

Decision rhythm

The first dimension is decision rhythm. The senior leader who is running the role is the senior leader who is making decisions in the room, with rationale, and moving the queue. The senior leader who is being run by the role is the senior leader who is deferring decisions, calling for more data, lengthening the queue. The pattern is visible by week six in most appointments and is the single most predictive of the six dimensions.

What good looks like is a leader whose direct reports leave one-on-ones with clarity on what was decided, what was tabled, and what they are now expected to deliver. What derailment looks like is a leader whose direct reports leave one-on-ones with the sense that the conversation was useful but the decision was deferred, again. The week-eight diagnostic is a direct one: ask the leader to walk through the three hardest decisions they have made in the last month, and listen to whether the answers are concrete (specific decisions, with specific consequences) or general (themes, directions, intentions). The leaders who can name three concrete hard decisions in their first month are, in our experience, the leaders who are running the role.

Calendar mix

The second dimension is calendar mix. The new leader's calendar in weeks four to ten is the most reliable signal of where the leader thinks the role's leverage is. In our placement data, the leaders whose calendars showed roughly sixty percent of time on team and operations, twenty-five percent on stakeholders and fifteen percent on strategy work were the leaders who were, by month six, embedded. The leaders whose calendars showed seventy percent of time on stakeholders, with the team and operations work delegated to direct reports, were the leaders whose senior teams, by month four, were quietly running the operations layer without the leader and around the leader.

What good looks like is a calendar where the new leader is visibly present with the senior team, in the operating reviews, in the field, in the customer conversations that matter. What derailment looks like is a calendar dominated by chair, principal-stakeholder and external-stakeholder time, with the operating presence delegated. The week-eight diagnostic is to ask the leader to walk through last week's calendar by category. The honest leader will surface the imbalance themselves; the leader who is unaware of the imbalance is the leader who has not yet developed the operating instinct the role requires.

Team-first hires

The third dimension is the team-first hires. The first one or two hires the new leader makes, and the playbook those hires represent, are the most concrete expression of the new leader's operating thesis. The leaders who are hiring against the thesis they were brought in to deliver, with hires whose profiles match the new direction, are the leaders who are forming the operating layer that will execute. The leaders who are hiring against last role's playbook, with familiar names and familiar profiles, are the leaders who are reaching for the comfort of the operating layer they ran before, in a context that does not, in most cases, resemble it.

The pattern is recognisable: the new CFO who hires the financial controller they worked with at the previous group; the new CEO who hires the chief of staff from their last firm. These hires are not, in themselves, a derailment signal. They become a derailment signal when the hire's profile does not match the institution's actual need, and when the leader is not showing the willingness to widen the bench beyond the personal network. The week-eight diagnostic is to ask, of the first planned hire, why this person and why now. The answer that is grounded in the institution's specific need is the good answer; the answer that is grounded in personal trust capital, with the institution's need treated as a secondary fit, is the answer that warrants attention.

Stakeholder integration

The fourth dimension is stakeholder integration. The new leader who has, by week eight, spent meaningful one-on-one time with every member of the senior layer plus the key matrix partners (the senior figures in adjacent functions whose cooperation the leader will need over the next twelve months) is the leader who is investing in the relationship infrastructure the role will run on. The new leader whose stakeholder time is concentrated on direct reports plus a few favoured peers is the leader who is not yet building the breadth that the role's matrix complexity will require.

What good looks like is a stakeholder map that is being deliberately filled in. What derailment looks like is a stakeholder map that is being filled in by accident, with the strong relationships being the relationships that happen to form naturally and the weaker relationships being left as gaps. The week-eight diagnostic is to ask, of every name in the senior layer, whether the new leader has spent two hours with that person one-on-one. The chair who runs this question rigorously, against a complete senior layer list, will surface the pattern within a few minutes.

Public posture

The fifth dimension is public posture: how the new leader sounds in the first board meeting, the first all-hands, the first major external appearance. The pattern that good public posture takes is recognisable: confident on operations, modest on strategy in the early weeks, with strategic claims earned through the operational specifics rather than asserted as direction. The pattern that derailment takes is the inverse: bold strategy statements in the first board meeting, thin on operational specifics; the all-hands message that lands with the senior team as ungrounded; the early external appearance that produces internal eye-rolls.

This is the noisiest of the six dimensions because some leaders are, by personal style, less verbal in the early weeks and the absence of bold public posture is not, in itself, a signal of caution. But the leader who comes out of the early public appearances with the senior team's confidence reinforced is, in our experience, the leader who has read the room correctly. The leader who comes out with the senior team quietly comparing notes about the gap between the strategy claims and the operational reality is the leader who has not. The week-eight diagnostic is to ask, after the first all-hands, how the senior team read the message.

Energy and bandwidth

The sixth dimension is energy and bandwidth. The new leader who is, by week ten, visibly running the role with reserves remaining is the leader who has built the operating model the role will sustain. The new leader who is, by week ten, visibly being run by the role, with no reserves, is the leader whose operating model has not yet absorbed the role's actual demand. The pattern is most visible in the way the leader handles unexpected demands: a regulatory development, a senior team conflict, an external-stakeholder request that arrives unplanned. The leader with reserves handles these as additional work; the leader without reserves handles them as crises.

The bandwidth dimension is sometimes mistaken for a personal-resilience question. It is not. It is an operating-model question. The leader who is being run by the role is, in most cases, the leader who has not yet built the discipline of delegation, the cadence of operating reviews, and the personal time-management structure the role requires. The week-eight diagnostic is candid: ask the leader how many evenings in the last week ended with capacity remaining, and listen for whether the honest answer is none, one, or two-or-more.

Three intervention patterns when signals are bad

Reading the signal is the first half of the work. The second half is the intervention. In our experience, three intervention patterns produce credible outcomes when the week-eight read shows two or more dimensions in the red. The right pattern depends on the specific signals, the institutional context, and the leader's capacity to recalibrate.

The recalibration conversation

The first pattern is the chair-led recalibration conversation. The conversation is direct, structured, and held in week ten or eleven, with the framing made explicit: the chair has seen specific signals on specific dimensions, the chair is sharing those signals candidly, and the chair is asking the leader for the leader's own read on each dimension before any joint conclusion is drawn. The conversation works when the leader has the self-awareness to hear the signals, the operating instinct to translate them into adjustments, and the personal authority to make the adjustments without losing standing with the senior team.

The framing matters. The recalibration conversation is not a performance review. It is not a warning. It is an early-stage course-correction conversation, run while the cost of correction is still low. The chair who runs this conversation in the right register makes it possible for the leader to receive the signal as constructive; the chair who runs it as a warning shot turns the conversation into a defensive one, with the leader spending the next month managing the chair's perception rather than recalibrating the operating model. The right framing is something close to "I am sharing what I am seeing because I want you to succeed, and I want to make sure I am reading the situation correctly before either of us draws conclusions." The wrong framing is anything that the leader could hear as the start of an exit conversation.

The structural reframing

The second pattern is the structural reframing. The reframing is the recognition that the signals are bad not because the leader is the wrong leader, but because the role itself is the wrong role. This pattern is more common than chairs typically expect. In our placement data, roughly a quarter of the early-derailment cases were resolved by adjusting the role rather than replacing the leader. The most common adjustments are: redrawing the reporting line (most often by adding or removing a reporting relationship that was not properly defined at appointment); reducing the scope (most often by separating out a function that was over-loaded into the role); or adding a senior partner (most often a strong COO or CFO who can hold the operating layer the leader was structurally going to struggle with).

The structural reframing is, in most cases, the right answer when the signals concentrate in one or two dimensions rather than spreading across the full six. If the leader is strong on decision rhythm, calendar mix and stakeholder integration but weak on team-first hires and energy, the underlying issue is often a scope problem rather than a leader problem. The chair who reads this correctly can resolve the situation with a structural adjustment that costs the institution very little; the chair who reads it as a leader problem will end up running an unnecessary replacement search.

The early off-ramp

The third pattern is the early off-ramp. The off-ramp is the recognition, after the recalibration conversation has been held and either the leader has not been able to recalibrate or the structural reframing has not produced the necessary change, that the right answer is to part ways in week twelve rather than month eighteen. The off-ramp is, in most cases, the most expensive pattern in raw terms (search fee, settlement, public optics) but the cheapest pattern in cumulative terms.

The mathematics is direct. A clean separation in week twelve costs the search fee for the replacement, a settlement (typically six to nine months for a senior role at this stage), and the optics of an early departure. A delayed separation in month eighteen costs all of those plus the year of organisational drift, the erosion of the senior team's confidence in the chair's judgement, the strategic decisions that did not happen in the intervening period, and the cost of an institution that is, in many cases, harder to recruit into the second time. The chairs who run the early off-ramp pattern, when the signals warrant it, do so because they have done this calculation and they understand which option is actually the more expensive.

The early off-ramp is also, in most cases, more humane to the leader. The senior leader who is removed in week twelve has the personal narrative of "the role wasn't right, we both saw it early, we made a clean call." The senior leader who is removed in month eighteen has the personal narrative of "I struggled, the board lost confidence, I was managed out." The two narratives have very different consequences for the leader's next role, and for the leader's personal account of the appointment.

What this looks like applied

A worked example, anonymised. A regional family group, mid-cap listed, hired a new Group CFO in early 2025 from a global financial services background. The appointment was widely seen as strong: credible CV, good interview process, full reference set, board enthusiasm. By week eight, the chair, working through the framework, identified two dimensions in the red.

Decision rhythm was the first signal. The CFO, in the operating reviews, was deferring most material decisions to "after the next review." The CFO's direct reports were leaving the reviews unclear on what had been decided, and the financial-controller layer was, in practice, making the operating-finance calls without clear CFO direction.

Calendar mix was the second signal. The CFO's calendar was heavily weighted toward external stakeholder time: investor relations preparation, banker meetings, capital-markets work. The senior finance team was being run by the senior controller, with the CFO present but not leading.

The chair ran the recalibration conversation in week eleven. The framing was direct: these are the two patterns I am seeing, here is the data behind them, what is your read. The CFO, who turned out to have substantial self-awareness, surfaced a third pattern the chair had not seen: the CFO had been hiring the team they intended to bring in but had not yet been able to backfill the senior controller relationship the institution actually depended on. The CFO needed eight more weeks to land the team-first hire that would let the calendar shift back to operations.

The chair gave the CFO the eight weeks. By month four, two of the three dimensions were green and the third was on track. By month six, the appointment was rated, by both the chair and the senior team, as embedded. By month eighteen, it was running. The intervention cost: one structured conversation, one explicit eight-week recalibration plan, and one chair who had read the signal at week eight rather than at month four.

The counterfactual is the same appointment with the chair reading the signal at month four rather than week eight. By month four, the senior team would have formed a view of the CFO as "external-stakeholder focused, light on operations." That view, once formed, takes substantially more than eight weeks to reverse. The intervention that worked in week eight, with the senior team's view still forming, would not, in most cases, have worked in month six with the senior team's view already set.

What good chairs do in week twelve

The good chairs run the calibration conversation in week eight and the recalibration conversation, if it is needed, in week ten or eleven. They do this because they have done the calculation on the cost of waiting and they understand that the conversation is harder, not easier, in month four. They do it because they have a framework for what they are listening for, rather than a vague sense of unease that they are not yet able to translate into specific feedback. They do it because they have a relationship with the new leader in which a direct conversation is possible without it being read as a warning shot.

The bad chairs do not run the conversation at all in the early weeks. They wait. They tell themselves that the new leader is still settling in, that the signals are noise, that the situation will improve as the role embeds. By month four, the signals have hardened. By month six, the senior team has formed its view. By month eleven, the chair is, in most cases, running the same conversation the chair could have run in week ten, except that the conversation is now an exit conversation rather than a recalibration conversation, and the institution is now eleven months deeper into the operational drift the chair could have prevented.

The discipline of the ninety-day signal is not difficult. It requires a structured framework, a chair willing to use it, and a calibration conversation held at the right moment with the right framing. The mathematics of the cost of waiting is what makes the discipline worth running. The chair who reads the signal in week eight saves the institution the year that the chair who reads it in month eighteen does not.

JOH Partners runs senior search and onboarding mandates across the Gulf, the United Kingdom and Singapore. For confidential conversations on early-tenure leadership transitions, contact the partners directly.