Alex ZonneveldHow much risk can you take, and when might you need your capital back? Those two questions can help shape how you put it to work.
Risk and liquidity are two important decisions in managing capital. Risk includes both how a strategy can perform and the structure around it: who holds the assets, who can trade them and who can move them. Liquidity is about matching the terms of an investment to when you may need your capital.
Access is the foundation that connects those choices to strategies. It includes how you participate, how you see what is happening and what authority you retain. The strategies you choose, their performance, fees and market conditions all influence the return you ultimately experience; it is never guaranteed. Your objectives, time horizon, costs and tax circumstances matter too.
The industry is optimising the wrong thing
Most of the wealth industry’s AI story this year has been about planning. Altruist’s AI tax agent knocked billions off listed wealth managers in February, Anthropic launched Claude for Financial Advisors in September, and Hazel added a full planning agent in October. All of this makes advisors faster at producing plans and reports, but none of it tells you whether the capital itself is in good shape.
If you look at what actually hurt wealth clients in 2026, planning had very little to do with it. Private credit funds that had been sold with quarterly liquidity capped redemptions at 5% while holders were asking for 10%, 15% and in one case around 40%. More than $500 billion of software loans had built up inside those funds without most holders realising it was there. Meanwhile an experimental AI agent with its own crypto wallet sent away hundreds of thousands of dollars because of a simple mistake. Each of these was a failure of risk, liquidity or access, and a better financial plan wouldn’t have prevented any of them.
Risk has two layers, and most people only look at one
When most people think about risk, they’re thinking about trading risk, which is how much a strategy could lose through drawdown, leverage, volatility or concentration. That matters a great deal, not least because losses compound against you: a 50% fall needs a 100% gain just to get back to where you started, which is why the downside of a strategy tells you more than its best year.
Underneath that sits structural risk, which most people never measure at all. It’s the question of who holds the assets, who can trade them, who can move them out, what happens if the manager or the platform fails, and whether the number you’re looking at was produced by someone independent. Structural risk is different in kind from trading risk, because a poor strategy tends to lose money gradually while a poor structure can lose all of it in a day.
That’s why family offices check the structure before they even look at a strategy. Before capital goes to an outside manager, they make sure the assets will sit with a custodian, that the manager can trade them but never move them out, and that the valuation comes from an independent administrator, and only once that’s settled do they start assessing how the strategy behaves. We go into this in more detail in Two kinds of risk, and most people only measure one.
Liquidity has to match your life, not just the fund’s terms
A fund’s liquidity terms describe what the fund is able to do, whereas your liquidity needs describe what your life is going to require, and managing liquidity is really about making sure those two things line up.
That sounds obvious, but in practice very few people ever write both sides down. Someone with a tax bill due in April and a property purchase planned for the summer might hold a semi-liquid credit fund that promises a quarterly exit, which works perfectly well in an ordinary quarter. In 2026, though, holders of Blue Owl’s technology lending fund asked to redeem around 40% of their shares, twice, and received 5% each time. The fund kept to its terms throughout. The mismatch was on the holders’ side, and nobody had ever shown it to them.
A family office avoids this by keeping a calendar of what the family will need and when, alongside a ladder showing how quickly each position could be turned into cash under its worst-case terms, and by never allowing the two to collide. There’s more on how that works in Liquidity is a match, not a feature.
Access means more than getting in
Access used to mean a single thing, which was whether you could get into a strategy at all. The best managers had high minimums, closed books and lengthy due diligence, which goes a long way to explaining why the average family office holds around 42% of its portfolio in alternatives, according to the UBS Global Family Office Report 2026, while most people hold almost none.
Today access has more to it than that. It matters who you arrive through, whether that’s directly, through an advisor or through someone you already trust. It matters what form you hold the capital in, whether a fund share, a vault you sign for or a token, and whether you see it in a quarterly PDF or live and traceable to the custodian. It matters who else can act on it, whether a co-signer, a family member or increasingly an AI agent, and what the whole arrangement costs. Good access lets you hold and act on capital the way you want without giving up control to get it, whereas poor access is the kind that’s easy to get into and hard to see inside. We set out the six parts of access in Access on your terms.
What AI changes about all three
None of these three factors is new. Family offices have always managed them, but they’ve done it with people: an operations lead checking custody, an analyst reading offering documents, an administrator reconciling valuations and a lawyer working through the side letters. That cost is the main reason this level of oversight has stayed with families who have nine-figure balance sheets.
What AI changes is the cost of that work rather than the nature of it. On risk, it can watch both structure and strategy continuously, checking that custody balances agree with what the manager reports, flagging when a drawdown approaches a limit you’ve set, and noticing when three different managers are effectively running the same trade. On liquidity, it can read every fund’s terms, place each position on a ladder and set that ladder against your own calendar, so you hear about a growing redemption queue before you’re part of it. On access, it changes the interface itself, because you can ask questions about your capital in plain language and, increasingly, let agents act within limits you’ve set, provided the structure underneath keeps the keys with you.
What AI shouldn’t do is make the decisions. Choosing managers, deciding how much risk to take and deciding when to redeem should stay with the person whose capital it is, because an oversight layer that starts making the decisions has stopped overseeing anything. In short, AI doesn’t change what good money management looks like, but it does change who can afford it.
How MakeBanc is built around the three
MakeBanc is capital orchestration infrastructure that connects investors with independent asset managers, and it’s built around risk, liquidity and access rather than around planning.
On risk, your assets sit with an institutional custodian in a vault you sign for, and the manager you choose can trade them but can never move them, while MakeBanc itself never holds the keys and you can check all of this on-chain for yourself. Every figure is labelled as self-reported, observed or verified, and the NAV is calculated by a third-party administrator. Strategies are classified by type, and historical drawdown is shown as reported by the manager, so the trading risk you take on is something you choose.
On liquidity, you can read a strategy’s terms in plain language before you allocate, and once you request a redemption you can see the date it will settle. On access, you arrive by invite from a manager or curator you already know, you can sign alone or with two or three named signers, you can see what the manager is able to see about you, and every cost is itemised before you commit.
MakeBanc doesn’t choose strategies for you, assess what suits you or allocate on your behalf, so the decisions remain yours throughout.
Frequently asked questions
What are the key success factors in managing money? There are three: risk, liquidity and access. Risk covers both structural risk, such as custody, counterparties and who can move assets, and trading risk, such as drawdown, leverage and concentration. Liquidity means matching how quickly each position can be redeemed to when you’ll actually need the money, and access means being able to hold, see and act on your capital in the way you choose.
What is structural risk in wealth management? Structural risk comes from the way assets are held rather than the way they’re traded. It includes custody, counterparty failure, who has permission to move assets and whether reported figures are independently verified, and because it can lead to a total loss, family offices check it before they look at any strategy.
How do family offices manage risk across external managers? They start with structure, making sure assets are held by a custodian, that managers can trade but not move assets out, and that valuations are independent. They then assess trading risk, including drawdown, leverage and overlap between managers, and keep monitoring both for as long as the capital is allocated.
How will AI change wealth management? AI is already making plans and reports cheap to produce, but its larger effect is likely to be making family-office-style oversight of risk, liquidity and access affordable for many more people, by automating document reading, reconciliation and monitoring while leaving the decisions with the owner.
Sources
- McKinsey & Company, The signal in the sell-off, April 2026
- Kitces, The Latest In Financial AdvisorTech, October 2026
- BIS Quarterly Review, Private credit’s software lending meets AI disruption, March 2026
- Alternative Credit Investor, Blue Owl private credit redemption requests fall again in Q3, 30 October 2026
- UBS, Global Family Office Report 2026; TechCrunch, Family offices are clamoring for AI investments
- The Crypto Times, AI agent accidentally sends $450K, February 2026
MakeBanc is a technology platform that connects investors with independent asset managers. MakeBanc does not provide investment advice, custody services or banking services. Past performance is not indicative of future results. Nothing here is an offer or invitation to allocate to any strategy.