The attrition data on women at mid-career is well-documented. What is less documented is what the room looks like in the months before the departure. Here is what I observe.
The mid-career attrition of women is not a sudden decision. It is the end of a process that has been visible for months.
The forty-eight per cent figure comes from multiple studies of women’s career trajectories in corporate India. It refers, approximately, to the proportion of women who at the mid-career inflection point, typically between eight and fifteen years into their corporate careers, either leave the organisation, significantly de-accelerate their career progression, or exit corporate life altogether in ways that are attributable to organisational culture and structure rather than to personal preference or family choice. The figure is well-documented. What is less documented is what the room looks like in the months before the departure: the specific pattern of organisational interactions and signals that precede the decision, that are visible to anyone trained to see them, and that are rarely acted on by the organisations that produce them.
I have been in a significant number of rooms with women who subsequently left. Not all of them, and not always before the decision was made, but enough to have observed a pattern that repeats with enough consistency to be worth naming. The pattern is not dramatic. It does not involve a single pivotal incident, a dramatic confrontation, or a visible crisis. It involves an accumulation of small signals, individually deniable, collectively legible as a message from the organisation: your contribution here has a ceiling, and the ceiling is lower than your capability. When this message is received clearly enough, and when the investment required to continue performing at the same level in its presence exceeds the return the person can reasonably anticipate, the decision to leave becomes straightforward. The question is always: why was it not obvious earlier, and why was it not addressed?
What I Observe
The first signal I consistently observe in the months before a talented woman leaves a senior role is a change in her MEETING PRESENCE. Women who were, six months earlier, actively contributing in senior meetings, raising challenges, completing their thoughts without visible self-editing, begin to contribute differently. The contributions become shorter. They are more qualified. They arrive later in the conversation, after the dominant voices have established the frame. The person is still fully capable of making the contribution she was making before. She has learned, through a series of interactions that are individually ambiguous but collectively clear, that the contribution will be received differently from how the equivalent contribution would be received from certain colleagues, and she is managing her contribution accordingly. This management is rational. It is also the beginning of a process that ends, if the culture does not change, in the organisation losing the person entirely.
The second signal is a change in the SPONSORSHIP CONVERSATION. Women who were, earlier in their tenure, being actively discussed for stretch assignments, visible projects, and promotion opportunities begin to fall out of the conversation. In the talent review language of Indian corporates, they become “solid performers who may not be ready yet.” The “yet” is important: it is a marker of sustained ceiling rather than a genuine developmental observation. The woman who was being discussed as a future leader eighteen months ago is now being managed within her current scope. She knows this, because she is senior enough to observe the conversations around her and to notice who is being discussed for what. The signal she receives from this change is precise: the organisation’s investment in her progression has diminished, and the reason offered, “not quite ready,” is not accompanied by a specific developmental plan that would close the gap.
The third signal, and the most important, is a change in the WAY SHE TALKS ABOUT HER WORK in the contexts where she can be direct. In coaching sessions, in culture focus groups, in exit interview equivalents where honest feedback is sought, she describes the work with a tiredness that is different from ordinary workload pressure. It is the tiredness of performing in a context that does not fully credit the performance. It is the specific exhaustion of covering, of managing the gap between what the role could be and what the culture permits it to be, for long enough that the investment no longer feels proportionate to the return. This tiredness is not dramatic. It is quiet, and it is very easy to mistake for something else, burnout or personal difficulty or ambivalence about the career itself, all of which are more comfortable explanations than the one that requires the organisation to examine its own culture. The misdiagnosis is not always innocent.
What the Organisation Is Missing
The pre-exit signals described above are visible. They are visible to the people who know what to look for, and they are often visible to the person’s direct manager if the manager is paying sufficient attention. The question worth asking is: why are they so rarely acted on? The most honest answer involves three elements. The first is that the signals are individually deniable. Each one, taken in isolation, has an alternative explanation that is more comfortable than the one this article is pointing toward. The change in meeting presence could be workload pressure. The change in sponsorship could be a genuine development readiness question. The tiredness in the coaching session could be personal rather than organisational. The deniability of each individual signal makes it very easy to address each one with the most comfortable available interpretation and to miss the pattern that they form together.
The second element is that acting on the signals requires the manager or the organisation to name something uncomfortable: that the culture is making it harder for this person to contribute and progress than it should be, and that the organisation has some responsibility for that. This naming requires the kind of cultural honesty that most Indian corporate cultures are not yet built for, and it requires the manager to have a different kind of conversation with the employee than the performance conversation or the development conversation that most managers are equipped for. It requires a conversation about the employee’s experience of the culture, which is a conversation about the manager’s own behaviour and the organisation’s own norms, and that is a significantly more uncomfortable conversation than “you need to be more visible.”
The third element is structural: the organisation’s measurement systems are not designed to detect the pre-exit signals. Annual engagement surveys are too infrequent and too aggregated to capture the trajectory that this article is describing. Exit interviews, where they produce honest data at all, produce it too late to be actionable. The talent review process, which is where the sponsorship conversation happens, is not designed to surface the belonging deficit signals that precede the change in sponsorship allocation; it is designed to evaluate current performance and future potential, and neither of these is the conversation that the pre-exit pattern requires. Building the measurement infrastructure to detect these signals requires specific investment in qualitative listening, in regular one-on-one conversations about belonging and experience rather than only about performance and development, and in a data collection approach that can capture trajectory rather than only snapshot.
“Every woman who left an organisation I was working with had been signalling her departure for at least a year before she made the decision. The signals were visible. They were not seen because the organisation was not designed to see them.”
RK, Founder & Principal Consultant, ProventusHRWhat to Do
The most important practical response is to BUILD THE LISTENING INFRASTRUCTURE before the signals become a pattern. This means creating regular, structured conversations between senior women and their managers or HR partners that are specifically designed to surface belonging experience rather than only performance and development. The conversation is simple: “Are there things about how this organisation works that are making it harder for you to contribute fully than they need to be?” This question is hard to ask because the honest answer is uncomfortable. It is also the most valuable question available to any manager who is trying to retain talented women, because it surfaces the pre-exit signals at a point when they are still addressable rather than at the point where they have become a departure decision.
The second response is to examine the SPONSORSHIP DISTRIBUTION data with the same rigour applied to other equity metrics. Who is being discussed for what opportunities? Is the distribution equitable across gender, background, and identity? If not, what specifically is producing the inequity, and whose sponsorship behaviour is most influential in creating it? These questions are answerable with data that most organisations already have in their talent review process but rarely analyse through an equity lens. The analysis does not require a large investment. It requires the willingness to look at the data and act on what it shows.
The third response is senior leader behaviour change of the specific kind described throughout this series: the daily micro-behaviours of idea attribution, contribution equity, and meeting presence protection that signal to non-dominant group members whether they are genuinely included or conditionally tolerated. These behaviours do not produce overnight culture change. They produce, over time, a different quality of signal to the women who are currently in the pre-exit trajectory: a signal that the organisation is actively working to include them rather than passively observing them leave. Some of these women will stay because of that signal. None of them will stay unless it is produced.
“The most expensive talent conversation an organisation can have is the exit interview. The cheapest is the belonging conversation it could have had eighteen months earlier.”
RK, Founder & Principal Consultant, ProventusHRResearch Reference
Lean In India (2023). Women in the Workplace India. Catalyst (2023). The Power of Belonging. McKinsey & Company (2023). Women in the Workplace: Fifteen Years of Progress and Persistent Gaps.
The ProventusHR Perspective
ProventusHR’s Women emPower programme is designed partly to build the capability and community of women at the mid-career inflection point, and partly to surface and address the organisational conditions that are producing the pre-exit pattern described in this article. The programme is built to do both because neither alone is sufficient. Building women’s capability and resilience in an organisation whose culture has not changed to include them is building a stronger swimmer in a pool whose conditions are not safe. Changing the conditions without building the swimmers’ capability is changing the pool without supporting the people it was designed for. Both are necessary, and the Women emPower design reflects both.
The organisational conditions component of Women emPower works with three stakeholder groups: the women participants themselves, their direct managers, and the senior leadership team that sets the cultural conditions for the managers. The manager engagement is structured as a specific behaviour change programme rather than an awareness session, using the same design principles described in the inclusive leadership article elsewhere in this series. The senior leadership engagement is a half-day culture audit conversation facilitated by ProventusHR, which examines the specific organisational practices and norms that are producing the pre-exit signals, and produces specific commitments from the senior team to address them. This three-level design is what distinguishes Women emPower from women’s programmes that work only with the women and then return them to an unchanged environment.
For organisations experiencing the forty-eight per cent problem, ProventusHR’s starting recommendation is a structured listening exercise rather than a programme design. Specifically, a series of six to eight facilitated conversations with senior women at different tenure points, asking the specific belonging experience questions described in this article, and mapping the responses to identify where in the organisational journey the belonging deficit signals most consistently appear. This listening exercise takes two to three weeks, costs a fraction of a programme design, and produces the specific diagnostic intelligence required to design an intervention that addresses the actual problem rather than the assumed one. In most cases, what it reveals is both more specific and more addressable than the organisation expected.
Every engagement begins with a discovery conversation. No templates, no generic programmes; only work that fits the context, the culture, and the people involved.
Start a Discovery Conversation