Costs
Part of Restaurant marketing: signal versus noise
Restaurant marketing case studies: what to verify before copying
How to read restaurant marketing case studies: the disclosures that matter, how to check endorsements, and how to retest one claim against your own sales data.
What to take away
- A restaurant marketing case study is evidence to inspect, not a result to copy.
- The useful ones name the restaurant, the period, the baseline and the cost.
- A named restaurant with a published method beats an anonymous client.
- Separate the marketing change from the operations change that ran alongside it.
- Reproduce one claim with your own data before you buy anything.
What a case study has to disclose
A defensible case study answers six questions:
- What changed
- Compared with what
- For whom
- Over which period
- At what cost
- With what operational side effects
GAO's guide to designing evaluations sets out how a researchable question, a comparison, measurement and stated limitations shape credible evidence. A restaurant can borrow those questions without running a government study.
| Claim | Evidence needed | Usual omission |
|---|---|---|
| Sales rose | Comparable completed sales, same period | Price change, new site |
| More guests | Deduplicated guest count | Regulars switching channel |
| Better return | Full cost and incremental margin | Fees, discounting, labor |
| Stronger reviews | Request method, full distribution | Selective solicitation |
| Campaign caused it | A credible comparison | Weather, events, roadworks |
| Delivery grew | Net-of-commission sales per order | Platform fees of about 15% to 30% |
| A new audience arrived | New-to-brand guest count, not redemptions | Existing regulars using the offer |
Six restaurant case studies, with method, baseline and cost
Every figure below comes from a filing, a press release or a published menu price. Check the current figure as of 2027 before you quote it.
Dominique Ansel Bakery, New York. The Cronut launch remains the most documented restaurant demand spike of the past decade. The pastry went on sale May 10, 2013 at $5, limited to two per customer. Queues formed before the 8 a.m. opening and the daily batch sold out within about an hour in the first weeks.
The method was one limited item and press coverage the bakery did not buy. The baseline was a single-room pastry shop. Cost: recipe development and lost production on other items; no media spend is public. The limit: the bakery documents demand, not revenue, and the result leaned on novelty plus a Manhattan address. Copy the scarcity, not the queue.
Sweetgreen. The chain publishes its own impact reporting and has tested loyalty and delivery changes in named markets. Its quarterly filings give comparable restaurant sales, average unit volume and restaurant-level margin, the baseline most case studies never disclose. Read the filing rather than the press release, which drops the comparison.
That subscription, Sweetpass, sold for about $10 a month when Sweetgreen tested it in 2022.
Taco Bell. Doritos Locos Tacos, a co-brand with Frito-Lay, arrived in March 2012. Nacho Fries followed in January 2018 as a limited-time item and have returned many times. Yum Brands filings and trade press document both launches. The method in each case is a limited-time offer built with a supplier partner.
Cost: the co-marketing agreement is not public, so treat the outcome as real and the cost as unknown. Neither company breaks out what a single item added to the system total.
Dig. The fast-casual chain runs roughly 30 restaurants, mostly in New York City. It has published results from its own kitchen and loyalty work, including the effect of menu simplification on ticket time.
That is an operations result wearing a marketing label. Ask which cost line moved before you credit a campaign.
Cava. Cava listed on the NYSE in June 2023, and its filings disclose comparable restaurant sales separately from the contribution of new openings. Restaurant-level margin appears in the same document. Most agency case studies blend those two lines and call the total marketing lift.
Wingstop. Wingstop, ticker WING on Nasdaq, runs more than 2,000 restaurants worldwide and reports digital sales at roughly two-thirds of system-wide sales in recent quarters. Same-store figures sit in its quarterly shareholder letters. Most locations carry little or no dine-in seating.
Cost: app and CRM spending sits inside general and administrative expense, not in a marketing line. The letters are free to read, which makes this the cheapest item on the list to verify.
Check the endorsement before the result
The FTC's Endorsement Guides overview covers endorsements, influencers and reviews. When a case study quotes a happy operator, check whether the relationship and any incentive are disclosed.
The FTC advertising substantiation policy requires a reasonable basis before an objective claim goes out. That applies to a vendor's published performance numbers as much as to an ad.
The FTC guidance on soliciting and paying for reviews warns about fake feedback, selective requests, conditioned incentives and paid rankings. Apply those checks whenever a case study credits a review campaign.
For the wider set of worked examples, see these hospitality marketing strategy case studies. For the operational side, our restaurant marketing coverage tracks how the same campaigns land in the booking and POS data.
Reproduce one number locally
Pick the single claim that would change your decision. Rebuild it with your own covers, average check and labor rate. If the vendor will not supply the method, file the item as a lead for testing, not a verified result.
Your own last four quarters are the one case study whose baseline you can verify. Pull the same four weeks from each of the last four years before you copy anything above.
Write down which result would make you stop, and by when. Keep the case studies you rejected alongside the one you followed; the reason you set one aside can expire.
Common questions
Are vendor case studies reliable?
Sometimes. Verify the definitions, the comparison group, the full cost and the operational changes before you treat one as evidence. Sweetgreen's filings show a comparable-sales figure; a slide deck usually will not.
Can a restaurant copy a successful campaign?
Copy the question and the mechanism, then retest it against your own demand, capacity and economics. A $5 pastry and a two-per-customer limit travel; a Manhattan foot-traffic advantage does not.
What makes a case study persuasive?
A named subject, a comparable baseline, reconciled outcomes, full costs, and an honest list of limits.
