How To Assess Human Rights Risks In Your Work Culture With Gavin Heryng

How did you originally get started in L&D, and how did that eventually lead into the work you’re doing now around human rights and worker welfare consulting?

I’ve been working in L&D for just over fifteen years now. I originally started in a call centre environment, and like a lot of people in learning and development, I more or less fell into it, although I often joke that I didn’t fall into L&D, I jumped into it.

At the time, I was working in a role that involved a lot of call handling and admin work. I’d ended up in this strange sweet spot where I was good enough at the job that I could train other people how to do it, but I wasn’t so essential that the business would miss me if I stopped doing the role every day. 

An opportunity came up to train some new agency workers because there’d been an increase in workload, so I took it. That then snowballed into a couple of years of internal training support work. I spent a lot of time travelling back and forth to outsourced call centre operations in places like Rotherham, training their teams, and I also spent four weeks in India with an outsourcing provider training people on the admin side of the business.

What was interesting was that after a couple of years of effectively teaching other people how to do my job, I eventually reached the point where there wasn’t really a role left for me to do. Fortunately, I wasn’t made redundant at that stage, but I was redeployed back into the call centre environment. By then though, I’d already realised that I genuinely enjoyed the training and development side much more than being on the phones.

That pushed me toward getting a Level 3 teaching qualification, and before I’d even completed it, I’d already moved out into the wider world of external L&D work. Over the following ten years or so, I worked largely with outsourcing companies and within financial services, still doing a lot of contact centre-related learning and development.

Then COVID happened and I was made redundant, like many people were. I was fortunate in the timing of what happened next. I got my notice on a Friday, applied for a new role the following day, interviewed over the next couple of weeks, and before my notice period had even ended I’d already received an offer from the organisation I work with now.

That was the first time in my career where I shifted away from teaching internal staff how to do operational jobs and into acting as an external consultant working with businesses and organisations. Over the last five years, I’ve been helping companies, including a number of large corporate businesses, identify, mitigate, and where possible eliminate human rights risks by helping them build the right knowledge, skills, behaviours, systems, and internal processes around worker welfare and human rights.

On the human rights side of your work, you’ve told me that you help companies understand what constitutes modern slavery and exploitation. Slavery can mean  different things to different people, and I imagine a lot of businesses initially assume these issues don’t relate to them at all. So what are those conversations like when you start helping companies understand human risk properly?

A lot of the initial conversations I have with clients tend to begin in a similar way. For context, some companies work with us on one-off projects, while others commit to longer-term partnerships. But regardless of the arrangement, the first thing I often hear is simply, “We need training.”

You’ll probably hear this a lot from people you speak to in L&D generally. The request always starts with training. Earlier in my career, I used to instinctively push back against that because, like many people in learning and development, I’d think, “Training isn’t always the answer.” Over time though, I’ve realised the more productive approach is to say, “Absolutely, I can help with that,” and then start unpacking the deeper issue underneath.

So the next step is understanding where the request is actually coming from. What is the business problem? Who’s the audience? What specifically are people struggling with?

For example, they might say it’s the procurement team. They’ll explain that procurement staff don’t know what signs of human rights risks to look for when onboarding suppliers, auditing suppliers, or reviewing supply chain risks. That’s useful because now we’re identifying the actual performance issue.

From there, the conversation shifts into asking what people should be doing differently that they currently aren’t doing. A lot of the process I use comes from action mapping. It’s about identifying the actions we need to see, understanding what’s preventing those actons, and then determining what solutions are actually required.

Sometimes the issue genuinely is a lack of knowledge. Procurement teams may simply not know the indicators of supplier risk. But once you address the knowledge piece, you then need to ask what they should actually do with that knowledge.

That’s where things often become more complicated. Organisations may have identified a knowledge gap, but they haven’t necessarily thought through the practical escalation process. If someone spots a concern, what happens next? Who do they report it to? What does the line manager then do? How far does the issue escalate internally? At what point does it move beyond the company into involving outside agencies or authorities?

Very often, what starts as a training request reveals missing systems, missing escalation pathways, or missing remediation processes. Clients may have thought as far as “spot a concern and tell your manager,” but they haven’t fully considered what happens after that.

That’s why these conversations quickly move beyond training and into organisational systems and operational readiness. In terms of the language itself, one thing I’ve learned over the years is that it’s often more useful to shift the conversation away from the phrase modern slavery and toward broader concepts like human rights and worker welfare. Those terms are easier for people to engage with because the word slavery comes with a huge amount of emotional weight, assumptions, and misconceptions.

We still absolutely use the term where appropriate, but we try to help companies understand that we’re not literally asking people to go out looking for “slaves.” What does a slave actually look like? Realistically, what we’re looking for are signs of vulnerability, exploitation, poor worker welfare, or human rights concerns.

That makes sense. It really is about language and framing, which affects so many aspects of business generally. Thinking more broadly about worker welfare then, how do you think these kinds of conversations can influence workplace culture positively?

A lot of the conversations I have revolve around the difference between compliance and capability. Some brands treat this work purely as another compliance requirement, alongside things like fire safety, data protection, or anti-bribery training. For some businesses, especially smaller ones with limited resources, that’s understandable. Sometimes they simply need to tick the necessary boxes first, and that’s where they are on their journey.

In those cases, we’ll absolutely provide what they need whether that’s a fifteen-minute e-learning module or a one-hour webinar. But what we always try to do is gently move organisations beyond compliance and toward capability.

The real opportunity is helping companies identify the realistic human rights risks they’re exposed to and then building genuine capability to address those risks meaningfully. Because the organisation I work for is part of a social enterprise model with a strong purpose-driven focus, the work genuinely comes from a place of impact rather than simply trying to generate revenue.

When I’m delivering awareness sessions, especially facilitator-led ones, I will sometimes deliberately lean into the emotional weight of the topic because the phrase modern slavery immediately grabs attention. It prompts people to bring their assumptions, questions, and reactions into the room.

One statistic we often reference comes from the Walk Free Foundation’s Global Slavery Index, which estimates that nearly fifty million people globally are living in situations of modern slavery. At that scale, it can feel overwhelming and impossible to influence.

So what we try to do is narrow the focus. Rather than tackling every possible form of exploitation, we focus specifically on the risks most relevant to the organisation in front of us. Often that means focusing on forced labour within supply chains.

From there, we help people understand that while they may not be able to solve modern slavery globally, they absolutely can influence worker welfare and human rights within their own sphere of responsibility.

One of my colleagues, who comes from a very strong investigative and enforcement background, often makes the point that nobody walks onto a factory site and instantly identifies “victims of slavery.” It’s not that obvious. What you’re looking for are signs of health and safety concerns, worker welfare issues, unusual behavioural changes, or vulnerabilities.

If somebody’s performance suddenly deteriorates, if conflict emerges between workers, or if there are unexplained conduct issues, there may be deeper problems underneath. Those are the kinds of indicators we help people become more aware of.

Ultimately, what we try to reinforce is that most people are already doing many of these things within their existing roles. HR teams are already monitoring people concerns. Procurement teams are already assessing supplier risk. Frontline managers are already observing team behaviors. Health and safety teams are already looking for unsafe conditions.

What we’re really trying to do is give people a clearer lens through which to view what’s already happening around them. It’s like updating your prescription at the optician. Suddenly things that were already there become much clearer and easier to recognize.”

That’s a really useful way of framing it. More generally then, and obviously it’s always situational and multi-layered, why do you think learning so often struggles to translate into real behavior change?”

One of the biggest reasons is that companies focus far too heavily on information transfer. Even going back to my earliest days in L&D, a lot of the job was simply about giving people information and hoping that knowledge alone would change behaviour. The typical response from managers, and I’m using manager here as shorthand for non-L&D stakeholders generally, is that if people aren’t performing correctly, then they must not know what to do.

Sometimes that’s true, but it’s rarely the whole picture. Even when there is a genuine knowledge gap, there are usually additional skill gaps and behavioural gaps underneath it. People may need to know how to do something, but they also need to understand how they’re expected to behave while doing it. Does the task require empathy, urgency, confidentiality, care, precision?

And even if organisations manage to define all of that properly, they often still ignore the environmental side of performance entirely. That’s one of the reasons I love action mapping so much, because it explicitly considers environment alongside knowledge, skills, and behaviours.

Environment includes systems, policies, culture, tools, physical conditions, management structures, workflows, all the things that either support or prevent performance. A lot of companies still approach performance problems backwards. They focus immediately on what employees are supposedly lacking without asking how the organization itself may be failing to support the desired performance.

Companies can start at the wrong end of the problem. They begin with the assumption that people need training instead of beginning with the performance outcome they actually need and then working backwards to understand what systems, support, and conditions are required to enable that behavior.”

Another thing I hear constantly in these conversations is the tension around technology and AI. Some people are extremely optimistic, while others are much more cautious and insist learning has to remain fundamentally human-first. What are your thoughts on AI in L&D at the moment?

People are simultaneously overthinking and underthinking AI. They’re overthinking it in terms of what they believe it can do. I use AI tools myself, but mostly as assistants. They’re useful for helping me move past a blank page or quickly reviewing work I’ve already done. But tools like ChatGPT, Copilot, and Claude are not experts.

The reason I know that is because I’m experienced enough in L&D to critically evaluate the outputs they generate. Quite often the output is poor. Sometimes it’s useful precisely because it shows me what I don’t want, which still helps move me closer to a solution.

The danger is that people without expertise often can’t critically assess the quality of what AI gives them. Most people’s experience of L&D has historically been poor learning experiences. Somebody standing at the front of a room reading slides for an hour. So if AI generates something that looks like that, people assume it must be good enough because it matches their existing expectations.

The same applies across other professions too. I’ve experimented with things like vibe coding, and on the surface it can look impressive. But because I’m not an experienced developer, I don’t necessarily know whether the code underneath is robust, secure, scalable, or full of problems.

That’s where I think people are underestimating the risks. They’re mistaking the appearance of expertise for actual expertise.

At the same time though, I think people also underestimate the genuinely useful opportunities AI provides. Used properly, it can fill admin gaps, support workflows, speed up processes, and augment existing expertise. The key thing for me is that human expertise still needs to sit on top of the system. There still needs to be someone capable of checking, validating, and critically assessing the outputs.

That balance comes up in a lot of Training Impact Gap Project interviews around AI. 

Another area I’m interested in is impact measurement, particularly the balance between quantitative and qualitative data. I like hard metrics and evidence, but coming from a copywriting background I’m also interested in storytelling and emotional resonance. How do you think organisations can strike the right balance?”

It depends entirely on what you’re trying to measure and why. Historically, L&D has always been good at measuring attendance and collecting smile sheets. We can easily count how many people attended a session, whether they enjoyed it, and whether they rated it positively afterwards.

Quantitative data is often the easiest thing to collect. If I deliver seven workshops at a certain rate, that’s easy to measure financially. If ninety-five per cent of attendees report increased confidence afterwards, that’s easy to quantify too.

But meaningful impact is much harder to capture. I like to think I can make sessions engaging and enjoyable, even when the subject matter is heavy. I’ve had people tell me after sessions that they enjoyed the workshop, despite the fact it focused on modern slavery and exploitation. That tells me they connected emotionally and intellectually with the content.

You can absolutely collect qualitative feedback through surveys, comments, reflections, and discussions. Those things are useful for understanding how people experienced the learning.

Where things become much harder is measuring sustained impact over time. Because I work primarily with external clients, once participants leave the session and return to their normal work environment, I often lose visibility of what happens next. Even when you follow up, people are busy and focused on their operational work.

One of the biggest challenges in L&D generally is that organisations don’t properly establish baseline performance data in the first place. There’s usually a vague sense that performance isn’t where it should be, but no clear measurement of what the actual problem is, how much it’s costing, or what success would look like afterwards. Without that baseline, it becomes difficult to properly evaluate whether a learning intervention genuinely improved anything.

So for me, one of the biggest issues is that companies jump straight into solutions without first defining the performance problem clearly enough to measure meaningful impact later.

Are there wider trends or patterns you’re seeing in L&D right now that you’re either hopeful or concerned about beyond AI specifically?

I think  a lot of L&D trends are essentially the same problems with a new coat of paint. AI is the current thing everyone believes will solve everything, but I’ve been around long enough to remember when e-learning was supposed to solve everything, and when coaching was supposed to solve everything.

The technology changes, but the underlying problems often remain the same. I do think AI is going to fundamentally reshape a lot of jobs and workflows over the next few years, especially as automation becomes more capable of handling routine tasks. Virtual avatars, AI-generated video and automated systems are all improving rapidly. 

But the core issue still remains the same: the people using these tools often don’t fully understand human performance in the first place. No matter how advanced the tools become, businesses still need to understand what behaviours they need from people, what systems enable those behaviours, and what environments support performance. At the moment, AI outputs can often look impressive at first glance, but once you start interrogating them properly, the cracks begin to show.

One thing I do find hopeful though and this is entirely anecdotal is that I’m seeing brands asking for more human interaction again. More clients want in-person sessions and are asking for interactive and engaging learning experiences. 

Because while everybody now has access to off-the-shelf e-learning and AI-generated content, what’s becoming increasingly rare is genuine human conversation, connection, and reflection. People still want someone to ask thoughtful questions, challenge assumptions, and help them think more deeply.