The Glass Cliff: When You’re Promoted to Fail

EXECUTIVE COACHING FOR DRIVEN WOMEN • July 20, 2026

The Promotion That Felt Like a Warning

In my coaching work with driven women in leadership, I hear a specific kind of good news delivered in a strange tone of voice. Not quite excited. Not quite relieved. Something closer to bracing. A woman gets the title she has wanted for a decade, and in the same breath she tells me about it, she starts listing the reasons it might be a trap. I want to be honest about what I can and can’t tell her in that moment. I can’t tell her whether her specific promotion was designed to set her up. I can tell her that the pattern she’s naming has a name, a body of research behind it, and real limits to what that research proves.

Katie is forty-four and runs operations for a mid-sized logistics company. She had been the obvious internal candidate for the VP role for two promotion cycles running, and both times the job went to someone else. Then the company lost its biggest client, missed a quarter badly, and the sitting VP left for a competitor with better numbers to point to. Three weeks later, Katie had the title.

“I should be celebrating,” she told me on our second call, still in her car in a parking garage, the engine off, her laptop bag on the passenger seat where she’d tossed it after a board call that ran forty minutes over. “Instead I keep doing this math. Why now. Why me. Why this exact quarter.”

I want to be careful here, because the honest answer is not a clean one. Sometimes a promotion that arrives during a crisis is exactly what it looks like: recognition, finally, with unusually hard conditions attached. Sometimes it is something closer to what researchers call the glass cliff, a documented tendency for women to be disproportionately placed in leadership roles when the risk of failure is highest. And sometimes it’s both, tangled together in ways that are genuinely difficult to untangle from the inside of your own career.

This guide is not going to tell you that your promotion was rigged. It’s also not going to tell you the research is nothing to think about. What I want to offer instead is something more useful than either reassurance or alarm: a clear picture of what the glass cliff research actually says, where it holds up, where it falls apart, and how to make a clear-eyed decision about a high-stakes role without assuming a story about intent that the evidence can’t actually support.

What Is the Glass Cliff?

The term comes from a specific piece of research, not a general vibe about women and workplaces. Michelle K. Ryan, PhD, a professor of social and organizational psychology, and S. Alexander Haslam, PhD, a professor of psychology at the University of Queensland, coined it in 2005 after a UK newspaper claimed that women joining corporate boards was causing company performance to decline. Ryan and Haslam went back to the actual FTSE 100 data to check the claim, and found the opposite causal story: companies that appointed women to their boards had already been performing badly for months beforehand. The women weren’t causing the decline. They were being appointed into it.

It’s worth being precise about what this pattern is and isn’t. It is a statistical tendency observed across some datasets, in some sectors, under some conditions. It is not a universal law, and it is not proof that any particular woman’s particular promotion was engineered to fail. Those two things get collapsed together constantly in casual conversation about the glass cliff, and the collapse does a disservice to the actual research.

What the Evidence Actually Shows

I want to slow down here, because this is the section where I think coaching conversations about the glass cliff most often go wrong. People treat “the glass cliff is real” as a settled, uniform fact, and it isn’t. It’s a body of research with real findings, real limits, and real disagreement among the people who study it.

The original 2005 study looked at one dataset (FTSE 100 companies over one period) and found a correlation between women’s board appointments and preceding poor performance. That’s a meaningful finding, but a single archival study of one market in one time window can’t tell you how universal the pattern is. Researchers spent the next fifteen years trying to answer that question with more data, in more settings, using more methods.

The most comprehensive answer so far comes from a 2020 meta-analysis I keep coming back to when clients ask me whether the glass cliff is “real.” Thekla Morgenroth, PhD, and her colleagues pooled dozens of studies, archival and experimental, and the picture that emerged was more textured than the popular version of the story. In real-world archival data, they found only a small effect overall, and it depended heavily on which sector they looked at. There was no detectable glass cliff effect in management specifically, the exact domain where the concept was first identified. But the pattern did show up in politics, education, and nonprofit leadership.

I don’t share that nuance to talk anyone out of taking the pattern seriously. I share it because the specific shape of the evidence matters for what you do with it. If you work in management and you’re weighing a crisis-era promotion, the research base for assuming you’re on a glass cliff is thinner than the research base for a colleague in a political campaign or a nonprofit executive search. That doesn’t mean your situation is fine. It means the label doesn’t do the diagnostic work people want it to do, and you still have to look at your specific role, your specific resources, and your specific board.

Where the Research Gets Contested

What convinced me the glass cliff deserved a more careful treatment than a tidy villain narrative was reading about the studies that failed to find it at all. A 2019 analysis of German and UK companies looked for the same pattern Ryan and Haslam had identified in the original FTSE data and came up empty. The performance trend before a female executive’s appointment in that dataset was no worse than the trend before a male executive’s appointment. Same general question, different countries and years, no supporting evidence.

A 2024 Harvard Business Review conversation with researcher Sophie Williams puts language to something I see constantly in coaching sessions with women trying to figure out whether their own promotion fits the pattern. Williams names three practical factors worth checking before assuming you’re on a glass cliff: recent profitability and share-performance trends, the availability of internal support and resources for the role, and whether leadership at the organization has historically been uniformly male until this exact opportunity opened up. None of those factors proves intent. They’re closer to a checklist for how precarious a role’s underlying conditions actually are, separate from anyone’s motives in offering it to you.

I read a BBC Future piece on this research not long after a client described her own promotion in almost the exact language researchers use to describe glass cliff appointments, and the piece named the thing I’d been trying to say to her more precisely than I had managed on my own. The journalist interviewed Morgenroth directly, and Morgenroth’s own read on her data was refreshingly uncertain: in management, she said, the glass cliff may not exist at all, or there may be moderating factors her analysis didn’t capture. That kind of honesty about the limits of a finding is rare in popular writing about workplace research, and it’s exactly the posture I want to model for clients. The article also noted something worth sitting with: the effect appears smaller in countries with higher measured gender equality, like Switzerland and Germany, than in the US and UK, which suggests the pattern is shaped by broader cultural and institutional conditions rather than being some fixed law of how organizations behave.

The honest summary of fifteen years of research is this. The glass cliff is a real, replicated pattern in some domains and some conditions. It’s a contested, sometimes unsupported pattern in others, including in general corporate management, where it was first described. Selection effects, sample size, country, industry, and time period all move the needle on whether a given study finds it. That’s not a reason to dismiss the concept. It’s a reason to hold it as a hypothesis worth checking against your specific situation, not a verdict you can hand down about someone else’s intentions.

Five Roles That Get Mistaken for Each Other

One of the most useful things I can do in a coaching conversation about a difficult promotion is slow down long enough to figure out which of several distinct situations someone is actually describing, because the right response is different for each one, and they get talked about as though they’re interchangeable.

An ordinary stretch role. The job is genuinely bigger than what you’ve done before, the learning curve is steep, and it’s uncomfortable in the specific way that growth usually is. This is difficult, not designed to fail. Most promotions, even good ones, come with some version of this discomfort.

An under-resourced role. The title and the responsibility are real, but the budget, staffing, or authority to actually do the job weren’t included. This can happen to anyone, for reasons ranging from a genuine resourcing crunch to a hiring manager who didn’t think it through. It becomes a glass cliff concern specifically when the pattern of who gets the under-resourced version of a role tracks along gender or other demographic lines.

A restructuring or turnaround assignment. The organization is in real trouble, and the role exists specifically to fix it. These roles carry a genuinely higher risk of visible failure than a stable role does, and that’s true regardless of who holds them. The relevant question isn’t whether the job is hard. It’s whether you were given a realistic shot, meaning the time, authority, and resources a turnaround actually requires, or whether the difficulty was set up to guarantee an outcome.

Unlawful discrimination or retaliation. If you have reason to believe you were denied resources, support, or a fair evaluation because of your sex, race, age, disability, or another legally protected characteristic, or because you reported misconduct, that is a legal question, not a coaching question. I am not able to assess whether discrimination occurred, and I wouldn’t want to guess in a blog post at what your specific facts add up to. That determination belongs with an employment attorney who can review your actual documentation and your jurisdiction’s law.

Harassment or a genuine safety concern. If what you’re describing involves harassment, threats, or anything that makes you feel physically or psychologically unsafe at work, that is not something a leadership framework like the glass cliff should be used to explain away or manage around. That warrants its own support, through your organization’s reporting channels, an employment attorney, or a licensed mental health professional, separate from any coaching conversation about career strategy.

A performance problem. Sometimes a role genuinely isn’t going well because of factors within a leader’s control: a skill gap, a misjudged decision, a team that wasn’t built well. It’s uncomfortable to consider this alongside the glass cliff conversation, but a rigorous coaching relationship has to leave room for it. Not every difficult outcome in a hard role is evidence of a rigged system, and assuming otherwise can get in the way of the specific, practical adjustments that might actually help.

Most real situations aren’t a clean example of just one of these. A restructuring assignment can also be under-resourced. An under-resourced role can also involve a performance gap that a better-resourced version of the same job wouldn’t have exposed. The point of naming them separately isn’t to force a diagnosis. It’s to keep from reaching for the most dramatic explanation before checking whether a plainer one fits the facts.

Why the Pattern Is Easy to Feel and Hard to Prove

Caitlin is thirty-nine, a marketing executive at a consumer goods company, three months into a promotion she’d wanted since she started there. Her division had been flat for two years before her promotion, and the CMO who preceded her had left abruptly for reasons the company never fully explained internally.

“Everyone keeps telling me how excited they are for me,” she said in our fourth session, turning her coffee cup a quarter turn on the table between us, not drinking from it. “And I keep waiting for someone to also mention the part where the budget got cut eleven percent the same week I got the title. Nobody’s mentioning that part. I don’t know if I’m supposed to mention it either.”

What I notice in conversations like this one is how much certainty people want from a framework that, by its nature, can’t fully deliver it. Caitlin wanted me to tell her whether she’d been set up. I couldn’t tell her that, and I told her so directly. What I could do was help her separate the parts of her situation she could actually verify (the budget cut, the timeline, the resourcing relative to her predecessor’s) from the parts that were interpretation (why the board made this specific choice, what anyone intended). The first category she could act on. The second category, in the absence of something she could point to and document, was mostly going to keep her awake at night without moving her any closer to a decision.

This is the real difficulty with the glass cliff as a personal framework rather than a research finding. Research aggregates across hundreds of cases and looks for a statistical tendency. You are one case. You cannot run a regression on your own promotion. What you can do is look honestly at the concrete conditions of the role: the budget, the authority, the timeline, the support, the track record of who has held this role before you and what happened to them, and use that information to make the best decision available to you, without requiring an answer to the unanswerable question of what was in anyone else’s head when they made the offer.

I see a version of this same dynamic in executive coaching for women in BigLaw, where the partner track gets into your nervous system long before any single promotion decision arrives, and the pressure to read intent into every assignment becomes its own exhausting, full-time interpretive job. The skill that helps there is the same one that helps here: naming what is verifiable, and setting down what is not.

Both/And: A Pattern Can Be Real AND Your Choices Still Matter

The women I coach through this question almost always want one of two stories to be true. Either the situation is entirely structural, meaning nothing they do will change the outcome and they can stop torturing themselves over strategy, or the situation is entirely within their control, meaning if they just work hard enough and think clearly enough, they can out-perform any bad setup. Neither story holds up on its own.

Both things can be true at once. The conditions of a crisis-era promotion can be genuinely harder than a stable-era version of the same role, shaped by dynamics that have nothing to do with your competence and everything to do with organizational patterns documented across decades of research. And the specific choices you make once you’re in the role, what you negotiate for before accepting, how you communicate about resourcing, when you escalate a concern, still matter and still shape the outcome within those harder conditions.

Holding both of those at once is uncomfortable, because it removes the tidy explanation in either direction. It means you don’t get to fully blame the system, and you don’t get to fully credit or blame yourself. What you get instead is a more accurate, more useful picture: a real structural pattern that shapes the odds, and a real set of decisions still available to you inside it.

The Systemic Lens: Why Crisis Roles Keep Finding Women

Even accounting for every caveat in the research, something is happening at the level of the system that’s worth naming without collapsing it into any individual’s intent. Several mechanisms researchers have proposed don’t require anyone to be acting in bad faith to produce a lopsided pattern.

One is a version of what researchers call “think crisis, think female”: stereotypes about communication and people-management skills that make women look like a more intuitive fit specifically when an organization believes it needs to repair relationships or morale, rather than when it’s simply performing well and wants continuity. Another is what some researchers call a status-quo bias: organizations that have been led by men without much scrutiny of that pattern tend not to feel pressure to change anything until a crisis forces the question, at which point a woman becomes appealing as a visible signal of change. Neither mechanism requires a boardroom conspiracy. Both can operate through ordinary, largely unexamined habits of thought that shape who gets considered for what, and when.

Naming the systemic layer matters for a specific reason. If the pattern is structural, the fix has to include structure: succession planning that considers women for stable, well-resourced roles at the same rate as crisis roles, boards that ask what resources a new hire actually has before congratulating themselves on a “bold” appointment, and post-mortems on failed turnarounds that look honestly at whether the role was resourced adequately in the first place. An individual woman managing her own career well cannot fix a selection pattern that operates at the level of hundreds of boards making similar choices. That’s not a reason to give up on individual strategy. It’s a reason not to mistake individual strategy for the whole solution.

What Coaching Can and Cannot Help With

I want to be direct about the boundaries of what a coaching relationship, or this article, can responsibly do for you.

Coaching can help you get clear on the actual, verifiable conditions of your role: the budget, the timeline, the authority you were and weren’t given, and how those compare to the role’s history. It can help you think through what you’d need to negotiate before accepting a similar offer in the future, and how to communicate about resourcing gaps in a way that protects your credibility rather than reading as complaint. It can help you separate the parts of a hard situation that reflect your decisions from the parts that reflect conditions outside your control, so you’re not carrying blame that doesn’t belong to you or ignoring the parts that do.

Coaching cannot tell you whether you were the target of unlawful discrimination or retaliation. That’s a legal determination, and if you have specific facts that concern you, an employment attorney licensed in your jurisdiction is the right person to evaluate them, not a blog post or a coaching call. Coaching cannot tell you whether to confront your board, how to document a case for a future legal claim, whether to resign, or whether to go public with your experience. Those are significant, often irreversible decisions with legal and financial consequences, and they deserve counsel from people with the specific expertise to advise on them: an employment lawyer for the legal exposure, a financial advisor for the financial impact, and, if the situation is taking a serious toll on your well-being, a licensed mental health provider who can assess and support you directly. This article and any coaching built on it is educational content, not legal, HR, financial, or clinical advice, and it isn’t a substitute for any of those.

If you’re navigating something that involves real distress, whether from the situation itself or from what it’s stirring up, that matters and deserves real support. A coaching conversation about strategy is not the right container for a mental health concern, and a good coach should say so plainly rather than trying to hold everything at once.

Questions Worth Asking Before You Say Yes

If you’re weighing whether to accept a high-risk role, here’s what I’d actually ask, based on what the research and years of coaching driven women through exactly this decision suggest matters most.

What happened to the last three people in this role? Not the sanitized version. The actual outcome, and how the organization talked about it afterward.

What resources come with the title, in writing? Budget, headcount, decision-making authority, and a realistic timeline before your performance gets formally evaluated. Vague verbal reassurance is not the same as a documented commitment.

Is the difficulty of this role about the work, or about what you weren’t given to do the work? A hard job with real support is a different proposition than an impossible job with none.

What’s your actual exit position if it doesn’t work out? Severance, references, how a difficult tenure in this specific role tends to be discussed afterward at this specific company. This is a negotiation question, not a defeatist one.

Who else was considered, and why do you think it landed on you, specifically, right now? You may never get a fully honest answer. Asking the question, and watching how it’s received, still tells you something.

None of these questions can tell you with certainty whether you’re looking at a glass cliff. They can tell you a great deal about whether the specific role in front of you is set up in a way that gives you a real chance, which is the more practical question underneath the research label anyway.

If You Think You’re Already On the Cliff

If you’re past the decision point and already in a role that feels like this, a few things tend to help more than others. Get the resourcing gaps in writing, in your own words, sent to the people who can address them, not because it guarantees a different outcome but because it creates a clear, contemporaneous record if you need one later and it forces a direct conversation instead of a slow, unspoken erosion. Separate your professional narrative about the role from your sense of your own worth as early as you can, because the two get tangled fast under pressure, and untangling them after a hard outcome is much harder than maintaining the separation while you’re still in it. Build support outside the role itself, whether that’s a coach, a mentor who has moved through something similar, or a peer group, because isolation tends to make every part of this heavier than it needs to be.

And if, at any point, this starts to feel like more than a career strategy question, if you notice it affecting your sleep, your relationships, or your basic functioning in ways that feel outside your capacity to manage on your own, that’s worth bringing to a licensed mental health professional directly. That’s not a failure of resilience. It’s the same kind of clear-eyed assessment this whole guide has been asking you to apply to your career: naming a situation accurately so you can get the specific kind of help it actually calls for.

If you’re weighing a role like this and want to think it through with someone who coaches driven women through exactly this kind of decision, you’re welcome to look at executive coaching with me, or to start with Fixing the Foundations^™, my foundational course on the patterns that shape how driven women make high-stakes career decisions. The course cart opens September 8, and you can join the waitlist now.

I think often about what Katie said toward the end of our work together, once the initial crisis at her company had stabilized and the board had, for whatever mix of reasons, kept her in the role. “I stopped needing to know why they picked me,” she said. “I just needed to know what I was actually working with. That was the part I could do something about.”

That’s the distinction underneath most of this. You may never get a clean answer about why a hard opportunity landed on your desk at the exact moment it did. You can get a clear answer about what you’re actually working with, and that’s the answer that lets you make a decision you can stand behind either way.

Related Reading

[1] Ryan, M. K., & Haslam, S. A. “The Glass Cliff: Evidence that Women are Over-Represented in Precarious Leadership Positions.” British Journal of Management 16, no. 2 (2005): 81-90.
[2] Morgenroth, T., Kirby, T. A., Ryan, M. K., & Sudkämper, A. “The Who, When, and Why of the Glass Cliff Phenomenon: A Meta-Analysis of Appointments to Precarious Leadership Positions.” Psychological Bulletin 146, no. 9 (2020): 797-829.
[3] Oakes, Kelly. “The Invisible Danger of the ‘Glass Cliff.'” BBC Future, February 24, 2022.
[4] Eagly, A. H., & Carli, L. L. Through the Labyrinth: The Truth About How Women Become Leaders. Boston: Harvard Business Review Press, 2007.

References

Peer-Reviewed Research (Vancouver)

1. Ryan MK, Haslam SA. The glass cliff: evidence that women are over-represented in precarious leadership positions. Br J Manag. 2005;16(2):81-90. doi:10.1111/j.1467-8551.2005.00433.x.
2. Morgenroth T, Kirby TA, Ryan MK, Sudkämper A. The who, when, and why of the glass cliff phenomenon: a meta-analysis of appointments to precarious leadership positions. Psychol Bull. 2020;146(9):797-829. doi:10.1037/bul0000234. PMID: 32700940.

Books, Journalism & Cultural Sources (Chicago Author-Date)

• Eagly, Alice H., and Linda L. Carli. Through the Labyrinth: The Truth About How Women Become Leaders. Boston: Harvard Business Review Press, 2007.
• Oakes, Kelly. “The Invisible Danger of the ‘Glass Cliff.'” BBC Future, February 24, 2022. https://www.bbc.com/future/article/20220204-the-danger-of-the-glass-cliff-for-women-and-people-of-colour.
• “Why the Glass Cliff Persists.” Harvard Business Review, podcast audio, March 26, 2024. https://hbr.org/podcast/2024/03/why-the-glass-cliff-persists.

Medical Disclaimer