Two years ago, I was about to log off from my work at the research funder Wellcome when I got an e-mail from a senior leader. I remember tensing as I read it. I’d asked that Wellcome add its name to some general anti-racism commitments.

The leader didn’t refuse, but expressed doubts about the timing and capacity. Anti-racism wasn’t a priority.

I’d worked on diversity and inclusion at Wellcome and elsewhere for several years. I’d faced challenges as a woman of colour and was acutely aware of how structural racism had affected many working here or applying for our funds. The hesitation stung, but speaking up was part of my job, so I persisted. After a couple of days’ discussions, the leader agreed.

By mid-2020, people around the world were demanding drastic change after US police officer Derek Chauvin murdered George Floyd, a Black man, even as a pandemic further exposed structural racism — people from racially minoritised groups have lost lives and jobs at much higher rates than have white people.

Wellcome, like many others, made a public statement supporting the Black Lives Matter movement. Seeing a post labelled ‘Our commitment to tackling racism at Wellcome’ on our website felt surreal. It was like the door I had been pushing had fallen open.

Suddenly, I was allowed – encouraged! – to progress anti-racism work that I’d wanted to do for years. I am inspired by how millions of people protested for change, but it’s horrifying that it took so much for organisations to get to this point.

By spending £1 billion ($1.4 billion) a year to support science that improves health, Wellcome holds power. We don’t want to use it to uphold racial inequity. To help turn intentions into actions, I’ve spent the past year leading development of Wellcome’s anti-racist principles, guidance and toolkit, co-created with staff and external experts and published this week.

It’s part of a broader anti-racism programme I co-lead at Wellcome with Tunde Agbalaya that includes advising leaders, developing training and supporting colleagues. The framework boils down to five principles.

Prioritise anti-racism. Doing this work properly takes time. I stopped other work and resisted requests to complete the project faster. Sure, we could have settled with the first draft or adapted work produced at other organisations, but that wouldn’t have been as effective as something co-developed with staff and tailored to our organisation.

That said, we are providing our handbook publicly, with a licence that allows others to adapt it as they find useful.

Our principles had to work across Wellcome’s functions as a research funder, museum, library and more. After the protests of 2020, many were eager to participate. Collaborating across the organization kept staff and leaders enthusiastic. The co-creation process was crucial to the result.

Investigate racial inequity. Collecting data isn’t enough; findings must be put to work. That could include identifying issues, tailoring actions or ensuring accountability. For instance, Wellcome funds very few Black researchers.

Analysis showed that this wasn’t because of a lack of applicants – the inequity was in our funding success rates. That has informed how Wellcome designs new funding schemes and practices. Our goal was not to prove that inequity is a problem – that’s obvious – but to identify what to do about it.

Involve people of colour. Wellcome currently has an all-white executive leadership team. The lack of representation must not prevent people of colour being involved in decisions. For this programme, we wanted to include people across funding, research and Wellcome as an institution who had professional anti-racism expertise and lived experience of racism, without exploiting or harming them.

We were careful about how we established a staff forum and an external expert group: we specified the budget and workload up front, so people knew what to expect. I took on much of the work on developing the principles so that the groups could focus on advising through multiple rounds of feedback, edits and discussion.

The work is much stronger thanks to input from staff and experts. For example, we followed advice to tailor our principles to address leaders and funding committee chairs, because they have more power and accountability to make change.

Counteract racism. General diversity and inclusion initiatives rarely go far enough; targeted approaches speed up progress. For example, University College London designed its Research Opportunity Scholarships programme to support UK-resident Black, Pakistani and Bangladeshi postgraduate research students.

We prioritised on the basis of evidence of inequity. We also know that racism gets compounded by other forms of oppression, such as ableism, and so sought advice from the London-based charity Business Disability Forum.

Make measurable progress. Statements are only as good as the actions behind them. Holding leaders to account requires finding ways to measure progress. Outcomes, not intent, demonstrate whether actions are racist.

In our most recent inclusion survey, 22 per cent of staff from racially minoritised groups reported experiencing offensive language, jokes or behaviours from colleagues. Our next survey will tell us if we’re doing better.

We’ll continue to publish data on disparities in who we fund, and share anti-racism progress in our annual report.

There isn’t a quick fix. For centuries, countless people have done hard, essential work to fight racism. Now it’s our turn.

• A Nature magazine report /Opinion