Our society is diverse and, fortunately, this does not stop at the workplace. At the same time, companies still find it difficult to take advantage of this social diversity and to recruit people from a wide range of backgrounds, for a variety of reasons. But what if you would like to make your company more diverse, but don't really know what diversity actually encompasses and how you can measure how diverse your employees are? These are questions we are asked again and again by customers and workshop participants, so we have put together a few tips for you.
Diversity - what does that even mean?
Anyone dealing with equity and inclusion in the workplace will quickly stumble upon two topics: women’s equality and the inclusion of people with (severe) disabilities. Many companies have quotas or attempt to give them preferential treatment when hiring in order to reduce existing disadvantages. They’ll likely have a women's or equal opportunities officer or, in larger organizations, a representative for severely disabled employees. Why these two groups in particular? One reason is that their equity in the workplace is particularly regulated by German laws, e.g. by the Zweite Führungspositionen-Gesetz, the Entgelttransparenzgesetz and regulations of SGB IX. These laws provide specific requirements for companies that must be taken into account.
However, diversity goes far beyond empty phrases and paragraphs.
Diversity really is about the question: Which characteristics shape a person and the way they are perceived by others? Which perspectives, which expertise, which connections does someone bring with them as a result of these experiences? A diverse company is one that brings together people from many different backgrounds – not only in terms of gender and disability, but also in terms of origin, race, age, sexual and gender identity, religion, worldview, social background, neurodivergence, family circumstances, weight, beauty standards and much more.
Measuring diversity in a company - how to do it (legally)
If you look at this list of characteristics, you will probably quickly realize why it can be difficult to measure diversity: even if a trait shapes someone’s identity, you don't necessarily see it from the outside. So if you want to make diversity measurable in your company, you have to rely on the self-disclosure of your employees. No problem, just fire up the survey program and start a little poll! Unfortunately, it's not that simple, because employers are generally not allowed to ask their employees or applicants about their personal circumstances. This is a good thing, because it protects against discrimination – but it also makes measuring diversity in a company and identifying weaknesses in terms of diversity and discrimination more difficult.
However, that doesn't mean that such a survey is impossible to do. Rather, it is important that the information is provided voluntarily and anonymously, and that no conclusions can be drawn about individual persons. So while a survey of employees is the first choice for collecting data on diversity in the company, it needs to be well thought out.

Tips for your diversity survey
So how can you, e.g. as a human resources or people and culture manager, conduct a survey that provides you with valuable insights but is also legally sound?
First of all, make sure that the survey is completely anonymous (e.g. does not require entering a name or email address or other identifying information) and that all questions are answered voluntarily, e.g. by making participation voluntary to begin with or providing answer options such as “I prefer not to say” or “No answer”. It's best to check with your legal counsel on how to ensure that your survey is considered anonymous and voluntary, because as long as anonymity and consent are given, you are allowed to collect data that goes beyond the basic data necessary for the employment relationship.
Be transparent about why you want to collect this data, who will have access to it, and how you will ensure anonymity. This gives participants the security they need to answer honestly – because if 70% of the answers are “I don't want to say,” that won’t be much use to you.
Get a first impression of your employees before you start the survey. Are there any individuals who stand out because of a certain characteristic? If, for example, there is only one woman or only one Black person on the team, then this information can make it easy to track which data set belongs to them, and the survey would no longer be anonymous. Don't just think about visible characteristics, but also about things like positions and titles. For example, if you want to ask about positions and include “team leader” as an option, but you only have four team leaders, then it is possible that this could lead to conclusions about individual people. If you notice such cases, consider how you want to deal with them in the survey. Can you achieve your research goal if you do not include these characteristics or use larger categories? To stick with the first example: If there is only one woman in the company, gender as a category in the survey may be of little relevance and can be omitted. If there is only one Black person but several other employees who are also not white, a broader categorization - e.g. the options “BIPoC”, “white”, “other” - can ensure anonymity without completely leaving out the characteristic of race.
Be precise in your wording and/or supplement your questions with a brief explanation of what is meant. For example, if you want to know how many of your employees have experienced racial discrimination, it makes more sense to ask directly about that than about their migration background (which white people can also have). If you are not sure whether all employees would understand certain terms, either try to find a familiar term or explain it. (Our glossary can help you with this.)
Many identities do not easily fit into predefined categories. It therefore makes sense to always offer an option such as “other” or a free text field where participants can enter their self-designation. The latter has the advantage that it gives you more information than just the fact that the person's identity does not fit into your predefined options.
Think about the context. Even if you only want to collect demographic data, it can be useful to look at overlaps in the evaluation, which can provide you with interesting insights. For example, are there many women, but 90% of them are interns? Are all employees with disabilities in a particular team? Are your younger employees very international, but all older employees are German? This is especially important if you ask about experiences of discrimination – because if you only find out that people have been discriminated against, but not why, you cannot derive targeted measures.
And lastly: even if it is tempting, especially with the complex topic of diversity, don't overload your survey. Consider which characteristics or experiences are particularly relevant to your research interest and limit yourself to these. But consider ending the survey with an open text field where participants can share anything relevant to them that wasn't covered by the existing questions. This way, you can determine in the analysis whether you underestimated certain identities or experiences or didn't consider them at all.
I have figures on diversity – now what?
Making diversity measurable in your company is the first step from which you can derive follow-up measures. What do the employees who are already there need? Who is missing? How can you bring new perspectives into the company and reduce existing inequalities? If you need support in this, we at IN-VISIBLE offer you various consulting services - and are also happy to take over the entire design of your survey for you. Just write to us!