Summary
Research involving 780 UK working adults found that 55.5% had witnessed at least one of five forms of workplace unfairness in the previous six months. Although around three-quarters of witnesses reported taking some action, they were much more likely to respond privately afterwards than intervene at the time. Respondents who could identify a concrete response to a hypothetical incident were also substantially more likely to report having acted when they last witnessed unfairness (approximately 85% versus 27%). Dr Anthony Giannoumis argues that helping employees rehearse practical responses, supported by managers who model inclusive behaviour, could make workplace inclusion more effective. The study shows an association, not a proven causal effect.
Some of the cooling towards highly visible DEI activity is bad faith. Some of it is a receipt: the bill arriving for inclusion work that was performance. That part is the product of our own failure. As advocates we failed to make the case to businesses that inclusion is in their interest. HR is under growing pressure to show commercial value, and the usual programmes are not delivering
For decades we anchored the case in human rights and equality. These remain the most critical issues, and we shouldn't shirk our obligations there. But many senior leaders see inclusion as optional, so we have to persuade them to take responsibility.
Instead, inclusion ended up in one of two places. Either the lawyers own it as legal compliance, or it becomes a marketing hook to show the organisation's commitment to society. The first is far too compliance-oriented. The second is far too fragile to survive economic risk and uncertainty. It isn't seen as driving revenue or cutting costs, so it's the first thing to be cut.
There is evidence linking inclusion to innovation, revenue, bigger markets for accessible products, team performance and lower attrition. We argued for doing the right thing when we should have argued for doing the best thing for the business. We need to draw inclusion into basic business processes and say it plainly: this will improve your revenue and this will help reduce your costs.
So if organisations step back from formal DEI programmes, what needs to happen in everyday work for people to challenge unfair behaviour safely and constructively? My research points to a practical answer. People need a script and the backing to use it.
Everyday unfairness at work: what 780 UK workers witnessed
For my white paper The Script Gap I surveyed 780 UK working adults in May 2026. More than half (55.5%) had witnessed at least one of five kinds of unfairness at work in the previous six months. The most common were a concern dismissed or minimised by a manager (27.1%) and being left out of informal conversations (25.1%). Moments like these rarely become HR complaints. They come with plausible deniability.
I'm not immune. I consider myself an inclusive, considerate and empathetic leader, and I still talk over people, especially when I get excited about an idea. It's a habit I've internalised from years of being the idea guy and the leader, and there's nothing malicious in it. The student who taught me most about this marched down from the back of an auditorium with a notebook full of everything I'd done wrong. I wanted to put her in her place. Instead I shut up and listened.
A lot of leaders struggle with this. They're seen as the authority, the one who is supposed to talk the most. When someone else is contributing, it's easy to steer the conversation elsewhere, interrupt or simply not leave enough space.
Why conventional DEI training may fail to change everyday behaviour
I don't believe withdrawing the programmes themselves will have any real effect on everyday unfairness, because most of them weren't changing it. Most were general policies. The training was often rooted in anti-bias work, and even where it brought in something like intersectionality, it took an academic approach.
In my experience that training was great for a thought exercise. It was great for leaving people with a sense of moral superiority. It was an absolute failure at changing behaviour in the small moments where inclusion matters most: under stress, against a deadline, with no capacity left.
In the survey, around three in four people who witnessed an incident did something about it. Most of that action was private and after the event. Checking in with the colleague later (42%) far outstripped intervening in the moment (about 14%). We trained awareness when people needed a sentence.
What does make things worse is leaders going quiet. If the climate scares you I understand the instinct. Saying nothing feels safe. It isn't. When leaders go silent they hand the argument to the people making it in bad faith.
Knowing what to say when you witness unfairness
Every respondent was given the same scenario. A colleague is repeatedly interrupted in a meeting, looks frustrated and says nothing. What would you actually do? Those who could name a concrete response had acted about 85% of the time in their most recent real incident. Those who couldn't had acted 27% of the time.
How bold the response was made little difference. People whose response was to raise it with a manager had acted 94.6% of the time. For intervening in the moment it was 86.7%, and for a quiet word afterwards 80.7%. Even the most modest script went with roughly three times the action rate of having none. Both questions were answered by the same people in the same survey, so this is a strong association. It does not prove that a script causes action.
Under pressure our reactions are knee-jerk. They come from fight or flight. A script works like muscle memory. An elite footballer knows when to pass and when to take the shot because they've trained for it under all sorts of conditions. Inclusion needs the same rigour: training people for a specific behaviour in the moment.
How managers make it safer for employees to speak up
A good response should change the behaviour of two people: the one speaking up and the one who acted unfairly.
Asked what would most help them speak up respondents chose leaders modelling the behaviour (38%), knowing the organisation would protect them (37%) and trusting colleagues to back them up (35%). Training came fourth (21%). Low demand for training says little about whether it works, because people can't ask for a skill they can't yet picture using.
That puts managers at the centre. People are waiting to see their manager do the small thing in the small moment. A colleague gets talked over, and the manager says, "Hang on, she wasn't finished." That's the green light.
Managers also need to own it when they're the one who slipped, and to make clear that nobody will be penalised for speaking up. Often the person speaking up is junior and the person they're challenging is senior. They need to know they won't be the one reprimanded or passed over for promotion.
That clarity has to come from leaders as well as policy, and leaders have to mean it: "This is what we want. When you see something unfair, we expect you to say something. And this is what we expect you to say."
Challenging unfair behaviour without blame or shame
None of this is about gotcha moments, and that's why the script matters. Many people who want to call out unfair behaviour instinctively treat it as a gotcha: "You did something wrong. I noticed, and I'm going to shame you in front of everyone." Calling someone out turns performative when it's about showing what you know. I've seen people in DEI spaces call out bias while misunderstanding what bias is.
The point is to help the organisation work better. When someone is steamrolled in a meeting the organisation loses their contribution, in that moment and in the future. One action, or inaction, in a meeting travels fast. People learn that this boss isn't someone to cross, won't hear them out or doesn't care what the team says.
The response needs empathy for both people: the one who was talked over and the one who misjudged the moment. Interrupting someone, talking over them or closing a meeting before everyone has spoken isn't necessarily malicious. When we stop treating it as ill intentioned we stop treating it as a gotcha.
Someone with a track record of toxic behaviour is a different and bigger problem. This approach is about the more frequent slip-ups.
What the workplace inclusion research has yet to prove
The Script Gap is the first wave of a three-year study. A follow-up in 2027 is designed to test whether giving people a script changes what they do. The paper makes no gender-specific claims, because men made up only 163 of the 780 respondents. The full anonymised dataset, survey and codebook are published, so anyone can check the work.
How HR can help employees respond to unfairness at work
- Make the sentence visible. Decide what you want people to say when someone is talked over, and put it where they'll see it, on posters for example. It raises awareness without pulling anyone out of their work.
- Have leaders practise it. Use role-play, gamification and immersive practice with coaching. Build it into the training new leaders already get on handling difficult situations.
- Ask about it in surveys you already run. Most surveys ask whether people support fairness. Almost everyone says yes, and that number tells you nothing. Add the meeting scenario as one more question in your next employee survey.
Each of these is deliberately low-cost. Part of the problem DEI has faced is that its initiatives came with huge budgets. That's hard to defend when a business is looking at its bottom line and saying the money isn't there.
About the author
