“We’re Fighting Against the Tide”

Ask a room full of senior benefits leaders whether they have governance under control and the honest answer is: not really.

At a recent Ben roundtable, heads of benefits from some of the world's largest employers compared notes on how global benefits programmes are actually governed. Not how they should be. How they are.

“I’m glad people have said it’s a messy mix. I don’t feel so alone.”  Nobody in the room disagreed.

Leadership wants cost control and consistency. Local markets want autonomy. Central teams sit in between, trying to build frameworks inside organisations that were never designed with this kind of discipline in mind.

For most companies, the gap between how governance is described in strategy documents and how it plays out on the ground is significant. This piece draws on that conversation, and the collective experience of the people in it, to explore what good governance actually looks like in practice.

1. The governance gap is more common than you think

“Culturally, the organisation had zero governance in the past. We’re fighting against the tide.”

The gap between governance intent and governance reality is widespread. Acknowledging it is the starting point.

One participant described inheriting a global benefits function at a major employer, only to find that the central benefits function was around three years old — despite the business operating across dozens of markets for decades. Another arrived to discover that local managers had been expensing their own private insurance on corporate cards because they didn’t rate the company plan. A third had joined an organisation where, historically, nobody asked what a benefit cost before putting it in place.

“I asked my VP when I first joined: is cost something you look at historically? And it would just be, ‘We didn’t ask what the cost was. We just put it in place.’ It blew my mind that was even possible.”  

These aren’t stories about poorly run companies. They’re the normal state of affairs in large, fast-moving organisations, where benefits decisions tend to sit with whoever has the budget, the urgency or the strongest opinion.

The gap is widespread. Naming it is step one.

2. Good governance isn’t about control: it’s about better decisions

“It really has to be a value-added process. If it’s a bureaucratic tick-box exercise, that’s a failure.”

What does good governance actually look like in practice? The most useful definition that emerged: a decision-making process that gets the right people engaged, at the right level, for the right type of change. 

Not every decision needs board-level sign-off. But every decision should follow a clear path: one that accounts for cost, legal exposure, payroll implications, systems updates and the employee communications that follow any benefit change.

“Everyone needs to know who the decision-maker is. People jump straight to solutions instead of thinking through what they’re actually trying to solve: what’s the budget, what’s the fairness question, what does equity look like across markets? All of that comes together in the decision-making process.”

There are two ways to think about governance. One is control — catching things before they go wrong. The other is quality assurance — making better decisions in the first place. The second is more useful. 

Benefits are sticky. In some markets they become contractual, so you can’t switch them on and off. That’s precisely why the decision-making process needs to be robust: not because someone is watching, but because the cost of getting it wrong compounds over time.

3. Philosophy first, governance second

“It’s the destination and then the route. One is where you’re trying to get to. The other is how you get there.”

Many organisations build a governance framework before agreeing on a benefits philosophy. And that’s a mistake.

Without clarity on what the organisation stands for — its global minimum standards, approach to equity across markets and view on competitive positioning — governance has nothing to govern towards. You end up policing decisions without a shared view of what a good outcome looks like.

One participant described the approach they’d taken at a large global employer: before introducing any formal approval process, they developed a clear benefits philosophy and set of global minimum standards first. That gave country teams the tools to make decisions locally, and gave the central function a principled basis for pushing back when something fell outside the framework.

“What do we stand for? What is the consistency in how we apply our framework? If you’re coming forward to ask for things, this is how you need to ask for it.”  

Philosophy and governance are related, but they’re not the same thing. One defines the destination. The other maps the route.

4. Most benefit decisions don’t start with a business case. They should.

“It’s not what the people need. It’s: this is what the market is doing, so we need to do this.”

In many organisations, benefit decisions are reactive. A market benchmark lands, a competitor launches something, a country manager wants to drive engagement. The response is to move, often quickly, without asking whether the proposed change is solving the right problem.

“The local HR teams don’t even know all the benefits they have. That’s the place to start.”  

In some cases, participants found direct duplication: the same benefit provided through two different routes, with no one locally aware of it

One of the most significant governance shifts described was repositioning the global benefits function from a policing role to an advisory one — helping local teams build the business case for changes, rather than waiting to approve or reject what came through. That shift, one participant reflected, transformed the dynamic entirely. Local teams stopped avoiding the centre, and started using it instead.

The practical implication: building a business case requirement into the governance process from day one raises the quality of every decision that flows through it. It also changes the question from “can we do this?” to “should we do this, and why?”

5. Stop trying to prove ROI. Start proving value.

“I’d move away from ROI and shift to return on value — because then you start with: what does the business actually want to achieve?”

The ROI conversation is one that benefits leaders have been losing for years. The problem isn’t the answer: it’s the frame.

“What’s the ROI on life insurance? I’ll be a bit challenged to answer that.”  Traditional ROI works when you can isolate inputs and outputs. Benefits don’t work that way. There are too many variables, people and factors outside anyone’s control.

The more useful frame, the group concluded, is return on value — starting with the business outcomes the organisation actually cares about and working backwards. What’s the business trying to achieve? Attract better talent? Reduce attrition? Improve time to productivity? Once those goals are defined, benefits can be evaluated against them.

“We’re meant to be facilitating the business achieving its goals, not doing something for its own sake within benefits or HR.”  

The metrics that came up included attrition rates, exit survey data, talent attraction cost, wellbeing and absence figures, and time to productivity. None of these are perfect. But all of them are more persuasive to leadership than a theoretical cost-return calculation, because they connect what the benefits team does to outcomes the business actually cares about.

6. Where budget sits changes everything

“I’ve seen budgets sit locally. They then believe they can spend whatever they want.”

You can build a good governance framework, but if the people who own the budget don't feel accountable to it, the framework doesn't work.

This came up repeatedly as a structural issue. When budget sits with a country GM or regional VP, the local view often becomes: ‘It’s my money, I’ll decide how it’s spent’. The central team is left steering without financial control, asking for adherence to a process that has no real teeth.

At the same time, leadership is demanding cost optimisation and proof of value. Organisations want central visibility without centralising budget. That’s a difficult position to manage, and the tension rarely resolves without a deliberate decision about where accountability sits.

“A piece of the global governance structure needs to be how you’re choosing vendors and what you’re requiring from them. Not telling them who to use, but making sure the partnerships you have give you the data to evaluate what’s actually working.”  

7. A small central team can govern a large global programme

“The people at the centre are very incentivised to keep you happy because they’ve got millions in fees at risk. Whereas the local broker in a small market doesn’t give a stuff about it.”

Centralising governance doesn’t require centralising headcount. Several participants had found a different route: centralising the relationships and accountability at the top of the supply chain.

The logic is straightforward. A local insurer managing a small premium in a minor market has minimal incentive to enforce anything. A global broker or network insurer managing a significant multi-market relationship does. Put governance requirements — data reporting standards, API capabilities, benefit change approval protocols — at the central level, and let them cascade to local partners.

“The only way to punch above your weight is to work with a strong partner at the centre who then cascades requirements out to their local partners.”

Others described similar structures. One had built a contractual requirement into their global broker arrangement: no benefit change could proceed without sign-off from the central benefits team. “They make sure it has to be an agreement from myself or my team before any benefit change goes through.”  The broker became part of the governance structure — not an overhead, but an enforcement mechanism.

The principle extends to vendor selection more broadly. Requiring API integration, aggregated reporting, and clear SLAs as conditions of partnership puts governance into procurement before a vendor is ever onboarded.

The harder question sits underneath all of this. Most benefits teams are trying to govern programmes their technology can't actually see into. Without aggregated data across markets, providers and benefit types, governance is a conversation about what people think is happening, not what’s actually happening. That's the real ceiling. Frameworks help. Philosophy helps. But the limit on what any central team can govern is the limit of what its infrastructure shows it.

Recommended actions

The conversation didn’t end with a neat set of conclusions. That was never the point. What it surfaced was something more useful: an honest picture of where benefits leaders are actually struggling, and what the people who’ve made progress have done differently.

“Is anyone happy with their approach, or would they change to another given the choice?”  one participant asked towards the end of the session.

“Is it perfect? Probably not. But we’re broadly grinding in the right direction.”

The phrase ‘grinding in the right direction’ captures something real about the state of benefits governance. Most organisations aren’t starting from a position of strength: they’re catching up, often without enough resource, inside cultures that weren’t built for this kind of discipline.

The clearest takeaway from the conversation wasn’t a checklist, but a sequence. Get clear on philosophy before you build a framework. Build the business case requirement in from the start. Centralise relationships rather than headcount. Shift the conversation from ROI to return on value. And be honest — with leadership, and with yourself — about where the organisation actually sits on the maturity curve.

“The don’t-know-what-you-don’t-know territory is the hard stuff.”  But that’s where you have to start.”