If you have been working in HR, or have had behavioural-based management training, you will have come across Unconscious Bias or Diversity Awareness.  Indeed, there is a huge amount of Bias and Diversity training our there.  The McKinsey Corporation estimated it accounted for some $8bn USD in 2017. So that's a lot of courses and consultants.  However, we are also now seeing some question the value of these Unconscious Bias Training (UBT) programmes. Most notably the UK Civil Service. They very publicly ended theirs in 2020 and even advised other public-sector bodies to do the same. 

So what is going on? 
Do these programmes really work and what's the evidence? 

First, lets look at what is is Unconscious Bias 

In essence, it refers to the deep-rooted prejudices that we all have and absorb from living within society.  It tends to be more severe, the more deeply unequal a society gets.  It leads to instinctual behaviours and assumptions about others.  For example, a view that nurses must be women and engineers men.  Research suggests even people committed to equality can find it hard to overcome, as it is often unconscious and may not be manifestly obvious to the person doing it. 

How big a problem is it? 

It can be big. As an example, at the time of writing, there seems to an almost continuous stream of news from the USA about the behaviour of white police officers against both black and Asians. Its not just race, it affects views and behaviours on gender too. You can find plenty of research showing that companies, with higher proportions of women board members outperform the others.  It is also been well observed that diverse teams will tend to outperform homogenous ones, especially when solving complex problems. 

"When bias clouds the decision-making process of any human resource team in charge of an organisation’s selection and recruitment process. The negative effects are obvious. " (ref 2) 

So why did the UK Civil Service drop their programme?

The UK Civil Service stopped their programme in 2020. If not controversially enough, they also advised other public-sector organisations to do the same. However, its not just them. There are also others who have  questioned the effectiveness of UBT programmes. Perhaps, even more controversially, KPMG's UK Chair, Bill Michael, in an online public meeting; referred to unconscious bias as "complete and utter crap". His reasoning given being: "Because after every single unconscious bias training that has ever been done, nothing’s ever improved".  His comments were perhaps unwise as they ended up with his resignation... 

So, what is the evidence?

The UK Civil Service decision to drop UBT was based mainly on a report by the UK Government  Behavioural Insights Team. See link below for a summary (ref 1). In essence, it says whilst unconscious bias and diversity training are common in organisations trying to creating a fairer and more inclusive workplace. That it is also easy to procure, with many and varied training providers. But evidence shows that their training content and techniques is 'lacking' with the research review concluding; that such training interventions do not seem to be effective at improving diversity outcomes within workplaces. 

Some issues the report gave suggesting the limitation of UBT effectiveness were summarised as follows: 

  • People resent being made to do something and so are not receptive to the training 
  • The training brings to mind unhelpful stereotypes which people then act upon 
  • The training makes people think that the organisation has now solved its diversity problems and that any difference in outcomes or experiences are justified, or at least then not due to individuals’ biases.

So how then do we address Unconscious Bias? 

Perhaps the answer is to look beyond training. Achieving behavioural change is hard in any organisation.  Unconscious Bias and Diversity training in itself won't for example, increase representation of minorities in senior management. We suggest, such training may only be worth doing if it is supported and contributes to a longer-term sustainable programme.  By way of example, there is certainly work many organisations could do to reform their recruitment processes. These we know can have a major impact on minorities and diversity; especially for example in how job adverts are phased and presented. Ask yourself, do your job adverts encourage minorities to apply?  Staying just on  recruitment. You can a lot to improve selection to limit bias. For example, use Structured Interviewing - that really works well. Also what about having HR present a list of potential interview candidates to managers for scoring. With names, photos, sex, age and ethnic identifying information withheld.  Another technique we have heard that can work wonders, is using diversity to deliver more diversity.  In essence this is getting your staff and managers to have more interactions with greater numbers of diverse people. 

Lastly a method we hear works well (its used by Google if you need a recommendation, is to base your efforts around 4 elements: 

  1. Structure of Success (basically what is your objective, what does success look like); 
  2. Measure Results (be focussed on assessing how you are getting to your objectives); 
  3. Evaluate the subtle messages (symbolism is important, be aware of how actions send messages); 
  4. Hold everyone accountable (decisions need to be justified and bias called out). 

The overall message about is that most organisations will benefit hugely from more diversity and inclusivity.  But, achieving it is hard work.  There is evidence that training on its own will not deliver much. Indeed, you might be wise to question the benefit you get from those costly UBT and Diversity training courses. Other methods, including the few examples given here, may be more effective or at least in turn will help make your UBT training more effective.


1. UK Civil Service - Unconscious bias and diversity training – what the evidence says   https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/944431/20-12-14_UBT_BIT_report.pdf 

2. Harvard University - Babcock 2006