nouz lets you record a day's revenue in two completely different ways — and most owners use both, often on the same day. The reason isn't a design choice we made for fun: it's that the rhythm of a real shop doesn't always fit a neat product catalogue. Some days are clean (every sale is a known SKU at a known price). Other days are messy (a friend came by, a delivery driver paid €5 cash for a coffee, a regular asked for a custom drink off-menu). The two entry modes exist to cover both cases without forcing you to invent fake products just to log a number.
Which mode you reach for first depends on how mature your catalogue is, how busy the day was, and what you want to learn from Statistics later. The trade-offs are real but the cost of getting it wrong is small — you can always edit later, and the P&L math handles both modes identically once the entries are in.
01 The two modes
In nouz, every revenue row falls into exactly one of two categories:
- Manual entry — type a single number per cash + card column. No product reference. Fastest path, fewest fields. Two boxes (BAR and EC), maybe a note, and you're done. The trade-off: nouz can't compute per-product margin from a manual entry because it doesn't know what was sold.
- Product sale — pick a product from your catalogue, type a quantity, choose whether it was cash or card. nouz reads the snapshot price and COGS from when you set the product up, and fills in revenue + margin automatically. More setup-required up front, but every sale you log teaches nouz one more thing about your business.
You can find both on the same screen: open Revenue, pick a date, and you'll see two buttons — + Manual entry and + Product sale. They open different forms but write to the same date.
02 When to use manual
Manual entries are the right tool when the alternative — setting up a product just to log a single sale — is more friction than it's worth. Four common scenarios:
- Your first week. You haven't built your product catalogue yet. Don't let perfect setup block you from logging the day. Two numbers (cash and card) take ten seconds and your P&L starts working immediately. You can backfill products later — they'll apply going forward without disturbing the manual entries you logged in week one.
- Service revenue you don't track per-treatment. A hairdresser charging €45 for a cut doesn't need fifteen "haircut variant" products in their catalogue. One daily manual entry, summing all the cuts, is more honest and more maintainable than a fake product called "Generic haircut".
- The day was chaotic. Sometimes you only have the till total — maybe the POS export had a glitch, maybe a staff member forgot to ring up individual items and just took cash. A manual entry for the day total lets you record what you know, when you know it.
- One-off transactions that don't fit a SKU. A friend's wedding got you a €60 catering tip. A delivery driver paid €5 cash. The corner shop owner came by with a single bottle of wine you let them take for €20. None of these need permanent products in your catalogue — a manual entry captures the revenue cleanly.
The cost of using manual entries is that Statistics can't rank these revenue contributions by product — they're aggregated. If you find yourself wanting to know "which menu item is carrying us this month", that's the signal it's time to start logging the high-volume items as product sales.
03 When to use product sale
Product sales are the right tool when you sell the same thing repeatedly and want to learn from how it performs. Four scenarios:
- Your top-five sellers. The cappuccino you sell forty of every day. The sandwich that pays the rent. The haircut style booked three times a week. Setting these up as products takes five minutes once and saves you typing forever — type the quantity at close-out and revenue + COGS auto-fill.
- Anything where margin matters. If you're ever going to ask "is this product worth keeping on the menu?", you need product-level data. Statistics can only rank products it has data for — so things you don't set up as products are invisible to the margin-drift detector.
- Items with different tax rates. Coffee in-house at 20% VAT, coffee takeaway at 10% VAT. Setting them up as two products lets the till staff pick the right one at close — the per-product tax-rate override does the math automatically.
- Price experimentation. If you ever want to try raising a price by 50 cents to see what happens to volume, you need historical product-level data to compare against. That data only exists if you've been logging product sales.
04 Why they coexist safely
Under the hood, nouz stores the two entry types in different database tables: manual entries land in revenue_entries, and product sales land in revenue_product_entries. Both tables are scoped by location and by date. When the P&L computes a day's number, it reads both tables and sums them together — no double-counting because each row exists in exactly one table.
A worked example: Tuesday's entries.
Three product sales summed to €267,60. One manual entry added €48 of catch-all. The P&L reads them both and lands on €315,60.
The only failure mode is conceptual, not technical: if you log a manual entry for "the day's revenue" and separately log individual product sales that overlap with it, you'll count the same revenue twice. The math doesn't catch that — it just sees two rows summing to a higher total than reality. The fix is discipline: pick one mode per "chunk" of revenue, not both.
05 The mixed pattern most owners use
After a few weeks, most owners settle into a hybrid. They log product sales for the things they sell often and care about analytically — the cappuccino count, the sandwich count, the haircut count. Then they add one manual entry per day for everything else — tap water tips, the one-off €5 cash thing nobody can remember, the customer who paid for a regular's coffee on their way out.
This pattern gives you the best of both worlds: rich product-level data on the items that matter, and zero friction on the long tail. Statistics has enough to rank your catalogue meaningfully, and your nightly close-out stays under two minutes.
Was this article helpful?
Your vote helps us decide what to write next.