Choose a minimum basket by comparing the contribution from items guests add with the cost of the discount on every qualifying order. Start with a threshold just above a common order value, calculate how many guests would need to spend more, and test whether that behavior occurs often enough to pay for the offer.
A larger basket is not automatically a better order. If a guest was already going to spend enough to qualify, the discount reduces what you keep without necessarily adding a sale. The worksheet below makes that cost visible before you publish an offer.
Calculate discount cost and added-item contribution
First, define the offer. A product discount reduces the price of eligible food or drinks. That is different from a free-delivery threshold, which changes the delivery charge. Keep the two separate when you calculate results, even if your ordering setup can show both.
For a percentage discount, multiply the eligible basket by the discount rate. A 10% discount on $40 of eligible items costs $4. A fixed-dollar offer is simpler: a $4 discount costs $4 on each qualifying order, provided the full amount can be used.
Next, estimate what the added items contribute. This is their selling price minus the variable costs associated with selling them. Ask whoever manages purchasing and the menu for ingredient and packaging estimates. Ask the person managing delivery and payments for the relevant per-order or percentage charges. Use your own records where possible; averages are only estimates.
For example, an added $10 item with $4 in variable costs contributes $6 before the discount. If that addition unlocks a $4 discount, the immediate gain on that order is $2, assuming no other costs change. If the guest adds a low-margin item, the gain may be smaller or disappear.
Decide what counts toward the minimum before doing the math. Does the eligible basket include discounted items, modifiers or bundles? Are delivery charges, tips and taxes excluded? Ask your ordering provider how the threshold and discount are calculated, and use those same definitions in your worksheet.
Account for orders already above the minimum
Hypothetical worksheet: An owner is considering 10% off eligible food orders of $40 or more. Without the offer, 20 orders in a comparison period would already total $40 each. Other guests typically spend $30, and some might add $10 to qualify. The added items are assumed to contribute 60% of their selling price after variable ingredient, packaging and other incremental costs.
The table compares the offer with no discount for the same orders. It assumes each qualifying basket is exactly $40, so each discount is $4. Fixed operating costs are excluded because they are assumed unchanged. Exception costs, such as refunds, remakes and extra support time, are also excluded; include estimates for them in your own decision. This is an order-level contribution comparison, not a profit forecast.
| Orders adding $10 | Added-item contribution | Discount cost | Change versus no offer |
|---|---|---|---|
| 0 | $0 | 20 orders × $4 = $80 | −$80 |
| 10 | 10 × $6 = $60 | 30 orders × $4 = $120 | −$60 |
| 40 | 40 × $6 = $240 | 60 orders × $4 = $240 | $0 |
Read the last column as the contribution gained from added items minus all discounts paid. At 10 moved orders, the owner gains $60 from additions but gives up $120 in discounts. At 40 moved orders, the two amounts match. Under these assumptions, more than 40 such orders are needed to improve contribution before exception costs.
The shortcut is to divide the discount cost on already-qualifying orders by the net gain from each moved order. Here, $80 divided by $2 equals 40. This shortcut only works if the added amount, contribution and discount are the same for each moved order. Actual baskets will vary, so calculate by order or use realistic groups.
That break-even count is not a prediction. Check how many orders actually sit near $30 and how many guests could reasonably add an item. If only 25 orders are in that range during a comparable period, this offer cannot break even through those particular $10 additions alone. A different minimum, discount or menu mix may be worth modeling.
Also check orders that exceed $40 by a wide margin. A percentage offer costs more as an eligible basket grows; 10% of $60 is $6, not $4. A fixed-dollar offer limits the discount per order, but could still reduce contribution on orders that needed no encouragement. Compare both structures using your own basket distribution rather than assuming one is always better.
Check what guests see in the cart
Before a live offer, ask your provider whether a test environment or simulation can show the complete guest journey. A practice check must not trigger a real card payment, guest message or courier dispatch. If the setup cannot prevent those actions, ask the provider for a safe demonstration instead.
Using the hypothetical $40 minimum, try an eligible food basket at $39.99, then add an item that takes it above $40. The cart should show the minimum clearly, apply the intended discount only when the conditions are met, and display the revised total before the guest commits. Remove the added item and check that the offer no longer applies.
Then check cases that could change the calculation:
- A basket at exactly $40, including a modifier or bundle.
- An order with a delivery charge, tip and taxes, to see which amounts count toward the minimum and which receive a discount.
- An item or promotion that should be excluded, including what happens if another offer is already applied.
- A pickup order and a delivery order, if both are meant to qualify.
- A canceled or changed order, including how the discount appears in the reporting available to staff.
Agree on the expected result for each case with your provider before checking it. A cart preview can show confusing wording or a calculation problem, but it cannot tell you whether guests will add items during a real service. It also may not reproduce every live payment or fulfillment step. Resolve any discrepancy before enabling the offer.
Run a controlled test with a stop rule
Pick a comparison that fits your order volume. If your system can assign similar guests to an offer group and a no-offer group at the same time, ask how that assignment and reporting work. Otherwise, compare matched service periods, such as similar weekdays, while noting menu changes, holidays and other promotions. A before-and-after comparison is less reliable when those conditions change.
Set the test period and decision rules before launch. Record the number of direct orders, eligible basket values, discounts, item-level sales and estimated variable costs in each group or period. Count how many orders were already above the minimum in the comparison group. Do not treat every qualifying order in the offer group as an order the promotion created.
Use the worksheet to compare total contribution, not just average basket size. Estimate contribution on eligible sales after variable costs, subtract discounts, and compare the results across groups. If order counts differ, look at both contribution per order and total contribution; a higher average can coexist with a worse total. Record refunds, remakes and extra support separately, then include their estimated costs in the decision.
For a live pilot, choose a maximum total discount spend and a maximum acceptable contribution loss that the restaurant can afford. Stop if either limit is reached, if the cart applies the offer incorrectly, or if staff cannot fulfill orders as promised. Review at the planned end even if none of those limits is reached. With too few comparable orders, the honest result may be that the test is inconclusive.
Compare the offer with keeping prices unchanged, not just with another discount. You can also model a higher minimum or a smaller fixed discount before testing again. Change one main condition at a time so you can understand what affected the result.
Choose the setup you can measure
Your ordering tools need to make the offer understandable to guests and give you enough detail to calculate its cost. Ask whether you can set the eligible items, minimum and dates; prevent unwanted promotion combinations; see discounts by order; and export the sales details needed for your worksheet. Ask how a changed or refunded order appears in reports.
Software and fulfillment costs belong in the broader decision too. Distinguish a quoted processing fee from the all-in cost of operating direct ordering, which may include a subscription, delivery or dispatch, hardware and add-ons. Delivery charges paid by guests can offset restaurant expense, but they remain a cost to the guest. Review current terms on Fleksa’s plans page, checked November 2026, and request a written quote for your setup.
If you are also considering a new ordering channel, the online ordering buying guide covers broader purchase questions. The guide to comparing direct and marketplace contribution can help keep channel costs separate from the discount calculation. Your restaurant website should also make the offer terms easy to find before checkout.
Next step: Pull a recent basket-value report, fill in the worksheet with your own costs, and check the cart scenarios with your current provider. To evaluate another setup, ask Fleksa about its restaurant online ordering options and verified availability for your US location.
Frequently asked questions
Where should I set the minimum basket?
Start near a common basket value that guests could reasonably reach by adding an item. Then check whether the added contribution could cover discounts on both moved orders and orders already above the minimum.
Is a percentage or fixed-dollar discount better?
Neither is automatically better. A percentage discount grows with the eligible basket; a fixed discount has a predictable maximum per qualifying order. Model both against your actual order values and costs.
Can a cart simulation prove the offer will work?
No. A safe simulation checks presentation and calculation without creating a real payment, guest message or courier dispatch. Guest behavior and financial results require a controlled live pilot.
