Managing digital files: how to name them, and how to set up your folders

Hook

Boss, a test: that TNB bill from three months ago — you have two minutes, can you find it? If you cannot, the problem is not your memory. It is that your digital files are not being managed. Today we cover three things: how to name files, how to set up folders, and which traps to stay out of.

The filename is the index card

Start with what matters most: the filename. "scan0001.pdf", "WhatsApp Image 123" — those are the same as having no name at all. A good filename answers four questions before you open anything: whose is it, what kind of document, when, and how much? A bill named something like "ShopRent-BossTan-Invoice-2026June-RM3500" can be found by searching the name, the expense type or the amount. Two small habits make it better still. Write dates year-month-day, so the files line themselves up in time order when sorted. And keep the punctuation simple — hyphens and dots, not asterisks or slashes the system will not accept. Naming a document properly is writing the index card for your future self.

Folders: shallow, not deep

Now the folders. One principle: shallow, not deep. Trap one, the Russian doll — Company, then Accounts, then 2026, then First Half, then June, then Utilities, then TNB. Seven clicks before you see a file. Tiring to file into, worse to find in, and when somebody else takes over they get lost immediately. LTT's rule: two levels at most before you reach the file. Trap two, a folder per month — one for January, one for February. The date is already in the filename, so a monthly folder is a redundant drawer that only scatters one supplier's bills across twelve places. Trap three, giving the same thing a new home each period — this year's customer invoices here, and next year "start fresh" with a different structure. The continuity breaks, and to look at two years of dealings with one customer you have to open two folder systems and piece it together. Set the structure once and keep it: what changes is the year, not the logic.

The properties you cannot see also matter

Finally, a layer most people do not know about: file properties. Dates — the created date the computer records is the day you scanned it, not the date of the document. So the real date belongs in the filename, and you should not trust "sort by date". Format — a phone photo is a JPEG; convert it to PDF for proper archiving. It is harder to alter, it does not distort as it gets passed around, and it prints neatly. Size — scan settings set too high give you a single bill of tens of megabytes and the archive fills up fast; clear and readable is enough. And whether the text can be searched — a scanned PDF is only an image, and the words inside it cannot be found. Run it through OCR and the whole archive becomes a full-text searchable database. The difference is enormous.

How to start

You do not have to reorganise ten years of old records in one go. Three steps. One, open an inbox folder — every new document lands there first. Two, name and file it the same day — the "never overnight" rule from the fifth lesson, applied to files. Three, tidy the old files only when you use them — fix one each time you dig one out, and in six months it will be clean by itself. Find that too tedious by hand? As the last lesson covered, LTT's AI tool FileWhiz does exactly this — automatic naming, automatic filing. Our clients do not even do this step themselves.

To close

Remember: the filename is the index card, folders stay shallow, the real date goes into the name, and scans go through OCR. Once the archive is managed, the "complete" from the first lesson has a foundation under it — LHDN asks for a document and you produce it in two minutes; you want to look at something and you search it out in two seconds. Want the archive to manage itself? Grab a coffee first and talk about your business. For the accounting, come to LTT. I am LTT, helping SME bosses get their accounts straight. Follow us, and see you next time.