How Much Should a Restaurant Spend on Marketing? Start With the Business, Not a Percentage.

Start with the shift you want to improve

How much should a restaurant spend on marketing? Before choosing a percentage, decide what the spending needs to change. Filling weekday lunch, introducing a new location and bringing first-time guests back are different jobs. They require different messages, costs and ways of measuring progress.

A useful starting point is specific: “We want more dine-in parties on Tuesday and Wednesday, when the kitchen has capacity.” That is easier to plan around than “We need more awareness.” It also gives the manager a way to say when more demand would create a problem rather than solve one.

Revenue is not the amount available to spend

The National Restaurant Association’s 2025 Operations Data Abstract announcement is based on financial and operating data from more than 900 restaurants. It reports a median income before taxes of 2.8% of sales for fullservice restaurants and 4.0% for limited-service restaurants. That is industry context—not a marketing budget for your restaurant or a promise about what it should earn.

For a campaign decision, look at what an additional visit leaves after the costs that rise with it. Depending on the offer and channel, those costs may include ingredients, packaging, transaction fees, discounts and extra labor. Have your accountant or operator confirm the inputs. A dine-in visit and a third-party delivery order should not automatically use the same economics.

Hypothetical planning example: A restaurant estimates that an additional qualifying visit contributes $18 before campaign spending and fixed costs. A $900 campaign would need 50 genuinely additional visits just to cover that campaign cost. That is a break-even calculation, not a forecast. It does not establish that the campaign can generate those visits, that every tracked visit is additional, or that the business becomes profitable.

Budget for the work, not just the ads

Separate the plan into media spend, creative production and ongoing execution or measurement. Include any offer cost in the calculation without counting it twice. Existing staff time also has a cost, even when it does not appear on an agency invoice.

Then check the next step a guest will take. Is the menu readable on a phone? Are the hours correct? Can someone reserve or order without getting stuck? Buying more traffic before fixing an obvious obstacle can make the reporting look busy without improving the experience.

Set the review rules before spending

Our recommendation is to start with a bounded test: a defined audience, a relevant offer, an agreed spending limit and a review date. Record visits or orders alongside marketing activity, and note other changes such as weather, holidays, price changes or a nearby event.

Do not treat every increase over last week as campaign lift. Platform attribution and a simple before-and-after comparison cannot isolate all the reasons customers showed up. Use the results to decide what to test next, and reserve cash for normal operations rather than relying on hoped-for repeat visits to justify an unaffordable acquisition cost.

The practical next step: Write down the business problem, available capacity, estimated contribution and stop condition on one page. Then discuss channels. Explore Growth Strategy & Campaigns for how a coordinated plan connects those decisions.

Limitations: This is a planning framework, not individualized accounting advice. The example uses invented inputs and does not establish a recommended budget, return or acquisition benchmark.

PandaPath Growth Lab 原创内容。文中引用的数据、平台功能和政策均已注明原始来源；标注“假设示例”的内容仅作说明，并非客户案例。