The Experiences Industry Is Hitting Its Inflection Point

Industries do not turn on a single day. What looks in retrospect like a moment of change is almost always, in the evidence, a convergence. Several structural conditions that had been stable for years weaken at the same time. Several others, previously absent, begin to accumulate. Somewhere in that overlap the industry crosses a threshold, after which the prior shape cannot be held in place and a new shape becomes difficult to avoid. That threshold is what an inflection point actually is.

The argument of this essay is that the experiences industry is inside one of those crossings now. The argument is not that something dramatic happens in a particular quarter, nor that any specific event is decisive. The argument is that five structural conditions that held the industry’s shape in place throughout the digital distribution era are weakening simultaneously, that the market signals visible in public-facing sources confirm the transition is underway, and that the choices made at the supply-surface layer during this window have outsized and durable consequences, for the reasons Essay 9 described.

The conditions that held the prior shape

The experiences industry has had a remarkably stable shape since online distribution scaled. It is worth naming the structural conditions that held that shape, because each of them is now under pressure, and the simultaneity is the point.

Human-tolerant interfaces. The entire distribution graph assumed a human at the edge who would absorb incoherence (Essay 7). A misprinted pickup location, a stale availability cache, an ambiguous voucher number. All of these were quietly reconciled by the traveller standing at the desk. The interface did not need to be complete or deterministic because the human made it complete on arrival.

Fragmented demand matched to fragmented supply. The industry’s fragmentation has not felt dysfunctional because both sides of it are fragmented. No single distributor held enough demand to force structural coherence on supply; no single operator category held enough supply to force structural coherence on demand. Fragmentation was stable because the asymmetry between the two sides was small.

Interface as the primary competitive surface. Distributors and suppliers alike have, in the modern history of the industry, treated the interface as where competition happens. Booking flows, conversion, brand. This was rational as long as humans were the counterparty and the interface was the surface they interacted with.

Channel managers as a good-enough routing layer. The category that emerged as a distinct piece of infrastructure solved for shared integration cost and consolidated reseller access (Essay 6). It did not solve for authoritative real-time availability, end-to-end data integrity, or neutral settlement, but the margin of error those gaps produced was absorbed by the human-tolerant interface upstream. The routing layer was sufficient because the failures downstream were human-resolvable.

Connectivity treated as a cost centre. Because none of the above conditions pressured the connectivity layer, participants at all levels of the industry treated it as overhead. Something to minimise. Something to outsource. Something that did not repay strategic investment.

Five conditions. One equilibrium.

Why each condition is weakening now

The conditions are weakening not in sequence, but together.

Human-tolerance is being withdrawn. The subsidy the industry has run on is priced at zero in the agent era. An agent does not reconcile. It queries, receives, and either acts or fails.^¹

The buyer side is consolidating faster than the supplier side. The demand-side concentration visible in the major online travel aggregators has moved the industry away from the symmetric fragmentation that made the prior equilibrium stable.^² A smaller number of buyers holds a larger share of the flow. That asymmetry alone, absent everything else, would be enough to pressure the distribution graph.

The interface is commoditising under agent pressure. For the reasons Essay 9 set out, accumulated investment in conversion design and brand does not transfer to a counterparty that does not convert and does not respond to brand.

The channel-manager ceiling is becoming material. Channel managers were sufficient when routing could be good-enough and failures could be resolved downstream. With agents at the other end, routing has to be authoritative and failures cannot be resolved downstream. The routing layer is now being asked to do something it was not designed to do.^³

Connectivity is no longer a cost centre. If Essay 9 is right, it is the layer value migrates into when the interface commoditises.^⁴ The strategic cost of treating it as overhead compounds in every quarter the migration continues.

Five conditions. Five simultaneous pressures. Not from any single direction.

What the signals confirm

A structural argument about weakening conditions is worth only as much as the signals in the market confirm it. Five are visible in public-facing sources.

Standardisation activity at the connectivity layer. Arival’s Technology and Connectivity 2023 report documents the emergence of connectivity standardisation efforts within tours and activities, a category that had no equivalent work earlier in the digital era. The presence of an industry-level standards conversation is, in itself, a signal that the connectivity layer has become strategic.^⁵

Shift in how operators describe their technology priorities. The same Arival research tracks a notable move of connectivity and data quality up the operator priority list. Technology discussion among operators has shifted from point-of-sale and website, which dominated prior-cycle surveys, toward connectivity and supply-feed quality. This is the operator-side expression of the pressure described above.

A dedicated industry literature on agent-mediated demand. Arival’s Guide to AI in Experiences 2024 is the first industry-wide synthesis of how agent-mediated interaction changes what the supply side is required to present.^⁶ Earlier no such literature existed in the category. The fact that the industry’s primary research body has produced a guide is itself the signal; it is not a commentary on any single recommendation within it.

Observable moves at the buyer side toward structured supply. The largest distributors in the category have publicly communicated, over recent quarters, an investment focus on structured supply feeds, machine-readable catalogues, and direct real-time connectivity with operators, rather than on front-end optimisation alone. No specific company is named here; the observation is about a pattern visible across the set of major distributors, which is what makes it a signal about the industry rather than a claim about any one participant.^⁷

Capital flows into infrastructure rather than interface. The direction of investment in the sector has visibly shifted from interface-layer businesses (new booking apps, new marketplace front-ends) toward infrastructure-layer businesses (connectivity, data, settlement). This observation is qualitative, but it is consistent with what the prior essays predict: value migrates to the layer underneath when the interface commoditises, and capital generally follows the migration.^⁸

None of the five signals is decisive on its own. Together, they triangulate the same structural moment.

Why this is not the mobile cycle or the consolidation cycle

The experiences industry has had prior moments that felt, at the time, like inflections. An earlier mobile cycle, during which the interface shifted from desktop to smartphone. A subsequent OTA consolidation cycle, during which a small number of distributors acquired or squeezed out smaller ones.^⁹ In both, industry commentary described a transformation. In both, the underlying shape of the industry survived more or less intact.

The reason is structural. Both prior cycles operated at the interface layer. The mobile cycle changed the device through which the interface was consumed; it did not change what the interface had to do. The consolidation cycle changed the ownership structure of distributors; it did not change the underlying distribution graph. The five prior-equilibrium conditions described above held through both.

The current moment is categorically different because it operates at the supply-surface layer. The pressure is not on how the interface is shaped, or who owns the interface, but on whether the interface can continue to exist as the primary competitive surface at all. The five prior-equilibrium conditions are under pressure simultaneously, and four of them sit underneath the interface, at the layer the prior cycles did not disturb. A reader who assumes this cycle resembles the prior ones is making a category error, not a disagreement of degree.

The window, and why it is not symmetric

The argument of Essay 9 was that infrastructure positions, once held, are durable.^¹⁰ The value that migrated to the dominant search index, to the dominant fulfilment layer, to the leading hyperscalers, did not redistribute later because the layer underneath had been built at a scale a new entrant could not assemble. The same durability applies to whatever supply-surface layer ends up holding the coherence function in experiences. The choices made in the next several years about how supply is presented to agent-mediated demand are choices that compound. Early decisions about architecture, commercial terms, and participation requirements establish the shape, and the shape persists.

This is a structural observation about how infrastructure layers form, not a tactical prescription. The asymmetry of the window is what matters: a participant arriving late to the layer does not arrive at the same layer. The coherence function is either a neutral shared resource, a set of vertically integrated private resources, or some mixture of the two, and which of those shapes solidifies is being settled by the moves visible in the market now.

What this means for different participants

Each class of participant is facing a different version of the same structural question.

Large distributors. Already building or buying supply-surface coherence inside their own commercial perimeter. Their choice, expressed through public roadmaps and investment patterns, is consistent with the vertically integrated path from Essay 9. Whether they also participate in a neutral layer is an open question.

Smaller distributors. Facing a different choice. They cannot replicate the integrated supply surface at scale, and they may find the agent era rewards either specialisation (becoming a curation and trust layer on top of someone else’s supply surface) or participation in a neutral infrastructure layer that allows them to remain competitively relevant without owning the supply coherence themselves.

Suppliers at the top of the supply curve. Already have connectivity in a shape that is agent-ready or close to it. Their question is on what commercial terms their supply surface is visible to which agents.

Suppliers at the long tail. Facing the sharpest structural question, for the reasons Essay 8 described. Reachability by agents is not a given, and the route to reachability runs either through an intermediary layer or through a vertically integrated platform whose commercial terms are not neutral.

Infrastructure participants. Operating inside the window during which the infrastructure position in this category is, for the first time, forming at industry-wide scale. The opportunity is large. The window is not indefinite, for the durability reasons already described.

Close

What changes during an inflection point is not the industry’s visible surface. The booking flows remain booking flows. The marketing campaigns still run. The interface-layer conversation continues as though it were the central conversation. What changes is underneath. The five conditions that held the prior shape give way at once. The connectivity layer moves from overhead to strategic. The supply surface becomes the site of competition. The settlement is not yet decided; it is being shaped, distributedly, by the technical and commercial choices participants are making now. And the choices made at that layer solidify into an infrastructure position whose durability means the industry’s shape for the decade after the window is, in large part, set inside the window itself.

The next essay attempts the straight-faced exercise of describing what the experiences industry looks like in 2030, on each of the two paths that the inflection could resolve into. The exercise is useful not because the future is knowable but because the structural properties of each outcome are knowable, and reading them side by side makes the stakes of the current window visible in a way that a more cautious essay could not.