US restaurants lose meaningful revenue to ghost covers every service. A forensic look at what TCPA, ADA and health codes allow — and where WhatsApp fits.
The National Restaurant Association (NRA) publishes an annual State of the Restaurant Industry report along with regular operator surveys tracking sales, labor, food-cost, and reservation-behavior indicators. Broken-reservation rates — the industry term for reservations that never seat, which restaurateurs call ghost covers or ghost tables — vary substantially by cuisine, price point, day of week, and metropolitan market. Operator surveys and independent restaurant-industry reporting have placed ghost-cover rates at independent full-service US restaurants in ranges frequently cited between five and twenty percent of confirmed bookings, with weekend peak-service ghost-cover rates on the higher end of that band and weekday lunch service on the lower end.
The math for a mid-size US independent full-service restaurant compounds quickly. A fifty-seat dining room turning approximately two-and-a-half times per weekend service, at an average per-cover ticket in the $45–$85 range for a full-service concept, produces a weekend-service revenue exposure per empty reserved table in the low hundreds to over a thousand dollars depending on party size. Fixed overhead — front-of-house wages, prepared-mise-en-place food cost, kitchen line staffing, occupancy — does not pause when a reserved four-top does not seat. Unlike a walk-in-friendly service model where an empty table can be filled at speed, a full-service reservation-driven concept holds tables against the walk-in queue, meaning a ghost cover often loses two revenue moments: the reservation that never seated, and the walk-in that was turned away because the table was held.
Operator responses have consolidated around three mechanisms: confirmation reminders (a text or call twenty-four to forty-eight hours before the reservation), credit-card holds against a defined no-show fee (widespread across upmarket concepts), and short-window overbooking against a modeled ghost-cover rate (a technique borrowed from airline yield-management adapted for restaurant capacity). Each of these has legal implications under US federal telecommunications law and under state consumer-protection statutes that a restaurant sending automated reminder messaging must understand before turning the flow on.
The rest of this piece walks through the US regulatory framework — TCPA for SMS reminders, ADA Title III for reservation-channel accessibility, state two-party consent laws for reservation call recording, and the FDA Food Code for allergen and menu communication to patrons — and then covers where WhatsApp specifically fits (and does not fit) in a compliant US restaurant communication stack, with honest cost comparison against OpenTable, Resy, and Tock.
The Telephone Consumer Protection Act (47 U.S.C. § 227) governs calls and text messages placed to US mobile telephone numbers using automatic telephone dialing systems or artificial/prerecorded voice. For SMS-based reservation reminders, FCC interpretive guidance treats appointment and reservation confirmations as informational rather than marketing content, provided the message body stays focused on the confirmation and does not carry promotional solicitation. The applicable consent standard is prior express consent (documented at reservation intake) rather than the stricter prior express written consent required for marketing text messages.
WhatsApp is over-the-top messaging over a data connection and is generally not governed by TCPA in the way SMS is — WhatsApp messages are not text messages placed to a mobile telephone number in the TCPA statutory sense. Meta's own WhatsApp Business Platform opt-in requirements still apply and are stricter than TCPA in some respects; FTC Section 5 deceptive-practices standards apply regardless of channel; and state laws — the Florida Telephone Solicitation Act (FTSA), the California Invasion of Privacy Act (CIPA), and Washington's Consumer Protection Act — have been read broadly enough to reach automated business messaging targeting state residents. A restaurant sending automated reservation messages to Florida or California mobile numbers should have counsel review the consent flow before launch.
Two-party (all-party) consent for call recording. Reservation phone calls in twelve US states (California, Connecticut, Delaware, Florida, Illinois, Maryland, Massachusetts, Montana, Nevada, New Hampshire, Pennsylvania, and Washington) are commonly classified as all-party consent under varying statutory tests — with Connecticut applying differently to commercial vs private recording and Nevada framed by court interpretation rather than statute. Federal law (18 U.S.C. § 2511) permits one-party consent, but the strictest applicable state law governs when parties are in different states. A restaurant recording reservation calls for training or dispute-resolution purposes must open with a disclosed recording notice in any all-party consent jurisdiction.
ADA Title III accessibility. The Americans with Disabilities Act Title III (42 U.S.C. § 12181 et seq.) requires places of public accommodation, which include restaurants, to provide accessible communication channels for guests with disabilities. Department of Justice guidance and settled cases have applied Title III to reservation systems, requiring that a restaurant not force reservations exclusively through a channel inaccessible to a category of disabled guests (a voice-only phone line with no text or web alternative, or a chat-only flow with no accessible phone line). A WhatsApp reservation channel is a valid addition to the reservation stack but cannot be the sole channel without an accessible parallel option.
The operational consent stack most defensible under this combined federal and state framework captures four elements at reservation intake: written or unambiguous-conduct consent to receive reminder messages via the elected channel (SMS, WhatsApp, or email), the specific message frequency and content types the guest is opting into, an easy opt-out mechanism (STOP reply for SMS, opt-out link for WhatsApp), and an accessible parallel reservation channel that satisfies ADA Title III.
For a US restaurant deciding whether WhatsApp belongs in the reservation stack, the honest positioning is that WhatsApp is one channel among five (phone, SMS, email, WhatsApp, portal) and that channel choice should follow patron population rather than vendor pitch.
Patron WhatsApp coverage in the US is uneven. Meta and Statista have reported approximately 80–90 million WhatsApp users in the United States. Coverage is significantly higher in major-metro markets — Miami, Los Angeles, Houston, New York, Chicago, San Francisco Bay Area — and in Hispanic, Asian-American, and immigrant-and-international-community-serving restaurants. Coverage is materially lower in suburban and rural markets with an over-fifty patron demographic. A South Florida trattoria with a substantial Latino patron base will see meaningful WhatsApp adoption; a Midwestern family restaurant with a median-patron-age north of fifty-five will not.
Marketplace vs communication layer. OpenTable, Resy, and Tock occupy the marketplace-plus-communication layer: they deliver diner discovery through their consumer-facing apps and websites in addition to handling reservation intake and reminders. A restaurant heavily dependent on OpenTable app-driven walk-in cover-count is not making a channel-swap decision by adding WhatsApp — it is deciding whether to reduce reliance on the OpenTable marketplace layer over time. Many restaurants operate a dual stack: a marketplace vendor for discovery-driven bookings and a direct-channel layer (WhatsApp, SMS, phone, website form) for the diners who already know the restaurant.
SMS remains the workhorse for reminders in most US general-market restaurants. Deliverability is universal (any US mobile number receives SMS regardless of app installation), reported open rates sit in the 90%+ range across industries, and A2P 10DLC standardization has made compliant business SMS more accessible. Cost per outbound message via Twilio, MessageBird, or Sinch A2P 10DLC lands at approximately $0.0079 per US message plus 10DLC registration fees.
Email remains the default for longer-form patron communication — menu previews, private-dining inquiries, tasting-menu wine pairings, event announcements. Reported open rates in the restaurant vertical sit in the fifteen-to-thirty percent range across published email-marketing benchmark data. Cost is negligible at reasonable volume.
Where WhatsApp specifically earns a slot is with restaurants whose patron base already uses WhatsApp daily — urban metropolitan concepts, immigrant-and-international-community-serving restaurants, and specialty concepts with international patron flows (destination fine-dining in Miami, Los Angeles, or New York with international-travel bookings). For those restaurants, WhatsApp reservation confirmations read as normal contact rather than institutional messaging and drive materially higher confirmation response rates than SMS.
Honest positioning: WhatsApp is a valuable channel-mix layer for the patron segment that uses it, not a wholesale replacement for the reservation platform. The retention math still lives in whether the restaurant runs a two-touch confirmation pattern (twenty-four-hour and same-day) against whichever channel each specific patron actually reads.
For a mid-size US independent full-service restaurant averaging six hundred covers per month across a five-day service, the entrenched reservation-platform pricing and the a la carte messaging costs compare as follows.
OpenTable — the category leader for US restaurant reservation infrastructure — publishes tier pricing across its Basic, Core, and Pro offerings. Published tier pricing sits in the approximate range of $149 per month (Basic) to $299 per month (Core) to $499 per month (Pro) plus per-seated-cover fees typically in the $0.25 to $1.50 range depending on the source of the reservation (network diner from OpenTable app vs website widget vs phone). A restaurant seating six hundred network-sourced covers per month on the Pro plan accumulates a per-cover fee stack that materially exceeds the base subscription.
Resy operates on a similar subscription-plus-per-cover model with tiered pricing generally reported in the $249 to $899 per month range depending on features and support level, with the American Express-owned Resy positioning benefits (Amex card-holder promotion, Resy consumer app placement) as part of the premium tier value.
Tock operates on a differentiated model, with prepayment infrastructure (ticketed reservations for tasting menus and prix-fixe concepts) that eliminates the ghost-cover problem for concepts that adopt prepayment. Pricing is published in tiers that generally begin at $199 per month for basic operation and scale with prepayment transaction volume.
A la carte messaging costs for a restaurant running its own reminder stack via a general messaging platform:
Total stack cost for a mid-size US restaurant typically lands in one of three configurations: entrenched marketplace-integrated platform ($350–$1,200/month all-in depending on cover volume and marketplace mix), middle-tier platform plus messaging ($200–$500/month all-in), or direct-channel-only via general messaging platform ($50–$150/month plus staff time for template maintenance and calendar sync).
The economic decision is rarely dominated by per-cover platform cost. It is dominated by (1) how much of a restaurant's monthly cover count is discovered through the marketplace vendor's consumer app (a hard-to-replicate distribution channel), (2) whether the restaurant has bandwidth to operate a direct-channel stack with the required opt-in and template maintenance, and (3) whether the concept can adopt prepayment (which eliminates the ghost-cover problem structurally, at the cost of some patron-conversion friction). A cheap platform that leaves marketplace-sourced discovery on the table is not cheap; an expensive platform whose marketplace reach a restaurant does not need is over-scoped.
Data + numbers referenced in this article are sourced from these public documents:
Product page with honest feature list, "not for you if" filter, and live demo for this vertical.
See /for/restaurant →Not ready to sign up yet? Try the free demo →