A busy last half-hour can make staying open later look obvious. Yet some customers might simply move their usual purchase to the later slot, and serving for another hour can extend preparation, staffing and closing work by more than an hour. The decision needs a small, observable test.
Start with one proposed service window and a written question: would this change add enough contribution to justify the resources it uses? The method below reviews an extension to existing hours. It is not a complete sales forecast or a formula for choosing legal trading hours. All figures are original fictional examples.
Define the change and the evidence you need
Specify the outlet, dates, days of the week, current hours and proposed hours. State the menu and channels available in the extra window, plus the last-order time and when the team can actually finish. Choose a review date and name the person who can continue, revise or stop the trial.
Record completed orders and net revenue using the same treatment of discounts, refunds and taxes as your existing reports. Keep each transaction in one time window. A transaction count is not a guest count, and requests made while you were closed are evidence of interest rather than completed sales.
Read the existing pattern without assuming new demand
Review comparable days and the periods around the proposed change. Note promotions, events, holidays, stockouts, weather and unusual group orders. A full queue at closing may show a service constraint; it does not tell you how many customers would buy during another hour.
Ask what customers wanted to buy and when, and keep those accounts separate from observed transactions. Square's guidance on reviewing operating hours recommends examining sales and costs by hour and day, considering customer and staff needs, and tracking the result of a change. Those general principles are useful without assuming any particular software is installed.
Count the costs that change with the decision
For an extension, record the ingredient, packaging and transaction costs attributable to the extra-window sales. Then add the paid staffing, utilities and other recurring costs caused by the extension. Include additional preparation and closing time, even when it falls outside the public opening window.
Keep cost categories separate so an invoice or paid hour appears once. Do not allocate unchanged rent to the extra hour and call that an avoidable cost. Equally, an unchanged payroll bill does not prove that staff have spare capacity: record work displaced, overtime exposure, supervision and owner time for the decision review.
Record one-off trial costs separately. A pilot can consume cash while its recurring operating result is positive. Neither result is the café's total profit, and the calculation does not approve staffing, working arrangements or trading permissions.
Worked example: an extra evening hour
A fictional café tests 17:00–18:00 over eight specified services. It records 40 completed orders and INR 8,000 net revenue in that window. The cost records show INR 3,200 attributable variable costs, INR 3,000 additional paid staffing including added preparation and closing work, and INR 1,000 other recurring costs. These are totals for the whole trial period.
The owner also records an explicitly uncertain estimate of INR 500 contribution displaced from existing service. This means foregone revenue elsewhere less the costs avoided on those sales, not another deduction of the full displaced revenue. The recurring incremental result is INR 300. Subtracting INR 1,200 of one-off trial costs produces a trial result of minus INR 900.
| Original fictional trial, eight services | INR | Treatment |
|---|---|---|
| Net revenue in added window | 8,000 | Recorded sales |
| Attributable variable costs | 3,200 | Subtract once |
| Additional paid staffing | 3,000 | Includes added preparation and closing |
| Other recurring costs | 1,000 | Exclude amounts already counted |
| Displaced contribution estimate | 500 | Uncertain; explain its basis |
| Recurring incremental result | 300 | 8,000 − 3,200 − 3,000 − 1,000 − 500 |
| One-off trial costs | 1,200 | Kept separate |
| Trial result after one-off costs | −900 | 300 − 1,200 |
Treat shifted sales as an uncertainty to investigate
Compare the surrounding periods and whole-day results with the recorded baseline. Customers who would have bought at 16:30 may now buy at 17:15. Those orders are real, but they are not all new business. A before-and-after difference also reflects other changes, so it cannot by itself prove the extra hour caused an increase or decline.
Keep the displacement estimate, its supporting evidence and its limitations visible. In this example, changing that estimate from INR 500 to INR 800 would reduce the recurring result from INR 300 to zero. That sensitivity is arithmetic, not evidence that either estimate is correct. If an input is unknown, leave it missing and investigate rather than entering zero to obtain an answer.
Review service quality before making the change permanent
Alongside sales, record waits, unavailable items, unfinished closing tasks and effects on the next service. Check whether the approved menu, staffing, safe operating arrangements and capacity support the extended window. A positive spreadsheet result cannot compensate for an operation that cannot reliably deliver the promised service.
Eight services is an illustration, not a recommended sample size. Choose enough relevant observations for your decision and retain unusual results with their explanations. Communicate any approved trial hours consistently at the door and across the channels customers use; record when those changes took effect.
Keep the decision and its conditions on record
At review, retain the original plan, observed results, unresolved assumptions and the authorised decision. If the trial continues with a smaller menu or different hours, create a new version rather than rewriting the earlier result. Assign remaining questions to an owner and a date.
Closing an existing window requires a separate review of revenue lost, costs genuinely avoided, sales transferred and effects on adjacent service. Do not simply reverse this extension calculation or assume every allocated cost disappears. The useful outcome is a documented decision about one specific change.
Put it into practice
Your action checklist
Your progress saves in this browser. No sign-up needed.
0 of 8 complete
From reading to doing
Opening-hours trial and incremental cost review
The Opening Hours Trial worksheet records the scope, baseline, trial conditions and cost evidence. Its single-period calculation subtracts attributable variable costs, added paid staffing, other recurring costs and an explicit displaced-contribution estimate from extra-window net revenue, then shows one-off trial costs separately. Missing, text or negative required amounts withhold the results. Manual sections retain surrounding-period observations, service effects, uncertainty and the authorised decision. It does not forecast demand, establish causation or approve opening hours.
Open the Opening Hours Trial tab in the launch workbook. Keep a copy for your own business.
Download the free workbook See what’s inside the toolkit →Sources & further reading
- How to Optimize Your Hours of Operation (Square, George Lee) ↗
Published 30 January 2023, retrieved 7 November 2026. General context for reviewing sales and costs by time period, considering customer and staff needs, communicating changes and testing results. Product features and historical market claims are not applied. The trial, incremental-cost model and figures here are original; no Indian trading-hours or employment rules are stated.
Published by Keystonne Grow. Product links connect to the Keystonne ecosystem; check the linked product or service for its current scope.