Why POS Agents Need Gazeloft: Your Float Is Not Your Profit

Ask an agent how business went and the answer is usually a big number. "We did almost a million today." It sounds like success. It is not an income figure, and treating it like one is the single most expensive habit in this trade.

The arithmetic nobody does out loud

Take a normal, decent day at a busy stand:

• Withdrawals: ₦520,000 across 62 customers
• Deposits and transfers: ₦280,000 across 24 customers
• Airtime and data: ₦19,000

Roughly ₦819,000 "did" business. Now the part that is yours. Say the charges you collected across all of it come to ₦11,400. That is the income line. Everything else was passing through.

Then subtract what the work cost you:

• Terminal and provider fees on those transactions: ₦2,650
• Fuel for the generator and data for the terminal: ₦2,000
• Drawer short by ₦500 at close, nobody knows why

Real earning for the day: ₦6,250. Not ₦819,000. Not ₦11,400. ₦6,250.

An agent who believes the first number will spend from the float. An agent who believes the second will be confused about why the money never grows. Only the third number tells you whether today was worth standing there for.

Why the float punishes you quietly

Float is unforgiving because errors in it do not look like errors. They look like a normal balance. You dip ₦20,000 out of the drawer for something personal and intend to return it. Two days later the bank balance is fine and the drawer is short, and you cannot tell whether it is that ₦20,000 or a wrong entry or a customer who was overpaid.

By the time an agent feels the squeeze, it is usually not one big loss. It is many small unaccounted movements that nobody wrote down on the day they happened.

The only defence is to record the two sides of your money separately, every time: cash at hand and cash in bank. Every entry you make moves one, the other, or both. Do that consistently and a shortage shows up the same evening, while you can still remember the day.

The fees that eat the charge

The charge you collect is not the charge you keep. Depending on the transaction there is a provider fee, sometimes a debit fee, VAT on the fee, and stamp duty on certain amounts. Individually they are small. Against a ₦100 charge they are not small at all.

Gazeloft itemises them per transaction type, so a withdrawal, a deposit and a transfer each carry their correct cost, and you keep a shared list of your provider's fee presets instead of re-typing them. The result is that your profit line is what actually remains, not what you hoped to keep. If a certain transaction type is costing you more than it earns, you will see it in a week rather than in a year. Read the detail in how terminal charges are recorded.

Withdraw and send, in one honest entry

Every agent knows this situation. The customer wants ₦50,000. You withdraw it, but you give them ₦45,000 cash and send ₦5,000 to their account, or they leave some with you. Key that as two separate transactions and your books will never balance, because your drawer and your bank moved by different amounts than the sale suggests.

Gazeloft records it as one transaction that carries both figures: the amount sent and the cash actually handed over. The charge is typed by you, never quietly assumed. Your drawer is right, your bank is right, and your charge is counted once.

What the evening looks like

At close you open the app. It shows the day's charges, the fees against them, expenses, profit today, and your cash at hand next to your cash in bank. You count the drawer. The two numbers agree, or they do not, and if they do not you know tonight instead of next month. That is the whole end of day routine, and it takes about a minute.

Over weeks, the app also starts answering questions you never had time to ask: which hours are actually busy, what share of your business is withdrawals versus transfers, which customers come back. Agents use that to decide when to open, when to rest, and when to top up the float before the rush.

If you have staff

A second stand or an assistant multiplies the same problem. Give the worker their own login with only the permissions they need, and tag their cash to them. At the end of a shift you can see what passed through that person's hands without an argument and without an accusation. Records protect an honest worker as much as they expose a dishonest one, and most workers prefer it that way.

Start with one day

Do not migrate anything. Tomorrow morning, record your opening cash at hand and your bank balance, then record every transaction as it happens for one day. At close, compare the app to your drawer.

Whatever the gap is, that gap has been there all along. You are just seeing it for the first time.

Open a free POS space and measure one day properly.