Cash Flow Management for Visa Agencies

A visa agency's profit and loss can look healthy while the bank account tells a different story. The reason is structural rather than a sign of bad management: a large share of the money moving through a visa agency was never yours to spend.

Three things create the gap. Money collected for government fees sits in your account before it goes out. Work is often done days or weeks before it is paid for, or paid for long before it is delivered. And the workload is seasonal, so a strong quarter is followed by a quiet one in which fixed costs carry on regardless.

None of these is exotic. Together they explain most of the moments where an owner looks at a profitable month and still cannot make payroll comfortably.

The fix is not complicated either. It is mostly about separating three kinds of money — pass-through, earned, and not yet earned — and always knowing which one you are looking at when you check the balance.

When a client pays a total that includes the embassy fee, your balance rises by the whole amount. Only part of it is income. The distinction between a government fee and a service fee is an accounting distinction as much as a pricing one: the government portion is a liability from the moment it lands.

The amounts are not trivial. A US MRV fee across a family group, or the UK's immigration health surcharge on a multi-year work visa, can be several times your service fee on the same case. An agency handling a hundred cases a month can be holding a substantial pass-through balance at any moment, and it looks exactly like cash.

The classic failure is spending it. Not deliberately — it happens by looking at the balance, deciding you can afford a hire or a marketing push, and forgetting that a meaningful slice is committed to fees going out over the next three weeks.

Two habits prevent it. First, never record a blended total: keep the government portion as its own line on the invoice and its own field in your records, so the reportable number is service revenue rather than gross collections. Second, once volume justifies it, hold pass-through funds in a separate account and pay fees out of that. It costs nothing and makes the mistake structurally difficult.

Be careful in the other direction too. If you pay a government fee on a client's behalf before they have paid you, you are lending them money — and if the case is abandoned, you rarely get it back.

The default in visa services should be payment before work starts. Not because clients are untrustworthy, but because the alternative funds other people's applications out of your working capital.

Where full upfront payment is not realistic — high-value cases, corporate clients on invoice terms, long-running residence applications — take a staged structure. A common shape is a deposit at intake covering the government fee plus your assessment time, then the balance at submission. The principle is that at every point in the case, the money received covers the money already spent and the hours already worked.

Corporate clients are their own category. They will ask for thirty or sixty day terms and they are usually good for it, but they will also pay late in the exact month you need the money. If corporate work is a growing share of revenue, model it explicitly: revenue invoiced in September and collected in November is not September cash. Ask for a deposit on the first engagement with any new corporate client, whatever their size.

Make paying easy or the terms do not matter. Cards, local payment methods and a link the client can pay from a phone all shorten the gap between agreeing and collecting — collecting visa payments online with an itemised checkout removes the most common excuse for delay. A quote sent as a document with bank details at the bottom collects more slowly, every time. If you sell add-ons, an e-commerce style checkout also captures them at the moment of highest intent rather than in a follow-up conversation nobody gets round to.

Refunds are the most reliable source of cash-flow surprises in visa work, because they arrive after the money has been spent and usually after the government fee is gone for good.

Start from the fact that most government fees are not refunded on refusal. A US visitor visa refusal under section 214(b) does not return the fee, and the same principle applies across most categories: the authority charged for considering the application, not for approving it. If a client expects their whole payment back after a refusal, somebody set that expectation badly.

So write the policy before you need it, and write it by stage. A typical structure: full refund of the service fee if the client cancels before you have started work, a partial refund once assessment and document review are complete but nothing has been submitted, and no service-fee refund after submission because the work is done. Government fees are refundable only to the extent the authority refunds them, which is usually not at all.

Then put it in front of the client at checkout rather than in a terms page nobody opens. The refund conversations that go badly are almost never about the policy itself — they are about a client who genuinely did not know. A checkbox at payment plus one line restating it in the confirmation email prevents most of them.

Decide separately what you offer as a gesture. Some agencies offer a free resubmission after a refusal instead of a refund, which costs hours rather than cash and often produces a better outcome for everyone. Whatever you choose, choose it in advance — refund decisions made case by case under pressure are inconsistent, and clients compare notes.

A chargeback is a refund taken from you by the card network, usually months later, often with a fee attached, and typically decided on documentation rather than fairness.

In visa services they cluster around one moment: a refusal. A client who has just been refused, is out of pocket, and believes they were promised more than they were told is the profile. The second cluster is slow cases, where somebody with no visibility on progress concludes that nothing is happening.

The defences are unglamorous. Keep accepted terms tied to the transaction. Keep the itemised quote showing what the government fee was and what your service fee covered. Keep the message history — a thread showing you sent the document list, chased twice, submitted on a stated date and relayed the outcome is the single most effective piece of evidence you can produce. If that history lives in somebody's personal messaging account, you effectively have none.

Prevention beats evidence. Never promise or imply an approval; describe your service as preparation and submission. Send status updates so nobody has to wonder. Make a refusal a conversation rather than an email, and offer the next step immediately. Most disputes are opened by people who felt ignored rather than people who were cheated.

The mechanics of taking money safely — processor choice, itemised transactions, receipts and dispute handling — sit alongside this in our payment processing guide for visa agencies.

Visa demand is seasonal almost everywhere. Student intake deadlines, summer travel, religious travel seasons and holiday periods create months of several times normal volume, followed by months where the phone is quiet and the rent is not.

This makes a monthly profit figure misleading in both directions. The busy month overstates how well the business is doing, because a chunk of that cash pays for work still to be delivered. The quiet month understates it, because fixed costs continue while revenue pauses.

A rolling thirteen-week cash forecast fixes most of this, and it does not need to be sophisticated. One row per week: expected collections, government fees payable out, payroll, rent and software, marketing. Update it every week. What you are looking for is the week where the line goes negative — you can almost always fix that two months out and almost never two days out.

The other seasonal habit is to fund the quiet season from the busy one deliberately. Decide what proportion of peak-month service revenue stays untouched and move it out of the operating account the same week it arrives. Agencies that plan their peak season operationally often forget to plan it financially, then spend the surplus on the extra capacity they needed during the peak just as demand falls away.

At small scale, cash discipline is four habits rather than a finance function.

Separate the money. An operating account, a pass-through account for government fees, and a reserve. Three accounts cost nothing and remove an entire category of mistake.

Look weekly. A fifteen-minute review of cash in, cash committed and the next four weeks. Monthly is too slow to act on; quarterly is a post-mortem.

Hold the buffer in months, not amounts. Three months of fixed costs is the usual target for a small agency. In a seasonal business, measure it against quiet-month costs rather than your average.

Invoice and chase on a schedule. Late collection is a cash-flow problem disguised as an admin problem. Corporate invoices with nobody chasing them go unpaid for months, while a fixed cadence — a reminder on the due date, then day seven, then day fourteen — collects most of it without a difficult conversation.

None of this is advanced finance and none of it needs an accountant to start. What it does need is the numbers being visible without a day of reconstruction, which is where most small agencies actually fail.

Cash management gets easy when the data is already there. A few things are worth setting up deliberately rather than reconstructing later.

Record the government fee and the service fee as separate fields on every application, so service revenue is reportable without stripping out pass-through amounts by hand. Record when fees were paid out, not only when they were collected — the gap between those two dates is your exposure.

Track payment status per case as a stage, so unpaid or partly paid cases are visible in the same view as the case itself. An application sitting at submission with a balance outstanding should be impossible to miss.

Flag refunds and chargebacks as their own outcome instead of editing or deleting the original payment, so you can see refund rate by visa type and by market. A visa type with a high refund rate is telling you something about how it is being sold.

Keep the message history attached to the case, permanently and centrally. It is your dispute evidence, your onboarding material and your quality record all at once.

These are the financial cousins of the KPIs every agency should track, and the reason to instrument them is simple: the alternative is finding out during a bad week. If you want payments, government fees, refunds and per-case financials in one place instead of three, book a demo.