So if this is your year – and for many in the visitor attractions world, it needs to be – then that day might as well be today. Because by the time high season hits, around say April or May for most attractions in the northern hemisphere, your strategy should already be in play.

The reality is, the industry’s moment for hesitation has passed. If you’re still waiting on that source system upgrade first or worse, debating whether to bring AI – or even data – into your organisation, you’re not just behind the curve, you’re standing still while the world marches on. 

This isn’t about chasing shiny tech. It’s about responding to a sharper new status quo. There was a moment, particularly in 2021 and 2022, when budgets loosened, government grants flowed more freely and innovation funding found oxygen. By the end of 2023, that window had largely closed. Two years later, economic uncertainty, rising costs and tightened public and private funding have ushered in a new era, one where performance, not just experimentation, drives priority. What we’re living in now is not a temporary slump, but a lasting shift: a world of tighter margins, cautious capital and relentless pressure to deliver more with less. That means commercial performance, (or at least for cultural institutions, sustainability), has to come first. Every investment needs to earn its keep. Every initiative must tie to outcomes. And AI, done right, delivers exactly that.

By the end of 2025, adoption of AI tools had already become commonplace in enterprise contexts: roughly three quarters of organisations globally reported using or testing AI in at least one business function, nearly nine in ten saying AI was a regular part of how they work. Yet most have yet to embed AI deeply enough to see meaningful, organisation wide impact (McKinsey).

What this suggests is that we’re not even close to AI’s growth curve – we’re at the beginning of a decade-long transition in how organisations operate. Over the next two to three years, we’ll see significant ground gained as successful adopters move from experimentation to scaled deployment, automated decision making and strategic integration. But even as these foundations are laid, the broader enterprise AI era, where data and AI reshape revenue, operations and competitiveness, is just getting started.

Execution speed will look different everywhere. Despite the need to manoeuvre around regulatory constraints, union agreements, aging ride control systems and public accountability, the most important thing is to simply begin. 

Intelligence that moves the needle

AI has become a headline grabber across industries. In attractions, the buzz is real: global AI spending in wider tourism and hospitality is on the rise as operators explore everything from turning a painting into a poem, powering a chatbot, even robotics. All cool. But most are what we’d call fun but fringe. Interesting experiments. Great for social media clout. None will transform the balance sheet.

Here’s the rub: adoption isn’t the same as impact. In other sectors, research shows a widening chasm between organizations that talk about AI and those that actually get value from it. According to a global study, many enterprises struggle to scale AI beyond pilots because the business case is unclear, integration is poor, or their data isn’t ready. Only about 5 % of companies globally are truly deriving value from AI at scale (meaning measurable returns such as revenue growth and cost reduction), while the majority struggle despite investment (BCG).

This is the hard reality of modern digital transformation. Across sectors, organisations with solid data practices and strategic clarity are overwhelmingly more likely to see AI deliver revenue growth, operational efficiency and competitive advantage. Others? A pile of toys that gather dust and headlines that outstrip results. Or worse, a sprawling multi year, multi million dollar investment with no payoff, the result of a team handed too much runway to play with.

That’s where the distinction becomes critical for attractions. It’s not enough to experiment with AI. To capture value, AI must be anchored in practical applications that generate clear value and return, with data that’s fit for purpose – integrated, governed, contextualised to your unique performance signals.

When it creates real outcomes, it’s because it’s tightly coupled with core decision loops:

• Forecasting visitation and spend to inform budgeting, pricing, staffing, inventory and promotional decisions weeks ahead, including simulation and scenario planning
• Conversational insight so frontline and strategic teams alike can ask real questions and get actionable answers, immediately
• Guest voice synthesis at scale to turn thousands of reviews, surveys and mentions into themes and trends that executives can act on before a drop in satisfaction becomes a drop in attendance
• Demand modeling to understanding where demand is coming from, what drives conversion and how different audience segments behave, to inform smarter growth strategies, price sensitivity testing, targeted acquisition, campaign optimisation and long range planning
• Operational optimization from maintenance planning to queue management, as a productivity lever that tangibly increases throughput, improves cost management and delivers on guest satisfaction

Starting over strategy also means risk and governance can be informed through experience, rather than theoretical and prone holes in practice. Where intelligence intersects decisions relating to topics like staffing, maintenance or guest flow – leading practice should feel out guardrails and human in the loop controls, with an agile approach that lends itself to staged deployment.

Data first, always.

AI without the right data foundation is just a fancy feature, which might look impressive in a demo, but it won’t stand up to the rigour of real world complexity. And yet, it’s still far too common to see organizations start with the flash, or the easy: a glossy consultant deck, or a generic chatbot. A project with no data under the hood might feel like momentum, but it’s movement without direction.

On the flipside, if your board’s first move is to ask for a custom built data warehouse – stop. If your instinct is to call a Big Four consultancy for a 10,000 word data audit with the TL;DR headline of ‘do data & AI’ – what you already know, pause. AI strategy should be steeped in practical experience. Anything else is theatre. AI, when done right, should be hyper efficient, agile in the truest sense and ruthlessly focused on outcomes. Strategy will emerge from experience and results. Equally, getting to great data shouldn’t take two years and a million dollars (which will only make you late to the AI party) – rather weeks to activate and days to show initial value. It’s not about building from scratch, it’s about connecting the dots that are already in your ecosystem. The former is fun at first, but an easy way to burn political capital fast and the spearhead  rarely survives leadership changes. 

Industry analysts project that a majority of AI initiatives will be abandoned before they ever deliver meaningful outcomes, unless they’re built on a solid data foundation and achieve strategic yet specific alignment from the start. Up to 60% of AI projects will likely be shelved this year simply because they lack the right data foundation (Gartner).

This is where the deeper reality of AI comes into focus: the success of any model, generative or predictive, is inseparable from the domain knowledge and data context it’s built on. Machine learning doesn’t run in isolation, it runs on context, unified data and industry frameworks tuned to the nuances of the field. In visitor attractions, that means data shaped by attendance cycles, ticketing models, guest behaviour patterns, event calendars, operational constraints and experience design. Leading theory and emerging best practices from AI research (including Google’s data centric AI and the MLOps movement) reinforce this: AI systems are only as good as the contextual accuracy and operational proximity of the data they’re fed. In other words, the more specific, the more successful.

At Dexibit, we’ve built AI for visitor attractions from the data up. That means connecting the systems attractions already use and shaping that data around the realities of attendance cycles, seasonality, pricing and guest behaviour. On top of that foundation, we apply domain specific models for forecasting, demand, operations and guest feedback, enriched with contextual signals like events, calendars and industry benchmarks. The result is intelligence that’s close to real decisions, not abstract dashboards or novelty tools.

Most attractions don’t need an internal AI lab. They need a faster, cost effective and de-risked path from data to outcomes, with a partner who understands both the technology together with the operational applications and constraints of the sector. That’s the gap we’re built to fill.

Culture is the multiplier

Technology is only half the equation. Culture is the part that determines whether AI actually changes how an organisation works, the part that only develops with the keys in their hands. It can sometimes be slow, perhaps politically charged – but essential nonetheless.

By the end of 2025, while most enterprises had already introduced AI in some form, research shows only a minority are seeing enterprise level impact as a whole. One consistent finding across studies is that success correlates less with the sophistication of the technology and more with how widely it is used, trusted and acted on. In other words, AI delivers value when it shows up in day to day decisions by teams, not when it sits on a roadmap or in the hands of a favored few.

The teams pulling ahead are not the ones with the biggest AI budgets. They’re the ones putting AI into the hands of their people – at the front line, in operations, in marketing, in planning – and using it to answer real questions in real time. These teams are crying out for this enablement. Too often, they’re held back by their organization’s speed of execution to deliver. Surveys show 60 % of employees are using unapproved AI tools at work (so called ‘shadow AI’), and up to 93 % admit to using them without formal governance. 

Successful teams, immersed in data and AI, may just be getting started – but they’re building habits of curiosity. They’re lowering the friction between question and insight. They’re making it normal to ask why something happened, what might happen next, and what to do about it. This is where practical experience with AI matters: it’s a finance team who begin using machine learning and AI simulation as part of their planning process. A visitor services manager who reviews recommended replies to complaints rather than writing from scratch. A retail store manager who finds slowing SKUs to reposition. A marketing manager who has AI analyze their exhibition retrospective.

AI maturity doesn’t come from strategy documents or steering committees or even deep technical work. It comes from repeated use, from learning what works and what doesn’t, from sharing knowledge and embedding insight into everyday workflows. Over time, that creates a shift: data stops being something you report on after the fact and becomes something you collaborate and work with continuously.

In a time where margins are tight and expectations high, that cultural shift is no longer optional. It’s how organisations move from experimentation to impact and why the work of building AI capability needs to start sooner than later.

Strategy season has started. If AI and data aren’t in your strategic plan already, it’s time to ask: what future are we planning for? By the time July rolls around, it’ll be clear who heard the starting whistle and who didn’t – the teams that set themselves up for the season versus those that just talked about it. The difference will show up in the briefings, the guest experience, the bottom line. So let’s not just say ‘one day’. Let’s call today ‘Day One’.