The Power of Plurality
September 16, 2026 | AI governance & trust, AI in pharma marketing, HCP engagement, Patient insight, agency transformation, healthcare communications
A thought piece on plurality of voice, and rewiring the advertising agency of the modern age
Executive thesis
Every wave of new technology arrives promising the same thing: speed. But our clients’ problems haven’t been getting simpler — they’ve been getting more complex, more regulated, more publicly scrutinised and ultimately, more dependent on trust that takes years to build and moments to lose. AI is simply this decade’s technology of the moment; quantum computing, or whatever comes next, will address the same underlying tension just in a different accent: not just how quickly you can produce an answer, but whose judgement stands behind it.
Health communications agency 11 London built 11 Minds, our technology platform comprising virtual on-demand ad boards, around that distinction. Clients can use the Minds directly, but they are also what now underpins how we work as an agency.
AI’s real dividend for a business like ours is not a machine that replaces expert judgement, but one that gives that judgement more to work with — more perspectives, more challenge, more voices in the room. That plurality matters because the strongest answer rarely comes from one point of view, but from several being tested against each other, with people still responsible for the final call.
This article argues that the advertising agency of the modern age has to be rewired around that principle: not with fewer people necessarily, but with people working differently, and able to draw on a breadth of perspectives no single agency team could gather on its own.
That plurality carries a commercial payoff as well as a qualitative one, and it’s not one we want to underplay. By running scenarios through dissenting voices up front and in parallel, plurality can cut out the back-and-forth built into traditional sequential workflows. Catching an objection on day one instead of month three saves real time, and real money, long before a launch date is ever at risk.
1. Taking the room with you
‘Your job, Matt, is to take the room with you,’ my Irish boss told me somewhat cryptically around twenty-five years ago. What he actually meant was that a great insight is worthless in a vacuum: success depends on bringing a whole range of people along with it and building genuine momentum behind an idea rather than just having the idea itself.
That’s still true, arguably more so today. Our new ability to do so much, so quickly, is ironically creating bottlenecks around the humans who still have to make decisions[1]. We’re all sifting through more convincing answers than ever before, thanks to the plausible sheen of AI, and so the challenge is knowing which one to trust, what matters, and what to do next. Taking the room with you has never felt harder — and a lot of that is because trust in a single voice feels like a big risk.
2. Myself I must remake, one voice at a time
The same boss kept a line from Yeats framed on his desk: ‘Myself I must remake.’ It comes from a poem about an old man refusing to accept that, as his body gives out, he has nothing left to give; instead, believing that his experienced mind is still worth something. Advertising is also a mature sector, and several of our competitors have sold up or closed down. It’s a decline we don’t accept, which means working out what our new role looks like.
Two years ago we built Jake: a virtual patient avatar — nicknamed ‘hot Jake’ internally, on account of an unreasonably handsome profile picture — whose knowledge base drew on around 800 real diabetes stories. People loved talking to him, but he never quite built trust — he felt like Zoltar, the fortune-telling machine in the Tom Hanks film Big[2]: fascinating, uncannily accurate, and hard to know what to make of it.
A single AI voice, however well-informed, is still just one voice. Jake, we realised, we must remake. That was the real starting point for 11 Minds: a curiosity about what it would take for an AI-powered voice to actually earn trust.
3. From individual brilliance to plurality
Advertising has long celebrated singular genius: the strategist with the killer insight, the creative director with the defining line. That mythology still has power, but modern brand problems have become too interconnected for any one perspective to hold the whole answer — they sit at the intersection of behaviour, data, culture, regulation, medicine and public trust.
Jake’s problem was that he came across as a lone voice, and an AI one at that. A single voice is concentrated and, by its very nature, undiversified. So the fix wasn’t a better single voice; it was more voices, working together and – crucially from a trust perspective – each accountable to real human expertise.
That’s the idea we’re calling plurality: instead of asking ‘who has the answer?’, asking ‘which combination of minds needs to be in the room?’ A strategist, a clinician, a behavioural scientist, a regulatory specialist, a patient advocate, a client stakeholder. The answer gets stronger not because consensus is easy, but because disagreement surfaces early enough to be useful.
4. The plurality dividend: necessary tension, not comfortable consensus
This is where AI becomes genuinely useful to a business like ours: it makes plurality available at every stage of the work, while origination and validation remain human. That means building tension into the workflow by design: multiple points of view, marginalised groups and dissenting voices structurally represented, red-teaming and wargaming an idea before a regulator does it for you — happening in parallel rather than in a chain of single-perspective sign-offs.
- Insight stress-tested against several expert viewpoints at once, not that of a single planner’s judgement
- Creative developed for different audiences with different needs, preferences and risk appetites before the strongest versions are chosen
- Regulatory and safety review running alongside development, not bolted on at the end
- Dissenting views logged and interrogated, not glossed over for a tidy deck
That’s what it means to bring the room with you – not as a stakeholder-management exercise, but as an operating principle. It still requires work to capture these voices with enough authenticity that a persona speaks credibly rather than generically; that’s ongoing, not a one-off build. Done well, though, these voices live with you, get better with every project, and remember in ways individual people structurally can’t — without the fatigue or unconscious drift toward whoever spoke last in the room.
This isn’t just better practice, either: it’s faster and cheaper, and that commercial case matters as much as the qualitative one. Traditional workflows tend to discover disagreement sequentially: insight, creative, medical review, regulatory, legal, then back to creative when someone finally raises the objection that was always coming, Every loop costs time and budget. Run the same scenario through dissenting voices up front and in parallel, and that same objection can surface on day one instead of month three.
The dividend of plurality isn’t only better judgement; it’s less rework, fewer stalled launches and fewer expensive surprises.
5. The things we never say
There’s a persistent gap between what people say and what actually drives them, between the good reasons we give for our choices and the quieter, real motivations underneath. Most market research is better at surfacing the former. The latter may only become visible through behaviour, actions that seem inexplicable until you understand what was going on underneath.
That gap is exactly what we’ve built the Minds’ behavioural layer to close. It lets us dial some of the main human motivations up or down — attitude to risk, social good versus personal status, and others like them — so a persona reflects more than what a patient or HCP is willing to say out loud. A well-built persona, drawing on genuine testimony rather than an averaged summary of it, can hold onto that texture in a way a single facilitated conversation often smooths away.
It’s also why the richest insight often comes from a client’s own people and data, not a purchased dataset. In practice, that means drawing on Salesforce records and MSL conversation notes, the research reports already sitting unused on a client’s SharePoint, and the patterns already buried in their CRM — rather than a third-party panel that has never met their patients or prescribers.
A purchased dataset is almost always built on stated intention, gathered at a distance; a client’s own systems have been quietly recording actual behaviour for years, and nobody else has walked those miles.
6. Rewiring the agency
AI shouldn’t sit in a side room as a production accelerator, quietly cranking out banner variants while the real decisions get made elsewhere. At 11 London, 11 Minds is now embedded in our operating model, from the first insight through approval to every version that follows.
The Minds now power a ‘go-to-market’ platform that helps our clients replicate the dissenting voices and regulatory challenges of a launch into the process earlier — when there is still time to act on them. Multiple HCP perspectives inform creative versions and channel choices across an always-on omnichannel approach. The same principle applies to clinical trials, where different voices — including those from under-represented patient groups — can inform development, recruitment and retention.
The point isn’t to make more stuff. It’s to make better calls, earlier — and to be able to see afterwards why we made them. Which perspectives changed the recommendation? What risks did we weigh? What did we miss? If that reasoning is captured, the next launch, trial or month of audience activity does not have to start from a blank page. The work develops a memory of its own.
7. Not all plurality is equal
We’re not the only ones claiming a version of this idea, so it’s worth naming the main differences.
IPG’s DXTRA Health approaches the problem through breadth: a collective of more than a dozen specialist agencies — brand strategy, PR, experiential, research, public affairs — under one holding structure, so a client can draw on several agency perspectives on a project. It’s a genuine, long-standing form of diversity, built through headcount and brand portfolio. In our language, their plurality sits across the network and is assembled fresh brief by brief; 11 Minds is exploring what happens when multiple perspectives are also built into the workflow of the work itself.
US-based EVERSANA’s AI Accelerator takes the opposite route: an integrated operating system, built with Google, Amazon, Microsoft and Veeva, stitching AI tools across medical and legal review, drug safety monitoring, omnichannel execution and market access data into a single pipeline. It’s genuinely impressive infrastructure, with case studies focused on faster reviews, higher prescription lift and lower cost per interaction. What it’s optimising for, by its own account, is speed and efficiency. Our emphasis is on what happens around the judgement itself, keeping different answers in play until a person has weighed them.
11 Minds is trying to do something narrower than either of them: not a bigger agency, a faster pipeline for its own sake or a bigger clinical workforce by proxy, but structured disagreement built into a persistent layer that any piece of work has to pass through. We’re not chasing scale or throughput as ends in themselves the way the other two are. But tension, introduced properly and up front, creates a different kind of speed: surfacing objections in parallel, on day one, instead of discovering them over months of sequential sign-off.
8. Why not build it yourself?
It’s a fair question: if plurality is this valuable, why wouldn’t a pharma company just build its own version in-house? A few reasons, and not all of them flattering to us. The risk is that any system built inside one culture can inherit some of that culture’s assumptions. External challenge still has a role, not because agencies are inherently wiser, but because we are paid to come at the problem from another angle.
Holding a room full of contradictory stakeholders — medical, legal, regulatory, commercial, creative — and finding a way through has always been part of the agency skillset, long before AI got involved. We see the pattern across dozens of clients and categories, which is exactly the kind of repeated exposure that makes a plural model better rather than merely bigger. And this repeated exposure has meant that, over the last two years, we’ve already made the mistakes so our clients now don’t have to wrestle with the best way to select data, design ontology, apply decision controls, build a validation methodology, and orchestrate between different agents and layers.
It’s also why we’re don’t see this as a pharma-only model. We’ve started trying the same approach in renewable energy. On paper, it has very little to do with diabetes. In practice, the shape of the problem looks familiar: grid operators, planners, local communities, investors and environmental groups, all with different priorities; a heavy regulatory burden; and decisions where getting it wrong is expensive and public.
Health and energy are arguably the two defining issues of our time, and neither is well served by a single confident voice, AI or otherwise.
Conclusion: the right minds, in every room
The agency of the future won’t win by becoming less human. If anything, the opposite. While AI makes it possible to produce more and faster; it does not make judgement easier. Someone still has to decide what is worth saying, who it is for, what might be missing and when a confident answer deserves another question.
That, in the end, is the case for plurality: a bigger, more argumentative room around human judgement. The intelligence worth building isn’t one all-knowing voice, but many distinct ones, willing to disagree with each other in front of us – and the art of agreeable disagreement is perhaps something that wider society needs to rediscover. In the age of AI, the modern agency’s job is still to take the room with it. Not just once, for a workshop, but always: the right minds, in every room, every time.
Postscript: Why ‘Minds’?
The name isn’t accidental. We wanted an AI platform named after fiction’s most companionable superintelligences, not its most sinister. In Iain M. Banks’ Culture novels, the ‘Minds’ are the vast artificial intelligences that run the civilisation’s ships and habitats. Instead of just one cold, monolithic superintelligence there many characters, each with a temperament, a sense of humour and a self-chosen name, disagreeing with each other and with the humans they work alongside.
It’s a useful corrective to the more familiar science-fiction image of a single AI voice deciding for everyone. 11 London’s version of that idea, built as 11 Minds, is a set of specialist personas, each traceable to real human expertise, each with a distinct point of view, working in productive and agreeable disagreement rather than converging on one averaged answer.
Matt Hunt, CEO of 11 London · Makers of 11 Minds · 11-London.com & 11-Minds.com
—
[1] https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5147196
[2] https://zoltaruk.com/big/
This content was provided by 11 London
Company Details
Latest Content from 11 London
The Power of Plurality
A thought piece on plurality of voice, and rewiring the advertising agency of the modern age
Stories of sickness: applying illness narratives to health advertising
By Dr Stephanie Goley Introduction As advertisers, we know that health and pharmaceutical campaigns hold enormous influence over how the public perceives illness, wellness, and treatment. But too often, our…
Advertising For Women’s Health: Breaking Taboos and Driving Change
Our strategist Stephanie Goley, MSc, PhD takes a fascinating look at how advertising is breaking down stigmas around areas of women’s health such as menstruation and menopause – and helping…
Rain Free Days: A Migraine App Shaped by Real Experiences
Rain Free Days® is a free migraine support app created by Teva, 11 London, and the migraine community. It helps users track symptoms, spot patterns, and access personalised resources like…
11 London and B Corp
Three years ago, 11 London became the first UK health agency to gain B-Corp accreditation. CEO and Co-Founder, Matt Hunt reports back on life in the B-List…
King’s College London and 11 London collaborate on health psychology
We’re delighted to announce that we are working in partnership with the Health Psychology department at King’s College London. Professor Rona Moss-Morris is working with the agency on a shared…
How we’re making AI a part of life at 11 London
This article talks about how AI is being used more in advertising, showing how it can change the way things are done from planning out strategies to coming up with…
Metahumans – the future of empathetic education?
Technology is moving fast, and here at 11 London we’re moving with it. Now, Metahumans (powered by Unreal Engine 5) are just one of many innovations launching in the digital…
11 London: first UK healthcare agency to be awarded B Corp status
This month – May 2022 – 11 London was awarded B Corp status.
Ad tracking: the end of the road?
As digital marketers, we’ve been closely monitoring the privacy debates between Apple, Facebook and Google over the last two years.