Tags:
InsightsYou've got new markets on the roadmap. Or a funding round has landed and the question of how to run marketing across multiple countries is suddenly urgent. Maybe the marketing model you've been running has started to crack, and you need to decide whether to tighten central control or give local teams more room. It's a decision most global marketing leaders face more than once, and most will tell you there is no clean answer.
Across 35 years of global campaigns, I've seen both work and I've seen both fail. What you are really choosing is which trade-offs you are prepared to accept and what governance is needed to compensate.
The case for centralisation, and what it costs
The logic for choosing centralisation is straightforward. One brief, one set of brand standards, one production process. Assets produced centrally cost less per unit at scale, launch faster across markets, and arrive in-market with consistent messaging. For organisations running high volumes across many markets, the efficiency and brand consistency arguments are both real.
The trade-off is local visibility. Centralisation deprioritises local insight, local creativity, and the depth of market knowledge that sits closest to each audience. You run the risk that an asset or entire campaign won't resonate with the audience receiving it, that a local activation won't be right for its context, or that a claim that is legal in the originating country won't clear the bar set by a local market's regulator.
Revolut learned this in Italy in March 2026. The brand picked up an €11.5 million fine from the Italian consumer authority for advertising investment products as zero commission without making the full cost picture clear enough to Italian consumers. eToro had been fined by the same regulator in 2023 for almost exactly the same thing.
Centralisation deprioritises local knowledge. That is not a reason to avoid the model, but you do need to build the governance that compensates for the absence of local knowledge.
The case for decentralisation, and what it costs
Decentralising prioritises what centralisation trades away: local insight, cultural fluency, and the market speed that comes from teams who know their audiences without needing a brief to travel from HQ and back.
The trade-off is oversight. Decentralised models deprioritise the central visibility that catches decisions which should never have been made at local level: interpretations of the brand that contradict global positioning, regulatory risks that were not escalated, activations that might have made sense in a local market but didn't consider the impact of the work on a wider audience.
In May 2026, Starbucks Korea launched a promotion on the anniversary of the Gwangju Uprising, when tanks were deployed to violently suppress pro-democracy protesters. The local operator used the date to promote discounts on its Tank tumbler range with a tagline referencing the event. Within 12 hours, the local CEO had been fired, the South Korean president had denounced the brand publicly, and a new Korean verb, "tal-buck", meaning to quit Starbucks entirely, was spreading across the internet. Nobody at global HQ signed off on this. It was the local licensee's call.
Decentralisation spreads the responsibility for brand consistency between the central team and local market teams. To protect the brand globally, it is crucial to build governance with checks for the kind of risks local teams won't necessarily be aware of.
AI has accelerated both models and their failure modes
In 2026, any brand making this decision is in a vastly different context to five, even three years ago. AI is dramatically reducing production time. Translation across languages, locally relevant imagery without a photo shoot, voiceover without a recording studio; these are all capabilities that once required local production infrastructure. Increasingly, they can be handled efficiently without leaving the office, be that at HQ or in a local market.
What AI hasn't done is compress the time it takes to approve, review and quality-check that output. The constraint used to sit in production. Now it sits in the review layer, where thousands of assets can arrive against windows that were built for a fraction of the volume, leaving too little time to assess each one properly.
AI isn't creating new problems in centralised or decentralised models, but accelerating problems that already exist. When you can activate across 20 markets in 48 hours, the risk that caught out Revolut becomes even harder to prevent. When local teams have tools that let them produce and publish faster than ever, the oversight gap that produced the Starbucks Korea situation becomes easier to fall into without anyone upstream noticing.
This makes the governance question more urgent than ever.
What does good campaign governance look like in practice?
Governance is the structure of decision rights, accountability and review that controls how marketing work moves through an organisation and out to market. Whether you run a centralised or a decentralised model, it is the governance underneath that determines whether the model holds together. It is something we prioritise at Freedman and help global teams set up from the ground up.
In my experience, three things show up consistently in the organisations that hold quality and control as they scale, regardless of where they sit on the centralised/decentralised spectrum. The key is to compensate for each model's blind spots.
A single point of accountability.
One person, function, or agency with genuine authority across global, regional, and local stakeholders, who can make calls and is accountable for the outcome. 66% of marketers using 11 or more digital channels say maintaining brand consistency is a challenge. At that scale, consistency does not hold through goodwill. It holds through a clear owner.
Real regulatory and cultural review, before go-live.
Not a final-stage check, but a standing process that maps regulatory exposure by market at the start of every campaign, with local insight at an early stage to save rework down the line. This is the structure that closes the centralised model's blind spot.
As volume rises, this is also where review has to get smarter: not everything needs the same level of human attention, and the teams that hold quality at scale reserve it for the work where cultural judgement, regulatory exposure and brand significance actually change the outcome.
Documented decision rights.
This means defining what local teams can act on independently and what requires escalation. It is what closes the decentralised model's blind spot. Each part of the model needs to know what it is authorised to do. A local operator shouldn't be able to make a call that would damage the global brand without central sign-off and vice versa, a central team shouldn't be able to move forward without meaningful input from a local team.
Key takeaway
Centralised and decentralised marketing can both scale well across markets, but both can also fail in predictable and largely preventable ways. Centralised models deprioritise local knowledge and create the regulatory and cultural risk that only in-market visibility can catch. Decentralised models deprioritise central oversight and create the conditions for local teams to make calls that should have been escalated. Choose the model that suits your setup and your culture. But make sure you have a plan for the governance layer underneath that compensates for the necessary trade-offs of each model.
Frequently asked questions
What should a global campaign operating model look like?
There is no single right answer, but there is a consistent wrong one: an operating model designed for a smaller scale than the campaign now demands. Whether you run a centralised model or a decentralised/ hub-and-spoke model, it needs to define who owns delivery end to end, how regulatory and cultural review sits in the production chain, and what local teams are authorised to decide independently. The brands that scale international marketing well are not the ones that picked the right model. They are the ones that built the governance to catch what their model could not see.
What does good campaign governance look like for a global brand?
Good governance means clear accountability at each stage of production, not just at sign-off. A single delivery lead with genuine authority across global, regional, and local stakeholders. Regulatory and cultural review built into the production chain before assets go live, not after the brief is locked. And documented decision rights that define what local teams can act on independently and what requires escalation. Without those three things, what holds campaigns together is individual judgement in individual markets, which is not sustainable.
At what point should a scaling brand move from a centralised to a decentralised model?
The trigger is usually one of three things: local markets are pushing back consistently on centrally produced work because it is not landing with their audiences; regulatory or cultural failures are appearing in markets where the central team lacked the local knowledge to anticipate them; or the volume of market-specific work has grown beyond what the central team can produce at the required quality and speed. The move to decentralisation is not primarily about giving local teams more autonomy. It is about building local knowledge into the production chain at the point where it is most useful, while keeping the governance structures that hold the operation coherent.
