Owning the Asset Class of One.
Why gaming assets underperform inside diversified portfolios — and the five levers owners control.
Somewhere in your portfolio there is a casino. It was acquired and is governed as hospitality yield. What it has quietly become is a regulated data-and-liquidity business with a building attached.
Somewhere in your portfolio there is a casino.
It is almost certainly grouped with something else — real assets, hospitality and leisure, alternatives, special situations. That category is where the asset inherits its reporting, its benchmarks and its governance rhythm, and for most holdings the arrangement works perfectly well. Gaming is the exception. It behaves differently enough from everything around it that it needs to be managed on its own terms rather than by the conventions of the bucket it happens to sit in.
It probably arrived through a real estate door, or a hospitality allocation, or as part of a larger transaction where it wasn’t the headline. It was diligenced by people who price buildings, covenants and cash flows. It came with an operator — selected for hospitality competence, inherited with the asset, or valued primarily for licence access and local relationships.
Since then it has behaved acceptably. Distributions have arrived. The operator reports gross gaming revenue, hold percentage, occupancy, covers, and the numbers look broadly fine. What none of those measures reveal is whether the thing that actually generates the value is compounding or eroding.
This is not a failure of oversight. It is a question of how the asset was categorised. These assets are almost always acquired and governed as hospitality yield. What many of them have quietly become is regulated data-and-liquidity businesses with a building attached. The operating model, the board pack, the metrics and the operator’s mandate tend to follow the original category rather than the current one — and all of it is fixable, from the owner’s side, without touching the property.
There is a structural reason this persists. In most portfolios the casino is the only one of its kind — no peer set to benchmark it against, no internal function built for it, no colleague down the hall who has run one before. Nobody builds specialist capability for a sample size of one.
Inherited from property, not from gaming.
The reporting conventions in this sector were inherited from property and hospitality, and most portfolio-held gaming assets still run on them. Footfall. Spend. Occupancy. Margin. Capex against plan.
What is almost never there: who is coming, whether they are worth more or less than they were a year ago, how many of them the business could contact tomorrow if it needed to, and what proportion of revenue would walk out the door if a single relationship manager resigned.
In a mature gaming asset, the largest determinant of enterprise value is not the property, the licence, or the floor layout. It is the identified, addressable, retained customer base. It is also the number least likely to appear in a board pack — not because owners don’t want it, but because the reporting was never designed to produce it.
The asset’s most valuable component frequently sits on someone else’s system, in someone else’s format, under someone else’s contract.
They are responding rationally to what you pay them for.
They aren’t hiding it. Management agreements are almost universally structured on gross gaming revenue or EBITDA. Neither line moves when customer equity erodes — that takes eighteen to thirty-six months to surface, by which point the cause is unattributable and the operator may well have changed. There is no term in the agreement that rewards building a durable, portable, institutionally-owned customer relationship, so it doesn’t get built.
The same incentive shows up in reinvestment. Comps, promotions, points and complimentaries are typically the largest discretionary lever on the floor, and in most assets they are governed by tier entitlement tables rather than tested against incremental response. Entitlement-based reinvestment flatters near-term revenue and quietly transfers margin to players who would have visited anyway. It looks like generosity. It functions as leakage.
And the relationships themselves usually sit with individual hosts, not with the institution. That is an owner-level concentration risk that appears nowhere in the P&L until the host resigns and the book leaves with them.
The natural instinct at this point is to look harder at the operator. The more productive move is usually to look at the agreement. The issue is rarely competence — it is what the operator has been asked to optimise, and that is a document the owner controls.
Every jurisdiction is converging on the same three demands.
Identified play, auditable transaction records, and demonstrable player protection. The mechanisms differ. Australia is pursuing account-based play and cash limits. The Philippines operates under PAGCOR licensing alongside AMLC and central bank payment obligations. Macau has been pushed toward mass market and non-gaming revenue under explicit government direction. Singapore’s regime is mature and exacting. Across the region, FATF-aligned AML standards continue to tighten.
The direction is uniform. No regulator anywhere is moving the other way.
The consequence for an owner is structural, not administrative. Anonymous cash play is being converted from a business model into a liability. Every anonymous transaction is simultaneously an AML exposure and a permanent hole in the customer record — the same record that determines what the asset is worth.
Compliance delivered through headcount scales linearly with volume: grow the business, grow the cost. Compliance delivered through infrastructure does not. Over a hold period, that divergence is material — and it shows up in the multiple, not just the cost line.
None of these require replacing the operator.
All five are owner-side decisions.
- 01
Establish the customer record as an owned asset
The single sharpest question an owner can ask is this: if we changed operator tomorrow, could we take the guest data with us — complete, usable, in a form we could hand to a successor? If the answer is no, or nobody knows, the most valuable component of the asset is not under your control. This reframes data from an IT topic into an ownership topic, which is the only frame that gets acted on.
- 02
Move reinvestment from entitlement to evidence
Reinvestment is capital allocation and should be governed like it. Test-and-control discipline on offers. Measurement of incremental response rather than redemption rate. Points and comp liability carried on the balance sheet with a stated basis, not buried in a footnote. This is typically the fastest margin recovery available in a gaming asset and it requires no capital.
- 03
Convert cash flows into account-based flows
Frame this as liquidity and wallet share, not as compliance. Account-based play produces three things at once: an auditable record, a longer and deeper spending relationship, and a customer identity that persists across gaming, food and beverage, retail and rooms. That last point matters most for diversification — non-gaming attach rate is impossible to grow without identity continuity across the property.
- 04
Shift compliance from headcount to architecture
Beyond the cost curve, there is a benefit rarely priced: an operator that can produce a complete transaction and player-protection record on demand negotiates with its regulator from a position of strength rather than exposure. In a licensed business, that standing is an asset.
- 05
Replace historic reporting with forward-looking player intelligence
Gross gaming revenue is a lagging indicator contaminated by luck. Hold variance can comfortably mask two consecutive quarters of customer erosion. The board pack should carry leading indicators instead: active member rate, visit frequency by cohort, populations at risk of churn, tier migration, and revenue concentration. Concentration is the one to start with. In most gaming assets a small number of players drive a disproportionate share of revenue, and where that concentration isn’t quantified it tends to surface for the first time in a buyer’s data room. Far better to hold the number first.
Owner-side operating capability.
Every lever above assumes someone at the asset can execute it. That assumption is usually where this stalls.
Owners tend to see two options. Push the operator harder — which runs into the incentive problem, and into the reality that the operator may not have this capability either. Or build gaming expertise in-house — expensive, slow, and hard to justify for a single asset in a diversified portfolio.
There is a third, and it is under-used in this sector because it has been normal in others for decades: owner-side operating capability. Sector expertise that sits alongside the operator but is accountable to the owner. It sets the metrics, interrogates the reporting, governs the reinvestment envelope, and holds the technology and data agenda on the owner’s behalf.
In most gaming assets, the interpretation of performance comes from the same party whose performance is being interpreted. The sector has long treated that as normal. In almost any other asset class in the portfolio, it would attract more scrutiny.
What will an acquirer discount you for?
A buyer diligencing a gaming asset in the late 2020s will ask for the customer record, the compliance audit trail, the reinvestment efficiency data, and the player-protection evidence. Where those can’t be produced cleanly, the response is a discount, an escrow, or a conditional close.
The reverse is less obvious and more valuable: auditable, portable customer data is itself a multiple argument. It converts trust the operator’s numbers into here is the evidence.
That is the floor of the argument. The ceiling is considerably higher, and it is the part almost nobody puts in front of this audience.
The upside case, rarely made.
Do the work described above and you don’t just end up with a better-run casino. You end up holding three things that are expensive to build, durable once built, and — this is the part that gets missed — reusable outside the four walls of the property: a verified customer identity record, a payments and liquidity rail, and a compliance and audit apparatus that satisfies a regulator.
Every operator in the sector needs all three. Very few have built them well, and fewer still have built them once, properly, in a form that can be reused. Once you have, three things open up that were previously unavailable.
The acquisition calculus changes.
A single-asset owner buying a second property is buying a second set of problems. An owner with working infrastructure buying a second property is buying distribution for a platform they already own — and the second integration costs a fraction of the first. This is the mechanism by which a collection of assets becomes a group with genuine synergy, rather than a group in name with shared reporting. It is also the reason the roll-up logic that works in dental practices, veterinary clinics and childcare has largely failed to translate into gaming: nobody built the shared layer first.
The infrastructure becomes a separable asset.
The capability you build for your own property can be licensed to operators who will never build it themselves — mid-market casinos, clubs, regional venues, adjacent hospitality. At which point you hold two assets with different characteristics: an operating gaming business, and a technology business serving the sector. Those are not valued the same way. Operating gaming assets are typically valued on EBITDA multiples reflecting regulatory risk, capital intensity and cyclicality. Recurring-revenue infrastructure businesses are valued on an entirely different basis. The specific spread is worth testing with your own advisers rather than taken from an article — but the directional gap is not controversial, and it is wide.
Adjacent revenue becomes addressable.
Once identity and payment ride on the same rail, the property stops being the boundary of the business. A partner and merchant ecosystem around the venue. Digital engagement that persists between visits rather than resetting at the door. Licensed online adjacency in jurisdictions that permit it. Digital assets and tokenised loyalty where the demographics support it. None of these are available to a business that only knows its customers while they are standing in the building.
The common thread: each of these is an owner-level opportunity, not an operator-level one. They require capital allocation decisions, group-level strategy and a multi-asset view. An operator running a single property under a GGR-linked management agreement has neither the mandate nor the incentive to pursue any of them.
A structural advantage, largely unclaimed.
There is a prevailing view — and we think it is wrong — that a family office or diversified fund holding a casino is a structural mismatch: under-resourced on expertise, over-exposed to a regulated asset. The opposite case is stronger, and almost nobody is making it to this audience.
The transition to account-based, data-native gaming operations takes longer than a typical private equity hold period. A fund has to demonstrate the uplift before it exits, which pushes it toward levers that show up fast and reporting that flatters the near term. The multiple arbitrage described above — build the infrastructure, prove it in your own asset, then license it to the sector — is a five-to-ten year proposition. It is structurally unavailable to a holder with a clock.
Patient capital has no such constraint. It can build the customer equity, hold the compounding, and capture the platform value on the other side. That is a genuine and rare structural advantage. It is also, at present, largely unclaimed.
For your next board meeting.
- 01
If we changed operator tomorrow, could we take the guest data — complete, usable, ours?
- 02
What percentage of our gaming revenue comes from players we can identify by name?
- 03
What is our reinvestment as a share of gaming revenue, and what is the measured incremental return on it?
- 04
What proportion of revenue sits with our top players, and what is the projected twelve-month attrition of that cohort?
- 05
If our regulator requested a complete transaction and player-protection record for a named customer, how long would it take to produce?
- 06
What has to be true about our data before any intelligence layer produces answers we can act on?
That is worth knowing in itself. Once you can answer them, the more interesting question is what you would build with the answers.
Disclaimer. The views and opinions expressed in this article are those of the author and do not reflect the official policy or position of any institution or organisation. Nothing here constitutes investment, legal or tax advice. While every effort has been made to ensure the accuracy of the information, the author, its affiliates and companies inferred in this document assume no responsibility for any errors or omissions.