A restaurant average check drops when customers buy fewer items, choose cheaper ones, use more discounts, or when more of your orders shift to lower-value times, like lunch instead of dinner. Each cause needs a different fix. Before you change prices or promotions, compare like-for-like orders to find out which one you’re facing.
Domino ’s is a timely case. On October 13, 2026, the company reported that U.S. same-store sales grew 0.3% in the third quarter. CEO Joe Jordan told Barron’s that customers, particularly lower-income diners, were buying fewer items per order, which put pressure on franchisee profits. A national pizza chain’s customers aren’t necessarily yours.
Your restaurant might not be losing the customer. It might be losing the drink, the side, or the second pizza. Treat that as a possibility to check, not a conclusion.
Are customers placing fewer orders, or spending less per order?
Weaker sales can come from fewer orders, smaller orders, or both. Separate them first. If completed orders fell, you’re looking at a traffic problem, and our guide on how to increase restaurant traffic covers that. If orders held steady and sales still slipped, the change is inside the order.
Before comparing periods, settle on one definition. This article uses restaurant average order value, calculated this way:
Average order value = food and beverage sales after discounts and refunds, divided by completed orders.
Leave out taxes, tips, and service or delivery charges. They can move for reasons that have nothing to do with what customers ordered. Make sure your sales figure and your order count cover the same orders, and handle refunds the same way in every period.
Watch the label, too. “Average check” isn’t one fixed measure. Some reports calculate average check size per guest, and some treat each payment as its own transaction. Here, it means spending per order. Count each order once, even when a table splits the bill across three cards.
Is your average check falling because your sales mix changed?
Sometimes nothing changed inside the order. The mix of orders changed instead. Here’s an illustrative example with made-up numbers:
| Period | Dinner orders at $30 each | Lunch orders at $15 each | Total orders | Average order value |
| Before | 70 | 30 | 100 | $25.50 |
| After | 50 | 50 | 100 | $22.50 |
The average order value falls 11.8%. Yet a dinner order is still worth $30, and a lunch order is still worth $15. Spending within each meal period didn’t change.
What changed is the mix. This restaurant lost 20 dinner orders and gained 20 lunch orders. More upselling wouldn’t necessarily fix that. The better question is why dinner orders fell. And the shift isn’t harmless: total sales dropped from $2,550 to $2,250.
So compare the same weekdays, meal periods, locations, and ordering channels. Then account for anything else that changed between the two periods, such as opening hours, holidays, or a promotion that ran in only one of them.
What changed inside comparable orders?
Once you’re comparing like with like, look inside the orders. Treat each finding below as a hypothesis, not a proven cause. A drop in beverage revenue, for example, doesn’t prove customers bought fewer drinks. They may have picked cheaper drinks, or a drink deal may have lowered the price.
| What you find | What to check | What to test |
| Fewer drinks or sides | Item quantities, plus the percentage of comparable orders that include at least one paid add-on | One relevant, optional pairing, such as a drink offered with your most-ordered lunch item |
| More lower-priced mains | Which items gained or lost share of orders | One clearly priced upgrade, while keeping your affordable choices on the menu |
| More discounting | The food subtotal before and after discounts, in both periods | A review of one current promotion before you introduce another |
| Missing or unavailable extras | Your actual customer-facing menu on each channel: unavailable items, missing modifiers, wrong prices | Fix the ordering problem first, then decide whether a campaign is still needed |
The last row is easy to miss. If a modifier group dropped off your online menu, or a dessert has been marked unavailable for weeks, customers can’t add what they can’t see. Go through the ordering flow on each channel the way a customer would before you spend money promoting anything.
Did the change improve the business, or just the ticket total?
Test one change at a time, in one meal period or one channel. Then compare the results with a similar trading period, using the same definitions as before. If you add a pairing, adjust a price, and pause a discount in the same week, you won’t know which one moved the number. Low-volume restaurants may need to run a test longer before a real pattern shows, because a few unusual orders can swing the average.
Track three numbers side by side: completed orders, average order value, and total contribution after relevant variable costs. Contribution is the money left after the costs that come with each order, such as food, packaging, payment processing, delivery, and any labor the change genuinely added. It isn’t net profit, because rent and other fixed costs aren’t in it. It does show whether the change paid for itself.
That’s the trade-off to watch. A higher average check isn’t automatically a win if orders or total contribution fall. And smaller orders aren’t automatically bad when they add business that pays its own way.
Start with one comparison and one test
Orders.co’s restaurant reporting brings sales from connected channels into one dashboard. You can filter sales by platform, top-performing items, and peak hours, then export reports to a spreadsheet. You still decide what the numbers mean, but the like-for-like comparison is faster to pull together.
Your next step: pull a recent period and a comparable earlier one, matched by weekday, meal period, and channel. Find the line that moved, whether that’s order count, items per order, item mix, discounts, or menu availability. Then choose one test from the table above and run only that one.
Methodology: This article draws on Domino’s third-quarter 2026 earnings release, and Barron’s reporting on Domino’s; the meal-period example is a hypothetical calculation and does not use Orders.co customer data.
