Fabric shop record keeping: what to write down, and in what unit

Six questions your shop has to be able to answer on any given day. The four records that answer them, the one decision about units that makes them add up, and how to run the whole thing in the notebook you already own.

Published 5 October 2026

Six questions a shop has to be able to answer

Record keeping gets taught as a virtue, which is why most people give it up by February. It is not a virtue. It is a way of answering six specific questions that cost you money every time you cannot answer them:

1. What do I have — of each fabric, in a unit I have actually stated?
2. What went out today, and for how much?
3. Who sold it, and when?
4. What did it cost me, so I know what I made?
5. Who owes me, how much, and since when?
6. What should be in the drawer at the end of the day?

Every record below exists to answer one of those six.

If a line you are writing answers none of the six, stop writing it.

The decision that makes everything else work

Fabric arrives in bales, is bought by the piece, sold by the yard, and — for curtain work especially — quoted in metres, because that is how it was imported. Four units for one cloth. A shop whose figures do not add up is usually not being robbed; it is often writing in four units and saying which one on none of the lines.

So make one decision before anything else, and make it once:

• Pick one unit of record. For most Nigerian fabric shops that is the yard.
• Write down what a piece of each fabric is, in that unit, and keep that definition with the fabric. A piece of lace that runs six yards is six yards every time it is counted, by everybody.
• Convert at the door, once. Metres become yards when the stock arrives and never travel through the shop as metres. One metre is 1.0936 yards; one yard is 0.9144 metres.
• Write the unit, in words, on every line. “8 pieces”, not “8”. This single habit removes the error that hides longest, which is a counting mistake that looks exactly like theft.

The four records

Four books, or four sections of one book, or four tabs. What matters is the fields, not the stationery.

1. What came in

Every delivery gets a line: the date, the fabric, how much in your unit of record, what it cost per unit, who you bought it from, and who received it into the shop.

The cost is the field people skip, and it is the one that decides whether you can ever answer question 4. Write what you actually paid for this delivery, at the time of the delivery. When the same lace costs you more in March than it did in January, you want both figures, not an average you invented later.

2. What went out

Every sale gets: the date and the time, the fabric, the quantity in your unit, the price, how it was paid, the customer's name if it was credit, and the name of whoever recorded it.

The time and the name look like bureaucracy on a Tuesday and turn out to be the whole point in November, when a figure is short and you need to know which day and which shift it went. A sale with no name and no time cannot be asked a question.

3. Money out

Fuel for the generator, transport, whoever carries the bales, rent, a stapler. If money leaves the drawer it gets a line, with the date and what it was for.

This is the record you are most likely to think too small to bother with, and skipping it is why the closing count never balances. Nothing you do to the sales side will fix a drawer that pays for diesel without writing it down.

4. What you are owed

Name, phone number, what went out, when, how much, and what has been paid against it since. Update it the day a payment comes in, not at month end.

Credit given in good faith and recorded nowhere is a hole most shops never total. It does not feel like a loss because everybody involved remembers it — right up until the month when they remember it differently.

What not to write down

A record you do not use is a tax on the person keeping it, and the first thing to be abandoned when the shop gets busy. You do not need customer addresses, you do not need a description of the weather, and you do not need a general remarks column. Keep the fields above, keep them short, and let the record earn its keep by answering the six questions and nothing else.

A rhythm that survives a busy Saturday

• Every day, five minutes at close. Count the drawer. Compare it with what the day's lines say should be there. Write both figures, including the difference, especially when the difference is not zero.
• Every week, one fabric. Physically count one — your fastest mover, or whichever one you have a feeling about — and compare it with what the records expect. One fabric a week means every fabric gets counted a few times a year without ever blocking a day.
• Every month, two numbers. What you sold against what it cost you, so you know your margin rather than your turnover. And the total owed to you, read out loud, because a list you never total is a list you are not really keeping.

Three things that undo good records

1. The unit drifts. One line in pieces among forty in yards, and a whole month of counting stops meaning anything.
2. Nobody's name is on the line. Then a gap is a mystery rather than a question, and the mystery usually gets pinned on whoever is standing closest. How to tell a counting mistake from theft.
3. A payment is recorded that was never money. A transfer that did not land is a sale on your books and cloth out of your shop. How to spot a fake alert.

When paper stops being enough

Paper is not a lesser system and anyone who tells you otherwise is selling something. One person, one shop, a few lines a day — a notebook is faster than any phone and it does not need a network.

Three things change that. Somebody else starts recording, and you now need to know who wrote which line, from somewhere other than the shop. You open a second location, and one book cannot be in two places. Credit grows past what you can hold in your head, and the owed list needs totalling rather than reading. Until one of those three is true, improve the notebook rather than replacing it.

Questions traders ask

General information for traders, not legal, tax or accounting advice.

How long do I need to keep my records?

Six years is the number to plan around. Nigerian tax practice has long worked to a six-year window for going back over a business's figures, and that window falls away where there has been fraud or wilful default. Be careful with what you read elsewhere on this: the framework those rules sat in was rewritten by the 2025 tax acts, which took effect in January 2026 and renamed the revenue service, so older pages citing particular sections may no longer be current. How any of it applies to you depends on how your business is registered, so take it to your accountant. The practical reading for a shop: a year of notebooks is not an archive, and burning last year's book to save space is the wrong economy.

Is WhatsApp a record?

No, and treating it as one is a common and expensive mistake. WhatsApp is a conversation. You cannot total it, you cannot sort it by fabric, you cannot count stock from it, and the one message you need in six months is somewhere in nine thousand. Use it to tell people things. Do not use it to hold what the shop knows.

My records are already a mess. Where do I start?

Do not try to reconstruct the past; you will spend two weeks on it and get something you cannot trust anyway. Draw a line at today, and start with one fabric rather than the whole shop — your fastest mover. Count it, write the count with its unit in words, and call that its opening position. Add one more fabric each week. What happened before today is history you can look at, not a ledger you can fix.

Do I need all this if I am alone in the shop?

Yes, but for different reasons. Alone, you are not recording to keep anyone honest. You are recording so you know which fabric actually earns, what you paid for the stock sitting upstairs, and who owes you — three things nobody carries accurately in their head for long.

What about cloth I take home myself?

It gets a line like everything else, with the date and the yards, marked as yours. Owners resist this, and it is the single thing that decides whether the record is a system or a suspicion. A count that excludes you is a count nobody else in the shop will take seriously, and it will be short by exactly the amount you took.

Two people write in the same book and it is chaos. What do I do?

One book, one line per movement, and a name at the end of every line — not a second book. Two books for one shop is how a shop ends up with two different truths and no way to choose between them. If the handwriting is the problem, rule the page into columns, so that a line cannot be written without its unit and its name.