Restaurant email marketing: the only audience you actually own

1. Why owned email beats any paid channel on unit economics

Because the acquisition cost for that guest was already paid, and writing to them again doesn't charge you a second time. QSRs generate roughly 71% of their sales from repeat customers, according to Restroworks (Restaurant Customer Retention Statistics 2024), and 39% of US restaurant visits in 2025 came from loyalty members, double the 2019 figure, according to LoyaltyPass (Restaurant Loyalty Statistics 2026). Read those two numbers together, because they tell a single cash story: most of your revenue already sits inside your database, waiting for a reason to come back, while you keep bidding at auction for people who may never walk through the door. The guest who arrives through an owned channel is also worth 45% more in lifetime value than the one who arrives through the website alone, according to Lightspeed (Online Ordering Statistics 2025). That arbitrage is yours. Some 29% of US restaurant traffic over the last twelve months arrived with a deal attached, according to Circana 2025 (via Restaurant Business), which means nearly a third of your volume has been trained to expect a markdown.

2. Blanket discounting is the fastest way to lift covers and sink EBITDA

When you send the same coupon to your whole list, you hand margin to people who were coming anyway and teach everyone else never to pay menu price. An email that pushes the 32% food cost dish instead of the 24% one can close the week with more covers and less profit, and your manager will report it as a win because they measured opens rather than contribution margin. The rule we apply at Masterestaurant is simple and unpopular: no discount campaign ships until someone signs off on how much margin they are willing to give away, and to which exact segment. In this band the call is to spend nothing on software and everything on capture discipline. The threshold I ask for is concrete: 800 consented addresses within twelve months, captured at reservation and at the POS, a bounce rate below 3%, and no more than two sends a month, one recommending the menu and one tied to a real neighborhood date.

3. Under 500 thousand dollars a year: a clean list and two sends, no expensive platform

That is enough to move repeat visits without a marketing payroll. The supporting lever here is the local listing, since a Google Business Profile gets seven times more views than the restaurant's own website, according to Malou (Local SEO for Restaurants 2025), and 57% of Yelp users contact or visit a business within 24 hours, according to Yelp 2026. Your email closes what the listing opened. This band does NOT get dropped from the plan: it gets simplified. An operator in this band sent me a P&L with the marketing line up 31% over twelve months and an average check that hadn't moved; all that money bought traffic he already had. The fix costs nothing extra: stop sending by commercial calendar and start sending by cycle. If your average guest returns every 23 days, the day-nine coupon gives margin away and the day-twenty message recovers a visit that was about to be lost.

4. Between 500 thousand and 1 million: break the calendar and let repeat-visit data decide

Set the threshold at three live segments, a cycle measured with POS data, and a ceiling of 15% of the list receiving a discount in any month. Everyone else gets menu and calendar content. With 12.2 million restaurant-to-consumer delivery users in Spain alone during 2025, according to Statista Market Forecast, your own list is the only thing keeping you from renting your clientele forever. Opens don't cover payroll, and they make a convenient alibi. From this size up, the only report you accept carries four numbers per campaign: attributed revenue on a seven-day window, incremental covers against a holdout group that received nothing, the mix of dishes sold, and that week's food cost variance. Without a control group there is no attribution, only narrative. The operating threshold I use with clients this size is 8% of monthly revenue originated in owned channels and a holdout of at least 10% of the list.

5. Above 1 million: demand incremental sales from your team, not open rates

Investing here makes sense because 78% of adults have already downloaded at least one food app, according to the National Restaurant Association, so your guest is used to hearing from restaurants; the question is whether yours contributes anything or simply competes for noise with another dozen. Once the business crosses this line, and I mean the media-chef profile or the large-format themed venue with a celebrity behind it, a product appears that the small band cannot monetize: the gift card. The restaurant gift card market was worth US$36,817 million in 2025, according to Business Research Insights, and 43% of gift card sales belong to cafés and restaurants, according to Capital One Shopping (Gift Card Statistics 2026). That money lands before the plate cost exists, it improves your working capital, and a portion is never redeemed. Here email stops being a reminder and becomes a shelf: seasons, experiences, chef's table, corporate catering.

6. Above 5 million: email turns into a revenue channel, not a communication one

Diego F. Parra puts it this way in the boardrooms Masterestaurant advises: if the owned channel has no revenue line of its own, it is still an expense in disguise. In a group, the problem stops being the message and becomes duplication. The same guest exists five times under five different addresses, each brand writes on its own, and you pay to reach a person who was already on the way. Before approving another peso of budget, require unified identity, a cap on contacts per person per week, and an attribution split agreed among the brands. The threshold I set is 90% of records deduplicated and a ceiling of three weekly emails per person across every banner. With loyalty program adoption projected at 80% of operators by the close of 2025, according to LoyaltyPass, owning a database no longer differentiates anyone; what separates one group from another is the INTELLIGENCE behind its segmentation and the nerve to stay quiet.

7. What happens if you don't build the owned channel within twelve months?

Here is what happens, step by step.

Your rent rises, platform commissions take between a fifth and a third of the delivery ticket, and you respond by buying more reach on social, where 51% of TikTok users dine out because of a restaurant's content, according to Restroworks 2025, and 63.1% discover products and trends there, according to The Influence Agency 2025. Reach works, yet it is rented and its price climbs with the auction. By month thirteen you will have more gross sales, less margin, and no list of your own; by month twenty-four you will depend on an algorithm that has already changed twice. The paradox of this trade is that the oldest channel, email, is the only one that survives platform shifts, because the address belongs to you. Start this week: pull the real count of consented addresses from your POS. <strong>The difference is not the tool, it is who decides the send.</strong> Before, the commercial calendar and the anxiety of a slow Tuesday decided; after, repeat-purchase data decides.

8. What actually changes between the before and the after?

A guest who returns every 23 days does not need a coupon on day 9: he needs a reason on day 20. That single timing correction, with no extra spend, is what moves guest LTV.

<strong>Before you measured opens; after you measure incremental sales and margin.</strong> Open rate does not pay payroll. What an owner above the 1 million dollar band must demand is how much the campaign sold, with what dish mix, and what food cost variance it left that week; if the email pushed the 32% food cost plate instead of the 24% one, covers rose and EBITDA fell. <strong>Before, discount was the strategy; after, it is an expensive exception.</strong> With 29% of traffic already conditioned on offers (Circana, 2025), giving margin away by email teaches your best guest never to pay full price. Risk mitigation here means reserving discounts to wake dormant contacts and selling frequents on content, access and experience.

9. What actually changes between the before and the after — in practice

<strong>Before, the owned channel competed with the platforms; after, it uses them as a front door.</strong> A guest arriving through an aggregator costs commission every single time; one who migrates to your own channel is worth 45% more in lifetime value (Lightspeed, 2025). Email is the migration mechanism, and in Spain, with 12.2 million restaurant-to-consumer delivery users in 2025 (Statista, 2025), that arbitrage is worth a full P&L line. <strong>Before, online reputation was luck; after, it is an engineered flow.</strong> An email sent 24 hours after a good visit turns experience into a review, and that profile pulls 7 times more views than your own site (Malou, 2025). I got this wrong for years: I thought asking for a review was begging, when it simply closes the loop on service you already delivered well.