Most of the family offices and corporate treasuries we talk to in the UAE don’t open with questions about strategy. They open with questions about control. That tells you something about who actually adopts algorithmic execution at the treasury level, and it’s not who the retail trading world assumes.
A treasury desk is not chasing returns the way a prop trader does. Its job is to manage cash, hedge currency exposure, and keep capital working without taking on risk that a board hasn’t signed off on. So when a CFO or a family office principal asks us about automated trading, the conversation looks nothing like a sales pitch. It looks like due diligence. Here are the questions that come up first, in roughly the order they tend to arrive.
“Where does the money actually sit?”
This is almost always the opening question, and it’s the right one. Treasury teams want to know that the capital stays in an account they own and control, at a broker or venue they’ve vetted, under their own name. An Expert Advisor running on MetaTrader 5 executes inside that account. It places orders. It does not hold, move, or have withdrawal access to funds.
Spelling this out early matters because the word “algorithm” makes some people picture handing money to a black box. The reality is closer to a rules engine that can only do one thing: send trade instructions to a brokerage account the client already controls. Once that distinction lands, the rest of the conversation gets easier.
“What happens on a bad day?”
No one signs off on a system because of its best month. They sign off because they understand its worst one. So the second question is usually about drawdown, and specifically about hard limits.
This is where we walk through the risk engine rather than the entry logic. Treasury clients care far more about the floor than the ceiling. They want to see a daily loss limit that stops trading when breached, a maximum position size tied to account equity rather than a fixed lot, and behaviour during a gap or a fast market. A system that can explain exactly when it stops is more reassuring to a treasury desk than one that promises a high win rate.
We tend to frame this in language treasury people already use: this is a risk mandate expressed in code. The parameters are the policy. If the policy says no more than a set percentage of equity at risk per day, the EA enforces it without needing someone to watch the screen.
“How do we know it isn’t just curve-fit?”
Sophisticated clients know that any strategy can be tuned to look perfect on past data. The ones who’ve been burned before ask this question directly, and they’re testing whether you’ll be honest about it.
Our answer is the validation process, not a backtest screenshot. We optimise parameters on one slice of history, then test the chosen settings on a later slice the optimisation never saw. If the edge survives that out-of-sample period, it’s more likely to be real. If it collapses, we say so and move on. We’ve rejected entire instruments for specific systems when no honest parameter region existed. Being willing to say “this doesn’t work here” is usually what earns the next meeting.
We also avoid picking the single best run from an optimisation. Instead we cluster the strong results and extract settings from the centre of a robust group, on the logic that a parameter set surrounded by other good results is safer than a lone peak that may be a statistical fluke.
“Who is accountable, and what reporting do we get?”
A treasury function answers to a board, an auditor, or a principal. So the question of reporting is not an afterthought. They want a record they can show someone else.
This means clear trade logs, periodic performance reporting in a format their finance team can read, and a defined point of contact when something needs explaining. For a family office, “explainable” often outranks “optimal.” A modest, well-documented system they can describe in a board meeting beats an aggressive one they can’t.
“Can this fit our existing mandate instead of replacing it?”
The most realistic adopters don’t want to hand their treasury operation to an algorithm. They want a defined, ring-fenced allocation that runs under rules they approve, alongside everything else they already do. That’s usually how it starts: a controlled portion of capital, a strict risk limit, transparent reporting, and a review period before anyone discusses scaling.
This framing matters because it sets honest expectations. Algorithmic execution at the treasury level is a tool inside a mandate, not a replacement for judgement. The desks that get value from it are the ones that treat it that way.
The pattern underneath all of it
If there’s a thread running through these questions, it’s that treasury and family office clients are buying control, transparency, and accountability before they’re buying performance. That’s a very different buyer from the retail trader comparing win rates, and it’s worth understanding before you walk into the room.
For us, it’s also a useful filter. The same things a serious treasury client asks for, defined risk limits, honest validation, and reporting they can stand behind, are the things any trading system should be built around anyway. The questions just make it explicit.
QRC Enterprise Systems develops white-label algorithmic trading systems for UAE corporates and family offices, with risk parameters and reporting built to fit an existing treasury mandate. If your finance team is exploring this, get in touch to arrange a briefing.
