POS reports a manager should read daily

A point of sale can produce dozens of reports, and most managers read almost none of them. The end of the night is late, the team wants to go home and the reports screen offers sales by category, by department, by terminal, by modifier and by hour, each with its own filters. It is easier to close the till and go.

That is a pity, because a handful of POS reports, read at the right moment, catch most of the problems a venue faces: a drawer that does not add up, a server voiding too much, a slow hour that needs fewer staff, a dish that has quietly stopped selling. The trick is knowing which few to read, and what each one should make you do.

This guide sets out four reports to read at close and three to read on a Monday morning, what each one tells you, what to do about it, and which of the rest can wait. Every venue and figure in it is an example.

Most POS reports go unread for the same reason: there are too many, and none of them says what to do. A manager faced with forty reports at midnight reads none. A manager who knows that four of them matter, and what to look for in each, reads those four in ten minutes.

The aim of restaurant reporting is not to know everything. It is to notice what needs action, while there is still time to act. Our guide to restaurant reporting: six reports, no more makes the wider case for reading few reports well.

Choose your reports by the decisions they support. If a report never changes what you do, stop reading it.

At the end of each service, four reports cover what the manager needs to know before leaving: the takings against a comparable day, the payments by method, the voids and comps, and anything still open.

Together they take ten minutes. They tell you whether the day went as expected, whether the money is all there, whether anything was given away that should not have been, and whether anything was left unfinished. Each has a clear action if something is wrong.

Read them every night, even when you are tired. The night you skip them is the night something goes wrong.

The first report is the day's takings, compared with a comparable day: the same day last week, or the same day last year. A number on its own tells you little; a number against a comparison tells you whether the day was normal.

Look for differences you cannot explain. A Saturday a fifth below last Saturday, with no change in bookings or weather, needs a question. A Tuesday far above normal may point to a large party you should plan for again, or to a mistake in the figures.

The action is simple: if the difference is explained, note why; if it is not, find out tomorrow.

The second report shows payments by method: cash, card, transfer and anything else you accept. These payment reports are how the manager checks that the money is all there.

Compare each method with what you actually hold. Card payments should match the card terminal's total for the day. Cash should match what is in the drawer, less the float. A difference of a few pence is normal rounding; a difference of a large amount needs finding before the end of the night, while the team still remembers.

The action is to reconcile every night. A gap found the same night is usually a mistake that can be traced; one found at the end of the month rarely is.

The third report lists the day's voids and comps: dishes removed from checks and dishes given away. It should show who did each one and why.

Voids and comps are normal: a dish entered twice, a steak sent back, a dessert on the house for a regular. But a pattern is worth noticing: one server voiding far more than the others, voids made after a check was printed, comps with no reason given, or comps on the same guests every week.

The action is to ask about anything unusual, without accusation, the next day. Most patterns have an innocent explanation; the few that do not are far easier to deal with early.

The fourth report shows anything still open at close: checks not closed, payments not recorded, tables still marked as occupied. An open check at the end of the night is either a guest who left without paying, a payment recorded somewhere else or a check someone forgot.

Find each one before you leave. A check left open overnight is much harder to resolve in the morning, and an unpaid bill discovered a week later is usually lost.

The action is to close or explain every open item each night, so the next day starts clean.

On a Monday morning, or whichever morning starts your week, three more reports look back over the whole week: sales by hour, the menu report and the average spend and covers.

These are not about catching problems on the night. They are about planning: staffing, prep, purchasing and the menu. They take half an hour, and they shape the week ahead.

Sales by hour shows when the money comes in, hour by hour, across the week. It is the shape of your trading day: when you are quiet, when you are busy and when the rush starts and ends.

Use it to plan staffing. If Tuesday lunch is quiet from two o'clock but the team stays until four, there may be hours to save. If Friday's rush starts at seven but the second chef arrives at half past, there may be an hour to add. Our guide to sales by hour: the shape of your trading day covers how to read it.

The action is to compare the shape with your rota, and change the rota where they disagree.

The menu report shows what sold during the week, dish by dish: how many of each, how much revenue each brought in and what share of sales it made up.

Use it for prep, purchasing and the menu. A dish whose sales have dropped for three weeks may need a new description, a better place on the menu or retiring. A dish that sells out every Friday may need more prep. A dish nobody orders is taking up space on the menu and in the fridge.

The action is to adjust prep and purchasing for the week ahead, and to note any dish that needs a closer look.

The last weekly report shows covers and average spend: how many guests you served and how much each spent on average. Together they explain the week's takings: was a strong week driven by more guests or by each guest spending more?

Watch the trend over several weeks rather than one. A gradual fall in average spend may point to fewer drinks, fewer desserts or more discounting. Our guide to average spend per head, split properly covers how to read it.

The action is to pick one change to try, such as a better dessert offer or a drinks recommendation, and measure it over the next few weeks.

A report is only as good as what was entered. Every so often, check the till's figures against something physical: the card terminal's own end-of-day total, the cash in the drawer, the bank deposit, a count of the bottles behind the bar. If the till says one thing and reality says another, find out why.

Differences usually come from how things are recorded rather than from anything missing: card payments recorded as cash, a check closed on the wrong table, a dish rung in under the wrong name. Each is easy to correct once found, and finding them keeps the reports trustworthy, which is the only reason to read them.

Many POS reports can be split by member of staff: sales, average check, voids, comps and discounts. Used well, they show who might need help, who sells well and who could share how they do it. Used badly, they become a league table that sets the team against each other.

Read them with care. A server with lower sales may have had the quietest section; one with more voids may be newer, or may simply work the busiest tables. Look for large, lasting differences rather than small ones, and talk about them privately and fairly.

Some figures are worth a glance during service rather than after it: how many covers have been seated, how many checks are open and how the evening is tracking against a normal night. A manager who sees at nine o'clock that the evening is well behind can act while there is still time: a push on desserts, an earlier close for one section or letting a member of staff go home early.

Keep the glance short. Service is for guests, and the full reading of the reports belongs at close.

Make reading the reports a routine, not a task. At close, the same four reports in the same order, every night. On Monday, the same three, at the same time. A routine takes less effort than a decision.

Keep a short note of what you found and what you did. Over a few weeks, the notes become a record of the venue's patterns, and a guide for any manager who covers your shift.

If the reports take longer than ten minutes at close, the problem is usually the system: too many screens, slow loading or figures hard to find. That is worth raising with your provider, or considering when you next choose a POS.

Keep a short record of what each night's reports showed and what you did about it: a line or two in a notebook or a shared document. Over a few months, it becomes a history of the venue's patterns, its problems and its fixes, which is invaluable when planning the next season or handing over to a new manager.

Many reports are useful occasionally and need not be read daily. Sales by category, by terminal or by department; detailed modifier reports; reports on individual guests; long-term trends over a year. Read them when you have a question they answer, such as when planning a new menu or reviewing a year.

Some reports are rarely useful at all, especially in a single venue. Do not feel obliged to read a report just because the system offers it. A report is only useful if it changes what you do.

Once a month, spend an hour on a wider view. Compare the month with the same month last year, and with the month before. Look at how payment methods have shifted, how deep discounts have run and how voids and comps have moved. Look at the menu over the whole month: which dishes carried the month, which faded and which never sold.

Monthly reports are where slow trends show: average spend creeping down, a growing share of discounted checks, a dish whose sales have halved since the season changed. They rarely need action that day, but they shape the next month's plan.

Whenever something changes, such as a new menu, a price rise, a new rota or a promotion, use the reports to see what it did. Note the date of the change, then compare the weeks before with the weeks after, using the same days and similar trading.

Without that comparison, changes are judged by impressions: the new menu feels busier, the promotion seems to work. With it, you know. A change that did not work can be undone; one that did can be repeated.

POS reports show what the till knows: sales, payments, voids, comps and dishes. They do not show everything. Bookings, no-shows, guest feedback, staff costs and food costs live in other places, and a full view of the business needs them too.

Where your systems share data, some of those can appear alongside the till's figures: covers from bookings, for example, or spend per guest. Where they do not, keep a simple weekly sheet that brings the key numbers together. Our guide to a restaurant dashboard you look at daily covers what belongs on one page.

A good report is quick to open, easy to read and clear about what it shows. It compares the figure with something meaningful, such as last week or last year. It lets you see the detail behind a figure, such as which checks made up the voids. And it uses the same definitions every time, so this week's figure can be compared with last week's.

When choosing a POS, look at the reports you will read every day, not the number of reports on offer. A system with seven clear reports is better for a busy manager than one with seventy confusing ones. Our guide to POS features that matter after the first week covers the wider list.

Every report should end with a decision, even if the decision is to do nothing. A performance indicator is only useful if someone acts on it.

Write down the action for each report as part of the routine: reconcile the drawer, ask about a void, change Thursday's rota, add prep for Friday, try a new dessert offer. Then check, a week or a month later, whether the action worked.

A manager who reads seven reports and makes seven small decisions each week runs a better venue than one who reads seventy and decides nothing.

Some figures are worth sharing with the team: the week's covers, the best-selling dishes, the drop in remakes after a change to the tickets. Shared figures help the team understand what matters and why.

Share them simply, at the pre-service briefing or on a board in the back, and avoid using them to blame individuals. A team that sees its own progress is more likely to keep it going.

Owners who are not in the venue every night still need the key figures. A cloud system lets them read the same reports from a phone, so the owner can see each evening's takings, voids and covers without waiting for a manager's message.

Agree which figures the owner will look at and when, so managers are not surprised by questions, and so the owner does not drown in detail. A few figures, read consistently, keep an absent owner close to the business.

An invented bistro's manager closes the till each night and goes home. The reports screen has dozens of options, and none are read. At the end of the month, the accountant finds the card takings short by a significant amount, and nobody can say why.

The manager starts a ten-minute routine at close: takings against last week, payments by method reconciled with the card terminal and the drawer, voids and comps by member and any open checks. On Mondays, they read sales by hour, the menu report and covers with average spend.

Within a fortnight, the routine finds the problem: card payments recorded as cash on one terminal. Within two months, the Monday reports have cut two quiet hours from the rota and added prep for the Friday rush.

In servearound, payment analytics show what was collected by day, how it was tendered and the method mix, what made up the total, revenue by hour, what was voided and comped, who collected it, voids and comps by member, and the top tables by revenue. Order analytics show checks opened, the average and median check, items and lines per check, the void and comp rates, when checks are rung across the week by day and hour, where the checks stand and the spread of check values.

The menu report shows every dish that sold, with its units, revenue and share, and the dashboard shows revenue, covers, spend per cover and the no-show rate against the previous period. Each report covers one venue for any range of dates, and bills can be downloaded as a spreadsheet. See Analytics.

Takings against a comparable day, payments by method, voids and comps, and anything still open.

Sales by hour, the menu report, and covers with average spend.

About ten minutes at close and half an hour on a Monday, if the system makes the reports easy to find.

Patterns: one person far above the others, voids with no reason, comps on the same guests every week.

A figure on its own says little. A comparison shows whether the day was normal.

Compare its shape with your rota, and change the rota where they disagree.

No. Read the few that change what you do, and the rest only when they answer a question.

The same reports, in the same order, at the same time, with a short note of what you found and did.

Most POS reports go unread because there are too many and none says what to do. Read four at close: takings against a comparable day, payments by method, voids and comps, and anything still open. Read three on a Monday: sales by hour, the menu report, and covers with average spend.

Make it a routine, write down the action each report leads to, check whether the actions worked, and leave the rest of the reports for the days they answer a question you actually have.