Ask ten freelancers how they categorise expenses and you'll get two answers, both wrong. Half use a single bucket — "business" — and learn at tax time that a year of undifferentiated spend tells you nothing, and tells your accountant less. The other half build forty categories, then spend longer deciding where a receipt goes than it took to earn it. The useful answer sits in between, and it's worth getting right once, because categories are the difference between a tax season you sort and one you reconstruct.
What a category is actually for
A category isn't filing for its own sake. It exists to answer two specific questions, and you should judge every category by whether it earns its place against them:
- What's deductible, and how? Your accountant groups your spend onto tax-form lines — Schedule C in the US, the relevant heads of your profit-and-loss in India. Categories that map cleanly onto those lines save real preparation time and catch deductions you'd otherwise miss.
- Where did the money actually go? The business question, not the tax one. Which parts of the business cost what — so you can see that software crept up, or that travel quietly ate a quarter you thought was lean.
A good category set answers both at a glance. If a category does neither, it's clutter.
Start with these
You don't need to invent a scheme from scratch. For most freelancers and small businesses, a short starting set covers the large majority of spend:
- Travel — flights, trains, taxis, hotels, fuel for business trips.
- Meals & Entertainment — client meals and the like, which are often only partially deductible, so they earn their own bucket (confirm the treatment with your CA or CPA).
- Office supplies — the consumables: paper, postage, small bits of kit.
- Utilities — phone, internet, the running costs of working.
- Rent — workspace, coworking, storage.
- Software & subscriptions — the line that quietly grows for anyone who works on a computer.
- Other — the honest catch-all for the genuinely one-off, so you're never stuck at the counter.
Starlog ships with most of these as defaults — Travel, Meals & Entertainment, Utilities, Rent, Office Supplies, and Other — precisely so you can start categorising on day one without designing a chart of accounts first. The point of a short starting list isn't laziness; it's that a category you actually use every time beats a precise one you skip half the time.
When to add a custom category — and when not to
Add a custom category when a meaningful, recurring slice of your spend doesn't fit the defaults and you'll want to see it on its own. A photographer adds "Equipment"; a consultant who flies constantly splits "Airfare" out of Travel; an Indian freelancer might group GST-bearing professional services so the input-credit working is easy to pull. The test is simple: will this category hold several receipts a year, and will I ever want its total by itself? If yes, make it.
Don't add a custom category for a one-off. A single conference, a one-time legal bill, a tool you tried for a month — those belong in "Other" with a clear note. A category that holds one receipt forever is just a worse note. The discipline is to resist filing perfectly: "Other" plus a description is the right home for the long tail.
How many is too many
There's no magic number, but there's a reliable smell test: if you hesitate at the counter over which category a receipt belongs to, you have too many. Categorising should be reflexive. The moment it becomes a judgment call, you've built a taxonomy fine enough that you'll start applying it inconsistently — the same expense landing in three different buckets across the year, which is worse for your totals than a coarser scheme used consistently.
For most one-person businesses, somewhere around six to twelve categories is the sweet spot: enough to be useful, few enough to be automatic.
Keep them consistent — and separate, if you run two things
Two failure modes worth naming:
- Drift. "Software," "Subscriptions," and "SaaS" become three categories because you named the same thing differently in different months. Pick one name per concept and reuse it. This is one quiet advantage of an app over a spreadsheet — you pick from your existing list instead of retyping, so the list can't silently fork.
- Bleed between businesses. If you run more than one business, each needs its own category set — the restaurant's "Supplies" and the rental's "Supplies" aren't the same line, and mixing them turns two clean books into one muddle. In Starlog categories are per-business, with no cross-business bleed, for exactly this reason.
Let the categories file themselves
A category is only worth anything if every receipt actually gets one, and that's where good intentions usually die — in the month-end marathon that never happens. The fix is to categorise at capture, in the two seconds you're already holding the receipt.
That's the workflow Starlog is built around: snap the receipt, and as you confirm the store and amount you tag the category right there, helped by suggestions based on where you've shopped before. Then the totals build themselves — a category breakdown in your insights, a per-report donut for a given trip or client, and a year-end export your accountant can actually use, already grouped. The category you set in two seconds at the counter is the report you don't have to build in April.
The takeaway
Expense categories aren't bookkeeping theatre — they're the structure that turns a pile of receipts into answers about what's deductible and where your money went. Start from a short default set, add a custom category only when a recurring slice deserves its own line, keep the list small enough that tagging is reflexive, and apply it the moment you capture each receipt. Do that and tax season becomes a sort, not a reconstruction — and you'll actually know what your business spent.