What Your POS Export Is Trying to Tell You
By PlainSight — Insightful Actions · Updated July 2026 · ~7 min read
Every point-of-sale system exports roughly the same bones: when a sale happened, what was sold, for how much, and by whom. That is enough to answer the questions most owners guess at — which hours actually make money, which items are worth their menu space, and where margin is quietly leaking. Here is how to read it.
Getting a usable export
In your POS, look for Reports, then a sales, transaction, or item-detail export to CSV or Excel. Two exports are worth having: an order-level file (one row per check or ticket) and an item-level file (one row per item sold). Order-level answers questions about checks and traffic; item-level answers questions about menu and mix.
A caution about gross versus net. Some exports show gross sales before discounts, comps, and voids; others show net. Check which you have before comparing months, because a change in reporting looks exactly like a change in business.
What to look for
1. Sales by daypart and day of week
What good looks like: a repeatable pattern you can staff against.
Group sales by hour and by weekday. This single cut usually pays for the exercise: it tells you where you are overstaffed against demand, which slow window is worth promoting, and whether a new offer actually moved the hours you targeted.
2. Average check (or average basket)
What good looks like: stable or rising, tracked separately by daypart.
A blended average hides the story. Lunch and dinner, or weekday and weekend, behave differently enough that mixing them tells you nothing. Read them apart and you can see whether an add-on push is really working.
3. Transaction count versus revenue
What good looks like: the two moving together.
When revenue holds but transactions fall, you are being carried by bigger checks from fewer people — fragile. When transactions hold but revenue falls, discounting or downward mix shift is at work. The divergence is the insight, not either number alone.
4. Item mix — your top and bottom sellers
What good looks like: a short head of strong performers and a manageable tail.
Sort items by units and by revenue; they rank differently and the gap matters. High-volume, low-margin items drive traffic. Low-volume, high-margin items justify their place only if they sell at all. Anything low on both is menu space you are paying to print.
5. Discounts, comps, and voids
What good looks like: a small, stable share of sales with an explainable reason code.
This is where margin leaks without anyone deciding to let it. Track discounts as a percentage of gross and watch the trend. A creeping rate usually means a policy that was meant as an exception quietly became the default.
6. Payment mix
What good looks like: a mix you have actually priced for.
Card, cash, and third-party delivery carry very different effective costs. If delivery orders have grown into a serious share, your real margin is lower than your menu prices suggest, and pricing for that channel deserves its own look.
7. Sales by employee or register
What good looks like: differences explained by shift and station, not by habit.
Check average check by server or cashier. Consistent gaps usually reflect who is offering the add-on, the upsize, or the second round — which is coachable, and cheaper than any marketing you could buy.
Four leaks that hide in plain sight
- Discount creep — a percentage point or two of gross, every month, is a real salary over a year.
- A dead daypart you keep fully staffed — labor is your most controllable cost and the hardest to notice wasting.
- A long tail of items that barely sell — each one adds prep, inventory, waste, and complexity for almost no revenue.
- Growth that is entirely one channel — especially a third-party platform that owns the customer relationship and takes a cut.
A simple monthly routine
- Export the last full month at order level and item level.
- Read dayparts, average check by daypart, and the discount rate.
- Rank items by revenue and by units; note anything low on both.
- Pick one change — a staffing shift, a discount rule, a removed item, one add-on to push.
- Next month, check that number specifically before looking at anything else.
Let PlainSight read your POS export
Upload your point-of-sale export and PlainSight detects the columns automatically, surfaces dayparts, item mix, average check, and top sellers, benchmarks you against your industry, and writes the next steps in plain English — all in your browser, with nothing uploaded.
Try it free on your own numbers →
Frequently asked questions
- Which POS systems does this work with?
- Any system that exports CSV or Excel, which is effectively all of them. The column names differ but the underlying fields — date, item, amount, employee — are consistent enough to read the same way.
- My export has dozens of columns. Does that matter?
- No. You only need a handful: a date or timestamp, an amount, and ideally an item and an employee. Extra columns can be ignored.
- Should I use gross or net sales?
- Be consistent, and know which one you are looking at. Net is usually the truer read of what you kept; gross plus a separate discount rate tells you where it went.
- Is my sales data private if I use PlainSight?
- Yes. Your file is processed entirely in your browser and never uploaded to a server. Optional AI features send only anonymized summary totals.
This guide is general information for business owners, not financial, tax, or legal advice. Figures described as “typical” are rules of thumb, not guarantees — always read your own numbers in context.