What experience makers can learn from Oxy Insight’s Barbican Immersive case study – and why research works best when it is built into the process, not bolted on at the end.

Speaker: Sonia Joao and Sam Rogers
Campfire: Using Audience Insight for Impact: A Barbican Immersive Case Study
Date: 20 May 2026
Format: Case study
Core question: How can audience insight shape an immersive experience without diluting the creative vision?

Intro: Making your experience work harder

On Tuesday 20 May, WXO members gathered on Zoom for Oxy Insight’s Sonia Joao and Sam Rogers’ Campfire, Using Audience Insight For Impact: A Barbican Immersive Case Study. The session promised something practical: how audience research helped shape Barbican Immersive’s strategy, proposition, communications and show development. It delivered exactly that.

This was not a Campfire about insight in the abstract. It was not a defence of research as a corporate hygiene factor. It was a detailed walk-through of how Oxy Insight worked with Barbican Immersive across a multi-stage research programme involving:

• Semiotics
• Accompanied immersions
• Focus groups
• Omnibus research
• A targeted quantitative survey
• A longitudinal audience panel

The original listing for the session asked: “What if you could use audience insight to make better creative decisions, and see the results in visitor numbers, ticket sales, and funding?” That is the commercial promise. But the more useful point came early in the session, when Sonia Joao made a necessary clarification.

Audience insight is not the whole of experience-making. It is one part of a larger process involving creative teams, production teams, business teams and venue teams. It should not be used to flatten creative ambition or outsource judgement to a survey.

Or, as Joao put it:

“It’s really important to say that we understand that audience insight is just one part of what makes amazing experiences. There are phenomenal creatives, production teams, people that are involved in the business side of productions as well. And we know that audience insight is just one little bit of that, and we’re really respectful of that.”

Then came the line that framed the whole session:

“Sometimes there’s a worry that audience insight comes along to green light or red light something or to say if something’s gonna work or not. And it really isn’t that. It’s about working together with teams, and just bringing in the audience and the target audience’s voice.”

That was the centre of this Campfire. Research is not there to decide whether art is allowed to exist. It is there to help teams understand who they are for, what expectations they are creating, how people behave across the journey, and where the experience can work harder.

The brief: make Barbican Immersive audience-centred

Joao and her colleague Sam Rogers were careful to place the work in context. Oxy Insight had worked with the Barbican for around three years, across brand positioning, audience understanding and exhibition development. This particular programme focused on Barbican Immersive, which had been identified as a key pillar in the Barbican’s future audience and programme strategy.

The brief had three main questions.

• First, how should Barbican Immersive show up as a brand proposition in the immersive space, both in the UK and internationally?
• Second, who is the Gen Z audience for immersive experiences? Not just the audience already attending, but the potential audience too.
• Third, how could insight inform show development, positioning, marketing and ancillary services?

This is already a useful lesson for experience leaders. The research was not commissioned to answer one narrow executional question. It was designed to support strategy and development across the full chain: audience, brand, programme, show, marketing and venue experience.

The work was conducted in the UK, with a subset in Japan. That international dimension mattered because some codes, meanings and expectations travelled across markets, while others did not. Joao later gave the example of the “sublime” as a semiotic territory that resonated in both the UK and Japan, while psychedelic or drug-culture codes worked differently.

The scale of the study was also important. Joao described it as unusually comprehensive.

“This was a big study,” she said. “Not all studies are this comprehensive. This literally covered everything. Typically, we might just do a semiotic piece or we might just do qual or just quant, but this was a real good one for comprehensive view of the opportunity.”

That matters because the point is not that every experience project needs this level of research. Most will not. The point is that different research methods answer different questions. The trick is knowing which question you are asking, when you are asking it, and what decision the answer is meant to inform.

The method was the message

Surprisingly, one of the strongest features of the Campfire was its structure. Joao and Rogers did not simply list their impressive methodology. They explained the logic of the sequence.

The programme moved from macro to micro, then from exploration to quantification, then into development.

It began with semiotic mapping to understand the cultural codes around immersive experiences and adjacent sectors. It then moved into accompanied immersions, watching real audience members attend real immersive experiences. Focus groups then explored drivers, barriers, occasions and positioning. Omnibus research sized the potential audience. A targeted survey quantified behaviours, motivations and attitudes. Finally, a longitudinal panel followed target audience members through the development process.

Rogers described the value of this cumulative model clearly:

“There are multiple stages, but they built on each other. So each stage built on those that came before and then informed the subsequent stages. So very much cumulative learning with our clients and our stakeholders involved all the way through.”

That is the real lesson. Research should not be a disconnected report that lands in the inbox. It should be a learning system. Each stage should make the next stage sharper.

There is a practical discipline here.

• Semiotics without audience research risks being clever but untested.
• Qualitative research without quantification risks being rich but hard to prioritise.
• Quantitative research without prior qualitative work risks measuring the wrong things.
• Longitudinal research without earlier segmentation risks following the wrong people.

The Barbican Immersive programme worked because the methods were not competing with each other. They were doing different jobs.

Semiotics: start with the culture, not the product

The first stage was semiotic mapping.

Joao gave a plain explanation of semiotics for those less familiar with the field. Semiotics studies signs and symbols that convey meaning through cultural association, not literal meaning. A colour, word, sound, image or object can all carry meaning depending on cultural context.

She used simple examples. Red in the UK often signals danger; in Japan, it may suggest fortune and good luck. A red rose commonly signals romance. The alternating notes associated with Jaws can create dread before anything is seen.

“Tiny little signs and symbols can really convey powerful meaning,” Joao said. “That’s what semiotics is about.”

For Barbican Immersive, the team explored the visual signs at play in the immersive sector. But they did not stop there. They looked at adjacent spaces too, including wellbeing, entertainment and gaming, because audiences do not experience culture in neat sector boxes. They move between categories. Their expectations are built across all of them.

The output was an impressive set of semiotic territories: cultural codes that could inform the proposition, visual identity, communications, language and tone of voice.

Joao brought one code to life: the sublime. It has roots in Romanticism, with individuals silhouetted against vast landscapes. It communicates vertiginous infinity, smallness, awe, floating and the scale of human experience against something larger.

“You can see how also these codes play out over centuries, literally,” she said, “which is why they’re so, so powerful.”

But the value of the research was not simply to identify the code. It was to test its usefulness. The sublime had emotional resonance, including in Japan, where examples such as teamLab also play with the relationship between silhouette and expanse. But there was a watch-out. The code did not always feel actively participatory, and audiences often want immersive experiences to give them something to do, not just something to behold.

This is a critical point. Semiotics can help a team understand what meaning a visual world carries. But meaning still has to be connected to behaviour. Awe may be powerful, but if the audience wants agency, awe alone may not be enough.

Joao gave another cross-market example. Drug trip, psychedelia and drug culture may resonate in the UK, but not in Japan in the same way. The implication is simple: do not assume your visual codes travel just because your format travels.

For experience businesses planning international expansion, this is not decorative detail. It affects proposition, campaign language, visual identity and expectations. The wrong code can make an experience feel dated, passive, juvenile, alienating or culturally off-key.

Accompanied immersions: watch what people actually do

The second stage moved from cultural codes to real behaviour.

Oxy Insight recruited 16–34-year-old immersive event attendees and spent four-hour accompanied sessions with them, visiting two immersive experiences. The approach was semi-ethnographic: observe people in context, then ask relaxed but detailed questions before, during and after the visit.

Joao described the point of this stage as “granular insight”. It captured the audience journey from pre-visit behaviour through every touchpoint to post-visit impact.

This is where research gets close to the truth of an experience. People are not always good at explaining what matters to them in the abstract. They are often much better observed in motion. Examples include:

• Arriving
• Queueing
• Scanning signage
• Deciding whether to read text
• Checking phones
• Following friends
• Hesitating at thresholds
• Getting overloaded
• Missing cues
• Finding delight

One of the most useful findings concerned the transition from the outside world into the experience.

Joao explained:

“What was so interesting was how important the moment of transition is from leaving the high street. Just that moment of going in is really important and actually needs to signal that there’s a transformation happening – and they want it to be really involving.”

That is a deceptively important point. The first threshold is not logistics. It is dramaturgy.

In a good experience, the transition starts before the core content. It tells the body and mind that the rules are changing. It helps the audience leave one mode and enter another.

Joao contrasted two examples. Jeff Wayne’s The War of the Worlds: The Immersive Experience, when open in London, handled this well. Huge doors opened ominously and excitingly into a space that felt aligned with the mood of the experience. By contrast, Refik Anadol’s Echoes of the Earth: Living Archive at the Serpentine Gallery in London was described as phenomenal in itself, but the transition from outside to inside felt abrupt. Visitors were immediately met with a large amount of wall text. They were not allowed to ease into the immersive state.

That is not a criticism of information. It is a reminder that information has timing. Wall text can help understanding and still damage the emotional handover if it appears at the wrong moment.

The Oxy Insight deck (embedded throughout this article) summarised several findings from accompanied immersions:

• The importance of the transformational moment
• A compelling and logical narrative and emotional arc
• The flow of sensory load
• The impact of wall text, tech and production quality.

The importance of the transformational moment, a compelling and logical narrative and emotional arc, the flow of sensory load, and the impact of wall text, tech and production quality.

These are practical design issues. They are also easy to miss from inside a production team, where everyone already knows the story, the space, the technology and the intent. Audience observation exposes the gap between what creators think they have signalled and what visitors actually receive.

Focus groups: why do people seek immersive experiences?

The third stage used focus groups to go deeper into motivations, barriers, journeys, occasions, brand positioning, white space and show concepts.

This is where the programme produced one of its most valuable outputs: a framework of core drivers within immersive experiences.

Joao explained that the team drew on developmental drivers from neuroscience: the things people seek out in order to grow and develop. They tend to sit across four areas:

• Physical and sensory: the desire to engage through sight, touch, sound, bodily awareness, vibration, movement, space and sensation.
• Communal: the social dimension of attending alone or together, exploring identity, connection, relationships and shared experience.
• Cognitive: the desire to learn, be challenged, reconsider assumptions or have perspectives overturned.
• Emotional: the desire to feel something about oneself, others or the world.

Through the qualitative work, Oxy Insight found a strong crossover between those developmental drivers and the reasons people seek immersive experiences. Joao used the physical and cognitive drivers to illustrate the range:

“Physical and sensory speaks to that desire for experiences that really allow you to physically or sensorially engage with whatever is going on about you. So whether that’s about hearing or feeling or touching or sensing vibrations, but it could also be about being aware of your position within a space.”

At the other end:

“A cognitive driver is where you are seeking out experiences that feed your desire to be cognitively challenged or your perspectives to be overturned or to cause you to review attitudes.”

This framework is useful because it avoids a common problem in immersive work: collapsing all motivation into “people want to be immersed”. That is too blunt. People may want different things from the same form. One visitor wants awe in the body. Another wants to learn. Another wants to connect with a friend. Another wants emotional release. Another wants something to post on social media, though that was not the focus of this particular framework.

The job of an experience team is not necessarily to serve every driver equally. It is to understand which drivers matter most for the proposition, audience and moment.

A show can be physically powerful but cognitively thin. It can be emotionally moving but socially awkward. It can be intellectually strong but sensorially underwhelming. It can be communal in intent but confusing in practice.

The above framework gives teams language to discuss those trade-offs.

Quant and qual: together at last

The fourth and fifth stages moved from qualitative richness into quantitative confidence.

First came a larger omnibus survey: a scheduled nationally representative omnibus survey of more than 2,000 UK adults. Rogers explained the value clearly. An omnibus survey works on a cost-per-question model and can provide a relatively efficient way to size audiences and profile them at market level.

For this project, the team used it to size the existing and potential immersive audience. They defined the audience as people who had attended at least one immersive event in the last couple of years at a relevant venue and were open to doing so again.

The phrase “relevant venue” is important. Rogers explained that the immersive category is wide (thus reaching familiar territory to Campfire veterans!), so the definition had to be “Barbican-shaped”. They focused on experiences in museums, galleries, event spaces and similar contexts – places that reflected what Barbican Immersive could realistically offer.

“We want to speak to people whose expectations fit with what the Barbican can realistically put on themselves,” Rogers said.

That is a lesson in research design. The audience you measure should match the decision you need to make. A broad immersive audience might be interesting, but not always useful. A Barbican-shaped immersive audience gives sharper answers.

The omnibus research showed a significant 18–34 audience, which aligned with expectations from the qualitative work. But it also showed a meaningful 35–54 audience, giving the team useful context for planning.

Then came a more targeted quantitative survey: a separate 15-minute online survey with around 500 people from the core immersive audience.

Rogers explained why this stage mattered. The omnibus was broad. This survey went deeper into immersive audience needs, wants, behaviours and motivations. Crucially, the team did not have to guess what to ask. The qualitative work had already produced the driver framework and audience language.

“We didn’t have to guess what to put in that survey. We had this wonderful framework that was already built from the exploration and all that learning. We could build statements and structured questions that spoke directly to those from the qual and put them in the survey using the language of our respondents as well.”

That is the best argument for blending qual and quant. Qualitative research helps you understand the world. Quantitative research helps you judge scale and priority. If you skip the first, the second can become precise nonsense.

The survey quantified the importance of different drivers and needs. At the top sat things like having fun, quality of audio and visuals, and seeing something new and innovative. But the list also included feeling inspired, learning something and feeling connected with others.

This gave Barbican Immersive a hierarchy of factors to consider when making programming, positioning and investment decisions. It also supported practical questions: how people find out about immersive events, how and when they book, what add-ons or enhancements appeal, and how price sensitive they are.

That range matters. Good audience research should be able to answer big strategic questions and small operational ones. In the real world, both matter. A proposition can be right and still fail if the booking journey, pricing, copy or pre-visit communication is wrong.

Longitudinal panels: keep the audience in the room

The final stage was a longitudinal panel, still ongoing at the time of the Campfire.

Oxy Insight recruited some of the strongest respondents from the earlier stages: people who were representative, useful for the Barbican’s target audience and able to provide good feedback. They were then reconvened roughly every two or three months through online focus groups.

Each round had a specific purpose. Some explored show concept refinement. Others looked at naming, introductory website copy, marketing strategy, comms, exhibition experience and later post-launch reflection.

This is where audience insight becomes a feedback loop rather than a one-off diagnosis.

Joao explained that the team could use the panel “exactly when it would have the best impact and fit with their development process.”

That timing is crucial. Research is far more useful when it meets a live decision. Too early, and it can be vague. Too late, and it becomes a post-rationalisation exercise. The longitudinal panel allowed insight to arrive at the points where it could affect naming, messaging, concept and experience development.

It also brought different stakeholders into contact with audience needs and wants. That is an underrated benefit of research. Sometimes the output is not just the findings. It is the shared moment of watching, hearing and discussing audience response together.

Impact: insight that did not sit locked in a report

Towards the end of a well-attended Campfire, Joao shared how the research had affected Barbican Immersive.

According to the deck, the work contributed to a common language across the team, a coherent framework for programming decisions, and a shared understanding of who Barbican Immersive is for. It was also connected to tangible performance around Feel the Sound in 2025:

• Meta conversions 303% versus target
• 42% first-time Barbican bookers
• 55% of web visitors aged 18–35
• 17,947 new mailing list subscribers

The numbers are useful. But the less visible impacts may be more important in the long run.

Joao said the research gave the team “a common language” with “a real audience-centred focus”. It gave them “a coherent framework for understanding and making programming decisions”. And it gave them “a really important shared understanding of who they’re here for, why they’re here, and what they’re here to give their audiences.”

That is often where good research creates organisational value. It does not simply answer a question. It changes the quality of internal conversation. It gives creative, marketing, programming, venue and leadership teams a shared map.

In experience organisations, this is especially valuable because decisions are often cross-functional. A show development choice may affect marketing. A marketing promise may affect visitor expectation. A visitor expectation may affect front-of-house. A front-of-house issue may affect post-visit memory. Audience insight can connect those decisions before the audience is in the building.

What experience leaders should take from this

The Campfire was practical because it did not sell one magic method. It showed how different methods work at different levels.

Semiotics helps you understand cultural meaning.

Accompanied immersions reveal behaviour in context.

Focus groups uncover drivers, barriers and language.

Omnibus surveys size the opportunity.

Targeted surveys prioritise needs, behaviours and motivations.

Longitudinal panels keep audience insight alive through development.

Not every project needs all six. But every project needs to know what kind of question it is asking.

If you are defining a proposition, you may need cultural mapping and audience drivers.

If you are improving the visitor journey, you may need accompanied immersions.

If you are trying to understand whether an audience exists, you may need market sizing.

If you are choosing between messages, benefits or show elements, you may need quant.

If you are developing a new experience over time, you may need a longitudinal panel.

The practical warning is to avoid using the wrong method at the wrong moment. Do not ask a survey to explain something you do not yet understand. Do not ask focus groups to size a market. Do not ask semiotics to tell you exactly how many people will buy a ticket. Do not ask post-launch evaluation to fix a pre-launch problem.

The deck ended with an invaluable checklist that should be printed out by anyone commissioning research:

• Are you asking the right questions at the right time?
• Is your team on board with, and engaged with, audience insight from the start?
• Are you combining methods to get a holistic picture?
• Are you bringing in perspectives from outside your sector?
• Does your research have a clear role at every stage, from concept through to launch?
• How can the insight live on inside your organisation, and not just sit in a report?

That final question is the one that matters most.

Audience research fails when it becomes a document. It works when it becomes a living tool: a language, a set of principles, a decision aid, a shared reference point and a way of keeping the audience present when the audience is not in the room.

The bigger point: audience-centred does not mean audience-led

The most useful idea in this Campfire was the distinction between audience-centred and audience-led.

Audience-centred work understands the audience. It listens carefully. It observes behaviour. It checks assumptions. It uses evidence to shape proposition, journey, messaging and development.

Audience-led work, at its worst, can become timid. It can chase stated preferences, flatten creative risk or mistake immediate reactions for long-term value.

Joao and Rogers were not arguing for the second version. They were arguing for the first.

This is important in immersive and experiential work because the audience may not know what it wants before it experiences it. People can struggle to describe new formats, unfamiliar interaction models or emergent forms of art and entertainment. But they can reveal needs, anxieties, motivations, behaviours and expectations. Good research does not ask the audience to become the creative director. It helps the creative director understand the conditions under which the work can land.

That is why Oxy Insight’s Barbican Immersive case study matters. It shows insight being used not to reduce ambition, but to support it. The goal was not to make safer work. It was to build a clearer proposition, understand the target audience, improve the journey and give the team better tools for decision-making.

In an Experience Economy that increasingly depends on mixed formats, younger audiences, new cultural codes and more complex visitor journeys, this kind of discipline is not a luxury. It is part of making ambitious work legible, accessible and commercially viable.

The best experiences are not designed by committee. But they are rarely designed well in a vacuum.

For more on this topic

WXO Campfires are member-only sessions, with members also unlocking access to the organisation’s full on-demand library of Campfire recordings. 

The recording, transcript and presentation from this Campfire are available to WXO members inside the membership hub.

You can check out listings for forthcoming Campfires here.

For more information on how to join the WXO, head here.