City Shift Finance: As a business grows, which financial planning and reporting processes require the most attention, and what has your experience been in establishing them?
Zezag: When a company grows, the first thing that breaks is usually the definitions: what counts as revenue, whose cost something is, which company in the group should be paying for what.
When you have two entities, a couple of people can keep all of that in their heads. When you have more than ten, everyone starts to have their own version of the truth, and every conversation about numbers turns into a debate about whose spreadsheet is right.
So I tend to build things in a certain order. First, visibility over cash: where the money is, in which entity and in which bank. Then the monthly close: one chart of accounts and a fixed timetable. Then intercompany charges and cost allocation. Only once that base is stable do budgeting and forecasting become genuinely useful. Otherwise you are forecasting from numbers that different parts of the business interpret differently.
In one of the groups I work with today, we started with accounting and audit, then added cash-flow controls and budgeting. When one of the trading companies later went through the process of becoming a member of an international exchange, the quality of the financial reporting and the ability to stand up to external scrutiny became very practical.
Good reporting turned out to be useful well beyond management. Exchanges, banks and auditors all look at it. It is an asset that opens doors.
City Shift Finance: In a business where financial performance fluctuates significantly, how do you distinguish short-term volatility from changes that require revisions to financial forecasts?
Zezag: In trading, swings in performance are an integral part of the business. One bad month is not a reason to rewrite the forecast.
I separate two questions: did the result change, or did the mechanism that creates the result change?
If the market was temporarily quiet and volumes were low, but the economics per unit of volume were unchanged, I would usually treat that as volatility rather than a reason to change the model immediately.
But if the overall result still looks fine while exchange fees have gone up, a strategy is losing its edge, or new regulatory costs have appeared, that is a signal even if it has not fully shown up in the P&L yet.
Two things help in practice. First, when we approve a budget, we agree in advance what would actually trigger a revision. Not “revenue is down 10%,” but “this specific assumption has changed.” Second, I stress-test the fixed cost base against a weak year. People and infrastructure cannot be cut quickly, so those are the costs that need to survive a bad year.
When I worked in asset management, the logic was similar, except the volatility came through AUM. I only built the permanent cost base around management fees. Performance fees were upside. I never wanted to build a permanent headcount around them.
City Shift Finance: How have you approached allocating shared costs, particularly technology, infrastructure and personnel expenses, across different business activities?
Zezag: Shared-cost allocation usually has three different jobs, and the biggest mistake is trying to solve all three with one table.
The first is legal and tax: who charges whom, and whether that holds up from a transfer-pricing perspective. The second is management: what does this business line actually cost? The third is incentives: what should count when bonuses are calculated?
For management purposes, I allocate direct costs by actual use wherever that can be measured: servers, colocation, market data. For the full economics of a business line, I show both direct and shared costs. But when I am evaluating a team or calculating bonuses, I do not make people accountable for costs they cannot control. Otherwise they end up arguing about allocation keys instead of doing their jobs.
In software businesses, engineers are often the biggest shared cost. I prefer allocating that cost by time and roadmap rather than by product revenue. And I always show the full development cost before capitalization. Capitalization can improve accounting profit, while it does not create cash.
In one international fintech group where I worked, there were more than 35 entities, and I built the intercompany charging system. In my current role, I am working on a bonus-allocation model where the same issue becomes very sensitive. When an allocation affects people’s money, it quickly becomes either fair or a source of conflict.
City Shift Finance: What challenges have you encountered in developing and maintaining budgets across international operations?
Zezag: FX and tax can be managed. The harder part is that the legal structure of a group and the way the business actually operates are two different things.
Teams make the decisions, while the money moves through companies in different jurisdictions. A budget built only around legal entities does not reflect how management actually runs the business. But a budget built only around teams does not tell you where the cash needs to sit, or when it needs to move.
So I normally build the budget in the management view first, and then translate it into legal entities through an intercompany funding plan. That becomes a separate part of the budget: who pays whom, when, which bank account needs the money, and how long the transfer is likely to take.
Then there is the local calendar: audit deadlines, capital requirements, bank-account opening timelines. When I worked in asset management, capital requirements at the management-company level affected how much money could actually be moved. In fintech it was even more restrictive. In one international fintech group, each local licence came with its own minimum capital and restrictions around moving money out of the entity. That directly affected where liquidity had to sit and when individual companies needed funding.
That is why in fintech I start the budget with a map of licences and cash, and only then move on to expenses.
City Shift Finance: From your experience, what makes financial reporting useful for executive decisions?
Zezag: For every number in a report, I ask one question: if this number were different, would anyone make a different decision? If the answer is no, it probably belongs in the appendix.
The report should start with the conclusion: what happened and what I think we should do about it. The numbers come after that.
Consistency of format matters more than adding another layer of detail every month. If management sees the same structure repeatedly, after a few months they can spot the trend themselves.
A useful report also needs things that are not sitting neatly in the P&L: signed commitments, the current expense run-rate, how many months of free cash are left.
And I like every business to have one unit metric that management knows by heart. In fintech, it might be revenue and cost per transaction. In asset management, cost per dollar of AUM. In trading, return per unit of risk, deployable capital or volume. In software, cost to serve one customer.
Headline numbers grow as the business grows, so they can hide problems. A unit metric tells you whether the business is actually getting better, not just bigger.
City Shift Finance: How do you evaluate competing investment priorities when technology, staffing and operational infrastructure all require additional resources?
Zezag: I do not start by comparing categories. I start with the constraint each request is supposed to remove.
At any given point, a business usually has one or two constraints that matter much more than the others. Money spent there can change the result. Money spent somewhere else can simply create capacity nobody uses.
The second question is how reversible the decision is. Cloud capacity can be switched off next month. A colocation contract may lock you in for a year. Hiring is the hardest decision to reverse.
When revenue is volatile, every permanent hire raises the break-even point in a weak year, so I look at hiring more critically than I would look at buying equipment of the same cost.
Instead of funding the whole project in one go, I fund it up to the point where the result can be tested.
Where there is an internal development team, the question often becomes build or buy. Building in-house can look artificially cheap because the salaries are already in the budget. So I ask for a three-year cost of ownership. What it costs to build is only part of the number.
City Shift Finance: How has the relationship between FP&A and senior leadership changed throughout your career, particularly in financial planning, performance management and resource allocation?
Zezag: When I started my career, finance was mostly asked to explain what had already happened: actual versus budget, the variance, the reason, done.
Now I am usually brought in earlier, while management is still choosing between options.
Part of that is simply because producing the numbers has become easier and cheaper. The value has moved away from preparing the data and towards explaining what it means.
But I think the bigger reason is trust. Finance earns trust when it brings the bad news first instead of hiding it until the end of the quarter.
What I find most interesting now is that finance increasingly helps create the rules, not just measure the result. In my current role, I am working on a bonus-allocation model. That is also resource allocation, except the resource is people’s money.
Finance is no longer only counting the outcome. It is helping decide what behavior the company chooses to reward.
