I first started working in the anti- slavery space on a counter trafficking programme in Timor-Leste. We were implementing a programme that looked very similar to counter trafficking programming around the world , which included providing technical advice to the government to develop counter trafficking legislation and a national action plan, support to NGOs to improve delivery of services, and training for police to better investigate crimes. One of my roles was tracking our impact, and while we were successful in achieving all our outputs, the answer to the question of whether we had reduced trafficking remained somewhat elusive.

Fast forward a few years, and to a new job at the Walk Free Foundation. When I was asked by a colleague, “what works”?, I had to pause. Could I hand on heart say that the counter-trafficking response was effective? What about the community empowerment approaches of the anti-slavery movement? Is there another, ‘easier’ way to approach this?

This basic question – what works? – became the cornerstone of a project called ‘promising practices’. We know that evaluation is key to understanding the impact and effectiveness of projects designed to prevent or address the harm connected with modern slavery. Systematic reviews of some trafficking programming had taken place, but there was not an attempt to really dive into that question of “what do we know works”, and just as important, “what do we know doesn’t work”?

In an effort to increase the effectiveness and efficiency of projects and programmes combating modern slavery, we created the Promising Practices Database. The Database collates impact and programmatic evaluations of anti-slavery and counter trafficking programming, and categorises these evaluations according to country, region, type of modern slavery, methodology, intervention, and whether the evaluation met its objectives or outcomes. It was developed so that project developers, researchers, and academics could quickly identify relevant evaluation work that had already been undertaken, and also seek to better understand what works – and what does not— through a simple search by country, target population, type or sector of slavery, or type of intervention. The theory is that we can learn from the evaluations already undertaken, even if that learning is that we need to rethink how to combat modern slavery.

To date, the Database contains 179 evaluations from the modern slavery and associated sectors. Accompanying the database is an overview paper, which draws out some descriptive statistics and some general conclusions about the current state of monitoring and evaluation in the anti- slavery and counter trafficking field. Following this are a set of policy papers that dive into different sectors or interventions targeting modern slavery, where we have made observations about which practices have been proven to work, which practices look promising, and which practices are ineffective. To date, we have policy papers on interventions relating to case management, and draft papers on labour inspections, and the impact of conditional and non-conditional cash transfers on child labour and child marriage. There are many more of these papers to come, while we also plan to incorporate the database into the Global Slavery Index government response database and link closely to the work of the University of Nottingham and other partners on monitoring and evaluation.

So, does the Database reflect what works? And if so, what is the answer? Somewhat naively, when we started this work I thought the truth was out there and once we had unlocked the information in these evaluations we would identify the panacea to combat modern slavery. The truth, unsurprisingly, is more nuanced – modern slavery encompasses many forms of exploitation, each of which requires a different response. However, all is not lost- community empowerment models look promising when preventing slavery and similar forms more akin to harmful traditional practices. Strengthening and enforcing the rule of law is promising when tackling the role of organised crime in human trafficking, and when ‘legal norms’ (or respect for the law) already exists. Businesses, and empowering consumers, also have a role to play in demanding better labour conditions for workers. The artificial intelligence revolution suggests that universal basic incomes could be helpful to protect those vulnerable to exploitation. Rather than a panacea, we should be thinking of a suite of responses.