7 reasons why your approach to diversity, equity & inclusion data may be problematic

Data is one of the most powerful tools organisations have for understanding inclusion, equity and systemic inequities. It helps reveal where disparities exist, track progress over time and inform targeted action. But collecting and using EDI data well requires more than a survey or a headline figure. It demands careful thought about how the data is gathered, interpreted and acted upon and a commitment to centring the people behind the numbers.
Too often, organisations fall into the same traps: collecting data that does not tell them much, drawing conclusions that miss the real story or treating the data collection itself as the outcome rather than the start of meaningful change. Below are some of the most common pitfalls to avoid, and a checklist to use to help you do better.

1. When the data analysed is not specific or intersectional, so it doesn’t really tell you much without generalisations
For example: “10% of our staff are Black” but you don’t know if they are all the same gender, age or anything else about them. Plus you don’t know whether they all work in one department or are the lowest-paid staff.
Without this level of detail, it is impossible to identify patterns of inequality or design targeted strategies that address disparities in representation, pay or progression.
2. When there is a lack of trust and psychological safety in the organisational culture, so people are dishonest or have omitted information during data collection
For example: when an engagement survey reads positively but people didn't add many comments and the quantitative data doesn't correspond with the high staff turnover and sickness-absence rates.
This often happens when employees do not believe the survey is truly confidential and fear that being honest could lead to backlash or negative consequences for their career, so they provide safe answers rather than truthful ones.
3. When it fails to centre those who are marginalised by focusing on the majority of responses being positive, rather than seeing the negative responses as an issue to explore
For example: “89% of our colleagues feel their voices are listened to... so we are doing great!” or "75% of our staff feel proud of our DEI work". Meanwhile, the 11% and 25% do not feel the same and could be more likely to be part of marginalised communities.
Focusing on majority views is the opposite of what DEI work is about. The goal should be to understand and close the gaps for those who feel excluded or marginalised, rather than celebrating results that may simply be masking deeper inequities.
4. When the data collection exercise is seen as an outcome in itself and uses up all of the DEI resources for the year
For example: software or consultants used for the data collection cost $$$ and your DEI budget is inadequate, and by the time new budget is allocated or the DEI team has capacity to take action, the data is actually out of date.
Data collection should never be the outcome. Analysis and action are, and if the process has taken too long or cost too much, the organisation may miss the window to act quickly, build momentum or deliver visible progress.
5. When someone's lived experience is not seen as enough to act upon, and instead, multiple surveys and focus groups are 'needed' to get examples from 'enough' people to prove that the organisation is impacted by systemic issues
For example: it has been a year since a Black colleague shared multiple experiences of racism in an exit interview and the company has spent the past 12 months trying to collect data that tells them it was a one-off problem so they do not need to do anything about it.
This is often a waste of time, as those who want to dismiss lived experience will never believe there is enough evidence to act because their resistance is not about evidence at all but about not wanting to make change.
6. When data collection is done in a way that is triggering for marginalised folks
For example: listening groups are conducted by untrained senior leaders who want to be educated on issues by those with lived experiences, but the senior leaders re-trigger the participants by asking unsafe questions and there is no after care for those who have taken part.
When handled insensitively, these exercises can do more harm than good, deepening mistrust and discouraging people from participating in future data collection efforts.
7. When there is insufficient or no communication about how marginalised people's feedback has led to changes, or worse, nothing is actually done with the information
For example: the data collection takes months and culminates in a final report that is not shared beyond senior leadership level and there is no buy-in to take forward the recommendations.
This lack of transparency erodes trust and signals that leadership views the exercise as a formality rather than a commitment to meaningful action.