Nov 4 2026 | theoutcastcollective
Two years ago a DEI audit in India counted things. A POSH committee in place, women by level, the unconscious bias module completed. In 2027 it has to answer a harder question: with global parents pulling back, boards asking for proof, and the leadership numbers slipping, is any of this changing how people are actually managed?
India did not follow the global retreat. That is not the whole story.
When American companies started cutting DEI in 2025, the question in Indian boardrooms was whether India would follow. Mostly it did not. Through 2025, DEI-related job postings in India rose by around a third. TCS, Infosys, Axis Bank, SAP India and Nestle India expanded programmes rather than renaming them. A survey of US-headquartered multinationals operating in India found that only 28 percent paused anything at all. Regulation did part of the work: BRSR disclosure, the Rights of Persons with Disabilities Act, the POSH Act and the Companies Act requirement for women on boards kept the function funded while American parents quietly dismantled theirs.
So the Indian picture looks like resilience. Then you read the 2026 Avtar and Seramount Best Companies for Women in India study, which covers 387 participating organisations, and the picture changes. Among the best companies, women’s share of corporate executive roles fell from 20 percent to 16.7 percent in a single year. Overall representation slipped from 35.7 to 34.6 percent. Pharmaceuticals dropped from 24.6 to 17.3 percent. Manufacturing sits at 13.1 percent. These are the organisations that apply to be measured. If the leading companies are losing ground at the top while hiring more DEI staff, the money is not landing where the outcomes are decided.
That is the real shift between 2025 and 2027 in India. Not whether organisations fund inclusion, but what they fund, and whether anyone can show it works.
Three gaps the Indian numbers expose
The headquarters gap. India-headquartered companies in the Avtar study have 31.4 percent women overall and 14.2 percent in corporate executive roles. US-headquartered companies operating in India have 41.1 and 18 percent. Other foreign-headquartered companies have 39.2 and 20 percent. The comfortable explanation is that MNCs import mature programmes. The harder reading is that Indian companies have built DEI functions without building the manager capability underneath them.
The sector and city gap. Professional services, ITES and GCCs lead at 38 to 46 percent. Manufacturing, which employs far more people, is at 13 percent. Bengaluru, Chennai and Hyderabad are near 40 percent. Delhi-NCR is at 28.5 percent. A DEI audit that reports a company-wide number is averaging a Bengaluru GCC with a plant in Pune and calling it a result. The audit has to be cut by site, by function and by the manager population, or it hides the thing it was commissioned to find.
The reason-for-leaving gap. For the second year running, health and wellbeing has overtaken childcare as a reason women leave. The old explanation for women’s attrition in India was the family. The data now says people are leaving because of how the work feels day to day. And how work feels day to day is set almost entirely by the person they report to.
Where the money went, and where it did not
Across the organisations we work with in India, three movements are visible regardless of whether the overall budget rose or fell.
From visible to embedded. The Pride month post, the Women’s Day panel, the externally branded programme. These were the first things cut by US parents and, in India, the first things the CFO began asking about. What replaced them, where anything did, was quieter and more defensible: second-career returnship programmes, which 83 percent of India-headquartered best companies now run; gender-neutral parental leave; POSH and grievance processes rebuilt so they are actually used; inclusion questions added to the engagement survey. Less to photograph. More to audit.
From gender to everything else. The Avtar inclusion index shows focus on generational diversity rising from 86 to 92 percent and on regional and cultural diversity from 77 to 83 percent. This is the right direction for a country where a Bengaluru team may hold five mother tongues and three generations, and where the quiet exclusions are usually linguistic and regional first. It also means the audit can no longer be a women’s representation report with a disability annexure.
From the central function to the line. Mphasis and upGrad are among the Indian companies that now hold managers accountable for team diversity inside the appraisal cycle. This is the right move and it exposes a problem nobody has resourced. You can make a manager accountable for an outcome. You cannot make them capable of producing it by writing it into their KRAs. Most organisations have shifted the accountability to the line without shifting a single rupee into finding out whether their managers know how.
Why the 2025 audit no longer answers the question
The standard Indian DEI audit was built for a world where the risk was absence, and where the regulator was the main reader. Is the POSH Internal Committee constituted and is the annual return filed? Is BRSR Principle 3 reported? Are there women on the board? How many women at each band? What percentage completed the bias module? It told leadership and the regulator that things existed.
In 2027 that dashboard answers a question nobody is asking. CHROs and boards are asking three different ones. What are we actually spending, and on what? Can we show that it changes anything? And if the global parent cuts our allocation by a third next year, what do we keep?
A presence audit cannot answer any of those. It can tell you that 94 percent of managers completed the training. It cannot tell you whether a single one of them runs a team meeting differently, allocates a high-visibility project differently, or says anything when a senior colleague makes a remark about a regional accent in a leadership review. And since those three things are where inclusion is experienced, an audit that cannot see them is measuring the policy and the paperwork while assuming the middle.
The POSH example makes the point sharply. An Internal Committee that has received no complaints in three years is reported as a clean record. In most Indian organisations it is a question, not an answer: either nothing has happened, or nobody believes the committee is worth going to. The audit has to be able to tell the difference.
Four things an Indian DEI audit has to establish in 2027
Where the effort is landing. Map inclusion spend and activity against three layers: leadership and policy at the top, compliance and process at the bottom, and the manager layer in between. In most organisations the first two account for nearly all of the money and the third for nearly all of the experience. That gap is the finding.
Whether the system produces fair outcomes or fair-looking process. Pull the last twelve months of stretch assignments, client-facing projects and discretionary rewards across one function and list who received them. Check whether promotion criteria were written before the names were discussed or after. A calibration process can be run by the book and still deliver the same six names every cycle, because each individual decision was defensible and nobody looked at the pattern. In India, where tenure and loyalty are rewarded instinctively, this is more common than anyone admits.
Whether the data is honest. Most inclusion data is self-reported, and most self-report instruments are built so that the desirable answer is obvious. A manager asked to rate herself on ‘I ensure all team members have equal opportunity to contribute’ will answer four or five, and she will mean it. The instrument has measured her self-image, not her behaviour. A 2027 audit should test every self-report tool it inherits against a simple question: could a motivated respondent pick the best answer without it being true? If yes, the data cannot carry the weight being placed on it, and the audit should say so.
What the organisation is resourcing versus what it is saying. Line up the commitments in the annual report, the BRSR disclosure and the careers page against the actual budget, headcount and manager time allocated to inclusion. Usually the commitments were written in 2023 and the budgets in 2026, and nobody reconciled them. Putting the two side by side gives leadership a choice: fund the commitment, or change it.
A frame for the 2027 audit
We structure it in five layers, and the weight has moved down the list compared with two years ago.
Governance and intent. Board composition, executive ownership, external commitments and whether they are resourced. This used to be most of the audit. It is now the shortest section.
Systems and compliance. Hiring, promotion, pay, grievance and POSH processes, examined for whether criteria precede names and whether the process is used, not just whether it exists.
Manager practice. The new centre of the audit. How managers actually distribute airtime, opportunity, recognition and the benefit of the doubt, measured with instruments designed to resist the polished answer. Scenario-based rather than agreement-scale. Checked for consistency. Private to the manager, aggregated for the organisation, so the data is honest in both directions.
Employee experience. Pulse data cut by team, site and manager, not just by gender. Attrition reasons read against the manager, because the wellbeing finding is a finding about managers even though it is reported as a finding about women.
Outcomes. Representation by level, site and sector over time, yes. But also the distribution of discretionary opportunity, the retention of people who raised a concern, and whether any of it has moved since the last measurement. One baseline is useful. A baseline and a re-measure is the only thing that proves a programme worked, so build the second measurement into the budget for the first.
Where to put a smaller budget
The Avtar study carries one number that should reassure anyone defending an inclusion budget to an Indian CFO: 79.2 percent of best companies, up from 70, report higher operating profit from inclusive practices. The case exists. What is missing is proof that their own spend contributes to it. That argues for a specific reallocation.
Fewer programmes, more measurement. One honest baseline of manager practice is worth more than three awareness campaigns, because it tells you which of the three you needed.
Fix the template, not the person. Where the audit finds that criteria are written after the names, change the appraisal and project-allocation forms so they cannot be. One intervention, at the point where the default happens, for every manager in the company. It costs almost nothing and it does not depend on anyone’s good intentions surviving quarter-end.
Develop managers by starting point, not by batch. A manager who defaults to the people he came up with needs something different from one who is strong inside her team and silent in the leadership review. The audit should tell you which is which. Then the development budget goes where the gap is, instead of being spread across every manager in a batch and changing no one.
The question the board will ask
In 2025 the question in India was whether the organisation would hold its DEI line while the world retreated. Mostly it did. In 2027 the question is whether anyone can show that it works, and what would be lost if the allocation stopped. A DEI audit built for compliance cannot answer that. One built for the manager layer gives leadership something it has not had before: a clear view of where inclusion is actually decided, an honest reading of how their managers lead, and a short list of changes that will hold whether or not the budget does.
If you would like to see how the manager-practice layer is measured, including how the instrument is designed so that the obvious answer does not score, we are happy to walk through it.
Sources
Storyboard18, ‘As global firms pull back, Indian companies double down on DEI roles and investments’, 20 June 2025.
Avtar and Seramount, 2026 Best Companies for Women in India study, September 2026.
Take the first step today. Schedule an exploratory consultation via WhatsApp at +91-9372177748 or email lakshmi@theoutcastcollective.com with our DEI experts and start building a workplace where everyone belongs.