The Word 'Estimated' Is a Confession: What Restaurant Marketing Attribution Can Prove
Estimated ROI is not automatically bad. The word tells you where direct observation ended and where a restaurant marketing calculation began.
“Estimated” on a restaurant marketing report is not an admission that the number is fake. It is an honest label for where direct observation ended and a calculation began, as long as the report shows what was observed and what the estimate still cannot prove.
The real problem is not the word. It is a report that quietly turns an estimated conversion into a restaurant visit, an attributed sale into incremental revenue, or gross revenue into profit.
A useful restaurant marketing ROI report keeps those claims separate instead of forcing one impressive number to answer every question.
“Yelp, Google, Meta, everyone wants to sell me ads and will make up ROI, but it's all guesswork.”An independent brewpub owner
Restaurant owners are right to distrust a return that arrives without an explanation. They should not reject every estimate. They should ask the report to stop exactly where the evidence stops.
A direct match proves a connection, not causation
The strongest person-level restaurant record connects a campaign response to an identified guest, a completed visit, and a paid, non-voided POS check. That gives you directly matched gross revenue.
The match proves that the recorded path ended in a purchase. It does not prove the guest would have stayed home without the ad. A regular could claim an offer they already planned to use. A loyal guest could click an ad before making their usual visit. Direct observation is stronger than a platform guess, but it still answers a narrower question than incrementality.
This is why tracking Facebook and Instagram ads to restaurant sales should preserve both the direct match and the limit of the claim. The check is real. The ad's causal effect still requires a comparison.
An estimate can answer a question that direct matching cannot
Restaurant ads often affect people who never identify themselves. Someone can see a food video, remember the restaurant, and visit days later without clicking, claiming an offer, or using the same payment identity. A direct-match report will miss that guest.
An incremental estimate asks a harder question: how much of the restaurant's sales or visit movement probably would not have happened without the campaign? A relevant baseline, a similar untreated location, or a carefully designed test can help answer it. The result must account for other changes such as holidays, weather, closures, prices, promotions, local events, and service problems.
The estimate can be more useful for a spending decision than the direct-match total because it attempts to measure the campaign's added effect. It is also more dependent on assumptions. Both facts can be true.
That is why “estimated” is not a weakness to hide. It tells the owner to inspect the comparison and uncertainty before treating the number as fact.
One number should not do four jobs
Restaurant marketing reports should keep these answers separate:
| Label | What the number can honestly say |
|---|---|
| Platform-attributed gross revenue | The platform credited these sales under its own rules |
| Directly matched gross revenue | A known campaign response connected to these settled checks |
| Observed restaurant movement | Sales, covers, transactions, or item sales changed during a relevant comparison |
| Estimated incremental sales | The calculation attributes part of that movement to the campaign under stated assumptions |
Contribution comes after the restaurant subtracts the costs required to produce and serve the additional sales. Profit depends on the full accounting definition. Neither should be hidden inside a revenue label.
A report can show every row. The problem begins when the largest platform number is called sales, ROI, and profit at the same time. For Google Ads, see what Store Visits estimates can and cannot prove.
A useful estimate shows enough work to challenge it
You should be able to see which sales or visits went into the calculation, which period or location supplied the comparison, what else changed, and how the result responds when the assumptions become more conservative.
If a restaurant made more money during the campaign, that is observed movement. If the estimate removes a reasonable expectation of what would have happened anyway, that is estimated incremental sales. If the restaurant then subtracts food, labor, discounts, fees, creative, management, software, and ad spend, the remainder is estimated campaign contribution.
Smaller claims make the report more useful because they show exactly what the evidence supports. “The campaign generated revenue” hides too many questions. “We directly matched these checks, observed this wider sales change, and conservatively estimate this portion was incremental” gives the owner something they can examine and use.
When a report cannot show the inputs or explain why the comparison is fair, “estimated” is doing all the work. When it can, the label protects the owner from false certainty.
Honest limits are part of the product
Feast can connect trackable campaign responses to supported POS purchases, show the settled checks that matched, and place that direct evidence beside wider movement in restaurant sales and visits. The direct total and the modeled wider effect remain separately labeled, so the owner can see which result was observed and which depended on a comparison.
Owners do not need perfect certainty before making a decision. They need a conservative answer whose limits are visible. If the lower end of a reasonable estimate still repays the campaign costs, the decision is much safer. If the campaign works only under the most generous assumptions, the uncertainty is the conclusion.
The same discipline applies to the return itself. A good restaurant ROAS must clear the break-even point created by the restaurant's contribution margin, not merely look large in a platform report.
The word “estimated” should make a report more trustworthy, not less. It tells you where to ask the next question.
Feast Revenue Analyzer helps you examine your marketing alongside restaurant sales, with seasonal context and the limits of the evidence explained. Understand what the numbers can support before choosing your next investment. Analyze my restaurant's revenue.