Diversity & Inclusion

I've Run D&I Training at Three Companies. Here's Why Most of It Didn't Work.

8 min read
I've Run D&I Training at Three Companies. Here's Why Most of It Didn't Work.
Diversity & Inclusion
💬 Personal View

Quick answer

Most D&I training fails because it is run as a box-ticking event: mandatory, generic, imported from global vendors and judged by attendance or satisfaction scores. What works is listening to employees first, connecting training to hiring, promotion and team decisions, making it safe to admit bias, and measuring belonging, not just representation.

Key takeaways

  • Mandatory attendance without buy-in produces completion data, not behaviour change.
  • Representation dashboards measure presence, not belonging or psychological safety.
  • Generic global modules miss Indian bias: caste, region, religion and age.
  • Start with listening, connect training to real decisions, and make honesty safe.
  • Training is a small part of a wider D&I commitment covering sponsorship, flexibility and money.

Let me be honest: I've sat through enough corporate D&I training sessions to fill a warehouse. And not just as a participant. I've designed and run them at three companies in the past decade. Each time, I went in optimistic. Each time, I watched the same pattern repeat. The slides were polished, the facilitators were earnest, and the post-training survey scores were pristine. But come Monday morning? Back to business as usual.

It took me until my second company to realize the real problem wasn't the content. It was everything around it.

Why does mandatory D&I training become a box-ticking exercise?

Because it is treated as a task to complete rather than the start of a conversation, and nobody asks employees what they actually need before it is rolled out.

In January 2020, fresh into my role as Head of People at a mid-sized fintech startup in Bangalore, I inherited a D&I plan from my predecessor. We had dates, budgets, and a shiny new unconscious bias module licensed from a Silicon Valley vendor. The problem? No one had asked the people working here what they actually needed.

We launched the training in February. Attendance was mandatory, a compliance requirement, as stated in the HR compliance sheet. I watched spreadsheets full of people take it in the background while they actually did their jobs. The net promoter score was a cheerful 7.2 out of 10. On paper, we'd delivered. In reality, we'd checked a box.

The fundamental mistake: we treated D&I training as a task to complete, not as the beginning of a conversation. Mandatory participation without genuine buy-in is just theater.

Why don't D&I metrics show whether training worked?

Because most dashboards measure presence, such as representation by level, rather than belonging, such as whether people feel safe speaking up.

At my third company, a manufacturing group with 2,500 employees across four plants in Gujarat. I decided to measure everything. Gender representation by level. Caste diversity in hiring pipelines. Retention rates for women and minorities. We produced a stunning dashboard.

It felt important. It looked important. It wasn't nearly important enough.

Here's what we missed: you can't improve what you're not measuring emotionally. We were measuring presence, not belonging. A woman in a meeting room isn't the same as a woman who feels safe speaking up. A diverse hiring cohort isn't the same as a workplace where everyone gets the same mentoring, sponsorship, and opportunity for growth.

"Training that doesn't connect to your actual workplace culture is just a productivity loss wrapped in good intentions."

Why do generic unconscious bias modules fail in India?

They make people feel stupid or defensive, and they miss India-specific bias such as caste, region, religion and age.

You know the one. "Scenarios for Unconscious Bias." A photograph of a woman in a IT uniform and the question: "Whose desk would you walk past without talking to?" And the 'correct' answer? Stop the assumed gender bias.

These modules make people feel stupid if they get it wrong. Worse, they make people defensive if they recognize themselves in the scenario. And they almost always miss the India-specific context: caste-based bias, regional discrimination, religious exclusions, age-ism in tech hiring. A generic global module doesn't land the same way when your workplace has its own complex history.

I finally understood this when a manager in our Vadodara plant told me, privately, "I don't need a California module to tell me I might have biases. I need to know how to actually talk to people different from me and what I'm unconsciously assuming." He was right.

What makes D&I training actually work?

Listen before you design, connect training to real decisions, and make it safe to be honest.

Looking back, here are the three shifts that actually started to move the needle:

1. Start with listening, not lecturing. Before designing training, I should have sat with 30 people from different levels, backgrounds, and teams. What do you experience that makes this workplace feel excluding? What would make it feel like you belong? Those answers shape everything.

2. Connect training to systems and decisions. D&I training divorced from hiring practices, promotion criteria, team composition, and leadership accountability is just inspiration. It needs to touch the actual machinery: how teams are formed, how budgets are allocated, and how success is measured.

3. Make it safe to be honest, and imperfect. The moment training becomes about right answers, people stop thinking. The moment managers feel they'll be punished for admitting bias, they stop admitting anything, which means they never address it. Psychological safety matters more than perfect knowledge.

What I did, and what I would do instead
AreaThe box-tick versionWhat started to work
Starting pointAn inherited plan and a licensed Silicon Valley moduleSit with people from different levels, backgrounds and teams and ask what makes the workplace feel excluding
ContentGeneric global scenarios and "correct" answersIndia-specific context: caste, region, religion and age
Link to the businessTraining as a standalone eventConnected to hiring, promotion criteria, team composition and leadership accountability
ToneRight answers; people fear admitting biasSafe to be honest and imperfect
MeasurementAttendance, a 7.2 out of 10 score, a representation dashboardBelonging, psychological safety and culture change

Is D&I training enough on its own?

D&I isn't a training program. Training is a small part of a much larger commitment. It's the leadership having hard conversations about succession planning and who gets access to the sponsor in the room. It's adjusting team composition and communication norms so that quiet voices are heard. It's designing work flexibility for carers, and ensuring LGBTQ+ colleagues don't have to hide. It's examining your procurement practices and your supplier base. It's about money, where it goes, whose ideas get funded.

The training that worked? It was embedded in that larger ecosystem. It wasn't a day out of your calendar. It was a conversation about how we actually work here, what we want to become, and what you have to do differently to get there.

This January, as companies across India are making their annual D&I commitments, I'm hoping more HR leaders avoid the path I took. Skip the eye-roll module. Ask people what they need. Connect training to real decisions. Make it about belonging, not just representation. And measure culture change, not compliance.

If you are rebuilding a programme along these lines, start with our diversity and inclusion training hub and the Diversity & Inclusion course. For the decision-level design of bias training, see unconscious bias training: what actually works.

Frequently asked questions

Why does most D&I training fail?

It is run as a one-off, mandatory box-tick with generic content, and success is judged by attendance or survey scores. Nothing connects it to the hiring, promotion and team decisions where inclusion is won or lost.

Should D&I training be mandatory?

Mandatory participation without genuine buy-in tends to produce completion data rather than change. Start by listening to employees so the training answers problems they recognise.

What should we measure instead of attendance?

Belonging and psychological safety, not just representation. A woman in the meeting room is not the same as a woman who feels safe speaking up, so measure who gets mentoring, sponsorship and growth opportunities.

Do global D&I modules work for Indian employees?

Rarely as they are. They tend to miss India-specific bias such as caste, regional discrimination, religious exclusion and age bias in tech hiring.

What diversity data do listed companies have to disclose in India?

SEBI's Business Responsibility and Sustainability Report (BRSR) format asks for employees and workers by gender, including differently abled employees and workers, the representation of women, and whether the entity has an equal opportunity policy under the Rights of Persons with Disabilities Act, 2016.

Sources

  1. SEBI circular: Business Responsibility and Sustainability Reporting by listed entities, 10 May 2021
  2. SEBI BRSR format (Annexure I to the May 2021 circular)
  3. Rights of Persons with Disabilities Act, 2016 (India Code)

Move Beyond Training Box-Ticks

Structured D&I programs that connect to your actual workplace culture, measure psychological safety, and drive real behavior change.

Request Demo
Diversity Inclusion Training IndiaDEI Corporate TrainingUnconscious Bias TrainingInclusive Workplace IndiaHR Training Programs