A POS report is a summary your point-of-sale system builds from transactions you have already taken: what sold, what it earned, who sold it, when, and through which channel. You are not short of them. Most systems ship with dozens, and most owners open one, the daily total, and close the rest forever.
That is understandable. A reporting menu with forty entries is not an invitation, it’s a wall. So this is a routine instead: five reports, in order, roughly twenty minutes, ending in one decision you make this week. Done monthly it tells you more than a dashboard you check daily and never act on.
It matters more than it used to. 42% of operators reported their restaurant was not profitable and 60% reported softer customer traffic (National Restaurant Association, 2026 State of the Restaurant Industry). In that environment the difference between operators is often just which ones look at their own numbers.
The Bottom Line
- Open five reports in this order: sales summary, menu performance, demand by hour and channel, discounts and voids, labour against sales.
- Rank menu items by margin contribution, not units sold. The volume list and the profit list are rarely the same list.
- Channel profitability is the report most operators have never run, and the one most likely to change a decision: marketplace commissions of roughly 15% to 30% per order make revenue-by-channel misleading on its own (LetMenu, 2026).
- Finish every session by writing down one change. A report that produces no decision was entertainment.
Lower your delivery costs by 23%
How we reduce costs:
- No delivery vehicle expenses
- Optimized local routes
- Pay-per-delivery model
- Average 23% delivery cost reduction
What a POS report is, and which ones you already have
Your system records every line of every ticket. That recording is the whole job of the machine, as our explainer on what a POS system is and how it works sets out. Reports are saved questions asked of that record. The names differ by vendor, but the families are consistent.
| Report family | The question it answers |
|---|---|
| Sales summary | What did we take, net of discounts and refunds? |
| Product or menu mix | What sold, how often, at what margin? |
| Daypart and hourly | When does demand actually arrive? |
| Channel | How much came from dine-in, pickup, our site, marketplaces? |
| Payment | Card, cash, wallet, gift, and what processing cost us |
| Labour | Hours and cost against sales in the same period |
| Discount and void | What left without being paid for, and who authorised it |
| Inventory | What theoretical usage says we should have versus what we have |
Two things decide whether these are worth reading. First, whether all your orders are actually in the system. If staff re-key marketplace orders by hand or leave them on a separate tablet, your reports describe a business you do not have, which is the real argument for the POS integrations that pull every channel into one queue. Second, whether you can read them anywhere, which is one practical gain of running the POS in the cloud rather than on a back-office machine you have to be standing in front of.
If you are still assembling the stack that produces this data, what a restaurant POS does end to end is the wider picture.
Step 1: Read the sales summary for the shape, not the total
Start with the sales summary for a full period: a month, or four complete weeks so weekday patterns line up.
Ignore the headline number. You already know roughly what you take. Look instead at four lines: net sales after discounts, average ticket, transaction count, and refunds.
The useful insight lives in which of those moved. Revenue down with transaction count flat means average ticket fell: a menu, upsell or pricing issue. Revenue down with ticket flat means fewer people came: a demand or visibility issue. Those two problems have nothing in common and they look identical on a single revenue line.
Write down which one you have before moving on. Everything after this step is an attempt to explain it.
Step 2: Rank your menu by margin, not by units sold
Open the product mix report and sort by units. Then sort by margin contribution, which is units multiplied by the margin on each item. Compare the two lists.
Four groups fall out, and each has an obvious action.
- High volume, high margin. Protect these. Never run out, never bury them on the menu.
- High volume, low margin. The dangerous group, because it feels like success. Reprice, re-portion, or re-cost the recipe.
- Low volume, high margin. A merchandising problem, not a product problem. Move it up the menu or give it to servers to recommend.
- Low volume, low margin. Cut it. It is holding prep space, stock and attention.
Most systems will not compute margin unless you have entered recipe costs, and that is the single highest-return hour of setup in your POS. If you have not done it, estimate the top twenty items on paper for now. An approximate margin ranking beats an exact volume ranking.
Delivery adds a wrinkle worth checking here: items that travel badly generate refunds and complaints that never appear in the mix report. Cross-reference your worst-travelling items against the refund list.
Step 3: Find out when demand arrives, by hour and by channel
Run the hourly or daypart report, then run the same period split by channel.
Delivery and pickup demand usually peaks later than dine-in demand, and that one fact reorganises a schedule. Roughly three-quarters of quick-service sales now arrive through online or phone orders rather than at a counter (Restaurant Velocity, 2026), which means staffing to the dining-room curve is staffing to the wrong curve.
Then do the channel arithmetic the report will not do for you. For each channel, take revenue, subtract commission, packaging and the labour minutes to pack it. Marketplace orders at 15% to 30% commission can be your largest channel by revenue and your smallest by contribution.
That calculation changes decisions. It tells you whether to push your own ordering channel harder, whether a delivery minimum needs to rise, and whether certain menu items should be priced differently on marketplace menus.
Step 4: Audit discounts, voids and refunds for patterns
This is the report nobody enjoys and everybody needs. Pull discounts, comps, voids and refunds for the period, grouped by employee and by hour.
You are not looking for a total, you are looking for a pattern. A manager comping the same amount every Saturday is a process. One employee with triple the void rate of anyone else is either badly trained or something worse. Voids clustered in the first ten minutes of a shift usually mean the ordering screen is confusing, not that staff are careless.
Three checks cover most of it.
- Void rate by employee, compared against the team median rather than an absolute threshold.
- Refunds by reason code, which is where delivery problems surface as cold food, missing items and late arrivals.
- Discount totals as a share of gross sales, tracked month over month. A rising share with flat traffic means you are buying your own volume.
Step 5: Compare labour to sales, then commit to one change
Finally, put labour hours and cost against sales for the same hours. Most systems will produce labour as a percentage of sales by daypart.
Look for the hours where the two curves disagree: staffed hours with thin sales, and busy hours running short. Shifting two hours of coverage from one to the other usually costs nothing and fixes service quality in the window that actually generates reviews.
Then close the session properly. Write one sentence: the change you are making, the number you expect it to move, and the date you will check. One change, not five. Five changes in a month make the next month’s report unreadable, because you will not know which one worked.
What to do when the reports contradict each other
It happens, and it usually means data is entering your business in more than one place. Marketplace orders entered by hand. A separate card terminal nobody mentions. Reservations and covers tracked in a platform that does not write back, which is worth checking when you are choosing a restaurant reservation system.
The fix is almost never a better report. It is one fewer place where data lives. Before you buy a reporting tool, close the gaps feeding the reports you already have.
There is one case where your own POS cannot answer the question: when you also sell wholesale into retail accounts and need to know what moved off someone else’s shelf. That is a separate category of software. Alloy.ai, for instance, unifies retailer POS sell-through and inventory data for consumer brands, with custom pricing on an annual contract. Different problem, different tool; your restaurant reports will never contain that data.
Make the routine stick
Twenty minutes once a month, same day each month, with the five reports in a saved folder. Two practical aids: most POS systems will email a scheduled summary so the session starts with the file already waiting, and most of this is readable on a phone between shifts, a habit covered in the apps worth keeping on an owner’s home screen. If your current setup makes reports hard to reach from the floor at all, that is one of the signs it is time to switch to an iPad POS.
Frequently asked questions
What is a POS report?
A summary your point-of-sale system generates from transaction data, showing what sold, what it earned, and how inventory, staff and channels performed over a chosen period. Common types include sales summaries, product mix, labour, discount and void, payment and inventory reports.
Which POS reports should I check first?
The sales summary for the shape of the period, then product mix sorted by margin. Those two answer most questions about why a month was good or bad. Hourly and channel reports come next.
How often should I review POS reports?
A monthly session of twenty minutes beats a daily glance. Daily totals tell you about yesterday; a month of data tells you about patterns, which is where decisions live. Check the daily summary for exceptions only.
Why do my POS reports not match my bank deposits?
Usually timing or a second payment path. Deposits lag by a day or two, processing fees come out before settlement, and tips and refunds settle on their own schedule. If a gap persists after accounting for those, look for transactions taken outside the POS.
Can POS reports show delivery profitability?
They show revenue by channel. Profitability needs you to subtract commission, packaging and packing labour yourself, or a reporting integration that does it. It is worth the arithmetic, because it is the figure most likely to change what you do next.
Start with one month
Pick last month, run the five reports in order, and give yourself twenty minutes. The first session is slower because you are finding where your system hides things. The second takes ten minutes.
The point is not the reports. It is leaving each session with one written change and a date to check it. That is what separates an owner who has data from an owner who uses it.
