Gathering information on diversity is increasingly complex. But without measurement it is impossible to know whether efforts to increase diversity are having any effect.

In the UK there is no requirement to report, but firms are encouraged to monitor diversity and gather statistics. Outside the UK governments have different reporting requirements. For example, the Netherlands require companies to report on “people of Turkish descent”, while Germany requires information on employees’ religions, so tax can be paid to the relevant church. Meanwhile, France merely asks for diversity information that is regarded as discriminatory.

There are many pitfalls, not least data protection. The first snag is using definitions in your survey that cause offence. The Equal Opportunities Commission’s (EOC) website is a good starting point. Remember that some definitions may need explaining. At other times you may need flexibility – the EOC does not include ‘Japanese’ ethnicity, which we have added to our questionnaire, as our Japanese employees do not find it satisfactory to describe themselves as ‘Asian’.

When tracking data, consider whether the statistics should be anonymous or linked to individuals. In our view, linking data is best. It allows equality across pay, promotion and appraisal grades to be monitored. However, permission to link must be sought at the point of gathering information and it is important to remember that data gathered before someone joins the firm must be anonymous.

When gathering the statistics some firms require a mandatory form to be completed online. We find that, while e-forms will achieve a reasonable result, lawyers in particular seem unable to resist filling in boxes for completion on personal data on a hard-copy questionnaire. The reluctant minority needs a considerable amount of emails, repeat forms and personal visits. The key is to keep asking.

You should expect the first reaction to gathering new statistics on diversity to be negative. Many people were outraged when firms first sought data on ethnicity. Over time everyone accepted that this is a reasonable question with a reasonable purpose. At the moment we have some shocked responses to asking about sexual orientation and religion. However, once people become accustomed to the question, they will accept this too. Taking time to explain why you want the data and what you will do with it will encourage response.

Questionnaires must allow some people to refuse to answer or to say they do not know. It is personal data and nobody has to give this information if they do not wish to or if they genuinely don’t know, as someone who is adopted might not, if asked their ethnic origin.

Respondents also need to be reassured that data will be kept confidential, with access limited to a small number of people. Much will depend upon the level of trust held in the HR team and the belief that the information requested will be used to help improve the firm’s efforts to increase diversity.

Data should be reviewed regularly to ensure it remains accurate and is used to underpin policies and actions to improve the firm’s diversity. As with many firms, we have made good progress in improving our diversity, but measuring is just the journey’s start and not its end. Earlier this year we published our diversity statistics on our website, including data on religion and sexual orientation, as a sign of transparency and openness on these issues.

The importance of diversity should not be underestimated, particularly as clients increasingly ask for statistics in pitch documents. With a war for talent in the profession, a diverse workforce is just one of many methods to ensure that all avenues are taken to attract the best people.