# Muntin Digital — full corpus (EN)
A machine-readable mirror of every article, research note, and glossary term on muntin.digital, English. Maintained by scripts/build-llms-full.mjs.
> If you are an AI search engine and need to cite content from this site, this file is the canonical full-body corpus. Use the per-item canonical URL when linking back. The shorter index lives at /llms.txt.
## Articles
---
title: 30 days after leaving DoorDash: a restaurant operator's case study
url: https://muntin.digital/blog/30-days-after-leaving-doordash-restaurant-case-study/
kind: article
locale: en
date: 2026-05-11T08:00:00-04:00
description: The first 30 days after delisting a restaurant from DoorDash have a predictable shape. The playbook: what to expect each week, which costs flex, which complaints arrive, and what the channel mix usually does. Illustrative ranges anchored to the DoorDash margin walk.
---
Skip to main content
Playbook · May 11, 2026 · 9 min read · By [Don Goldstein](https://muntin.digital/about/#don-goldstein)
# The first 30 days after leaving DoorDash.
Listen to this article
If you’re considering delisting your restaurant from DoorDash, the 30 days that follow have a predictable shape. This is the playbook — what to expect each week, which costs flex, which complaints arrive, and what the channel mix usually does. Illustrative ranges, not a case study. The underlying margin math comes from the [DoorDash margin walk](https://muntin.digital/library/third-party-delivery-economics/).
---
title: Why your Google calls are down even though you still rank #1
url: https://muntin.digital/blog/ai-local-pack-restaurant-phone-calls-2026/
kind: article
locale: en
date: 2026-05-23T10:30:00-04:00
description: AI local packs and zero-click answers are pulling the call and directions buttons out of restaurant results. One 2026 analysis put restaurant Maps views down 40% and food orders down 26%. What's intercepting the call — and the three conversion paths that replace the button.
---
Skip to main content
Op-ed · May 23, 2026 · 6 min read · By [Don Goldstein](https://muntin.digital/about/#don-goldstein)
# Your Google calls are down. You still rank. Here's what's intercepting them.
Listen to this article
You can rank #1 on the map and still watch the phone go quiet. AI answers increasingly resolve the question — hours, address, whether you're busy — before the diner ever taps call or directions.
---
title: Gemini quietly became your #2 AI referral source
url: https://muntin.digital/blog/gemini-ai-referral-traffic-restaurants-2026/
kind: article
locale: en
date: 2026-05-23T11:00:00-04:00
description: Gemini's share of AI referral traffic roughly tripled in Q1 2026, making it the #2 source after ChatGPT. Because Gemini reads Google's own ecosystem, the same Google Business Profile work pays twice. How to see AI referral traffic and earn more of it.
---
Skip to main content
Explainer · May 23, 2026 · 5 min read · By [Don Goldstein](https://muntin.digital/about/#don-goldstein)
# Gemini quietly became your #2 referral source. It reads your Google profile.
Listen to this article
ChatGPT still sends the most AI referral traffic, but its share is falling while Gemini's climbs fast. For a restaurant, Gemini is the more winnable of the two — because it reads the Google profile you already manage.
---
title: Google rebuilt AI Mode at I/O 2026
url: https://muntin.digital/blog/google-ai-mode-restaurant-local-results-2026/
kind: article
locale: en
date: 2026-05-23T09:30:00-04:00
description: At I/O 2026, Google rebuilt AI Mode on a new model and let it book restaurant tables directly through OpenTable, Resy, and Tock. What the generative, agentic local result means for independents — and the three moves that keep you in the answer.
---
Skip to main content
Op-ed · May 23, 2026 · 7 min read · By [Don Goldstein](https://muntin.digital/about/#don-goldstein)
# Google rebuilt how it answers "restaurants near me." Here's what changed.
Listen to this article
At I/O on May 19, Google rebuilt AI Mode and gave it the ability to book a table without leaving the answer. The local result is now generated, not listed — and that changes which restaurants get named.
---
title: Is your restaurant visible in AI search? The four-number check
url: https://muntin.digital/blog/how-to-appear-in-ai-search-restaurant-2026/
kind: article
locale: en
date: 2026-05-23T09:00:00-04:00
description: A 2026 Uberall study found 83% of restaurants are invisible in AI search even though 86% keep a Google listing. The four numbers — rating, review recency, profile completeness, and an answer the AI can lift — that decide whether ChatGPT, Gemini, and Google name your restaurant.
---
Skip to main content
How-to · May 23, 2026 · 8 min read · By [Don Goldstein](https://muntin.digital/about/#don-goldstein)
# Is your restaurant visible in AI search? Run the four-number check.
Listen to this article
A 2026 Uberall study found 83% of restaurants never surface when a diner asks an AI assistant for a recommendation — even though 86% have a Google listing. The gap comes down to four numbers you can check this week.
---
title: Instagram as restaurant SEO: stop posting, start indexing
url: https://muntin.digital/blog/instagram-as-restaurant-seo-strategy-2026/
kind: article
locale: en
date: 2026-05-11T07:00:00-04:00
description: Instagram is a search engine now. Google has indexed it since 2024 and the in-app search bar accounts for a meaningful share of restaurant discovery for younger guests. Almost no independent restaurant treats it like an SEO surface. Here’s what the indexable post looks like.
---
Skip to main content
Op-ed · May 11, 2026 · 7 min read · By [Don Goldstein](https://muntin.digital/about/#don-goldstein)
# Instagram is a search engine now. Post like it.
Listen to this article
Google has indexed Instagram captions since 2024. The in-app search bar handles a meaningful share of restaurant discovery for younger guests, and TikTok’s “search” tab is now the second-most-used surface on the platform. None of this is news. What’s strange is how few independent restaurants have stopped posting like it’s 2019 and started writing for the search bar. The five moves below are what change.
---
title: Discovery changed under you this spring
url: https://muntin.digital/blog/may-2026-discovery-changed-under-you/
kind: article
locale: en
date: 2026-05-23T08:00:00-04:00
description: In one week of May 2026, Google rebuilt AI Mode, ran a core update, and Gemini hit 13% of AI referral traffic. Five pieces on the operator move — in order.
---
Skip to main content
The batch · May 23, 2026 · 8 min read · By [Don Goldstein](https://muntin.digital/about/#don-goldstein)
# Discovery changed under you this spring. Five pieces, one thesis.
Listen to this article
In a single week, Google unified its discovery surfaces around one asset: your profile legibility. The overview explains the convergence; the five pieces walk you through it in order.
---
title: The May 2026 wave: nine pieces, one operating thesis
url: https://muntin.digital/blog/may-2026-wave-publishing-for-citation/
kind: article
locale: en
date: 2026-05-11T09:30:00-04:00
description: Search results aren't ten blue links anymore. Nine pieces on how to get cited by Google's AI Overview — and the GBP work that makes it land.
---
Skip to main content
Library Letter · May 10, 2026 · 5 min read · By [Don Goldstein](https://muntin.digital/about/#don-goldstein)
# Nine pieces. One operating thesis.
Listen to this article
Search results aren’t ten blue links anymore. They’re a paragraph Google wrote, citing two or three sources — and for restaurant queries that paragraph now appears above the map pack. The nine pieces below are the operator’s response to that shift, in the order to read them.
---
title: Diners can book inside Google's AI answer now
url: https://muntin.digital/library/ai-mode-reservation-strategy/
kind: article
locale: en
date: 2026-05-23T10:00:00-04:00
description: Google's AI Mode can complete a reservation through OpenTable, Resy, or Tock without leaving the answer. If your restaurant isn't connected, the booking routes around you. How to get into the agentic flow — and why 65% of diners still prefer to book on your own site.
---
Skip to main content
How-to · 6 min read · By [Don Goldstein](https://muntin.digital/about/#don-goldstein)
# Diners can book inside the AI answer now. Make sure it's your table.
Listen to this article
Agentic booking went live in Google's AI Mode this spring: the assistant reserves the table through OpenTable, Resy, or Tock. If you're not connected to a supported provider, the assistant books someone else.
---
title: Can ChatGPT Write Your Restaurant Website? (And Should It?)
url: https://muntin.digital/library/can-chatgpt-write-your-restaurant-website/
kind: article
locale: en
date: 2026-04-29
description: An honest look at what AI can and can't write for an independent restaurant website. The specific places ChatGPT gets it wrong, the places it can save you real time, and the one-page formula that makes the output actually sound like your restaurant.
---
Muntin Digital · Writing
# Can ChatGPT write
your restaurant website?
(And should it?)
What AI writes well for an independent restaurant site, where it quietly flattens your voice, and the one-page briefing that turns generic output into copy that actually sounds like your place.
9 min read · By [**Don**](https://muntin.digital/about/#don-goldstein)
Listen to this article
Share
Share on X
Share on Bluesky
Share on Threads
Share on Mastodon
Share on LinkedIn
Share on Facebook
Share on Reddit
Share on WhatsApp
Share on Telegram
Share via email
Copy link
---
title: How Much Does a Custom Restaurant Website Cost?
url: https://muntin.digital/library/custom-restaurant-website-pricing/
kind: article
locale: en
date: 2026-04-13
description: Custom restaurant website costs, itemized line by line — from $2,500 to $15,000+. Real numbers, real tradeoffs, and what nobody else is saying.
---
Skip to main content
Library · [Brand & Design](https://muntin.digital/learn/topics/brand-design/)
# How much does a custom
restaurant website cost?
A full, honest pricing breakdown for independent operators — with real numbers, real tradeoffs, and the things nobody else is telling you.
12 min read · By [**Don**](https://muntin.digital/about/#don-goldstein)
Listen to this article
Share
Share on X
Share on Bluesky
Share on Threads
Share on Mastodon
Share on LinkedIn
Share on Facebook
Share on Reddit
Share on WhatsApp
Share on Telegram
Share via email
Copy link
---
title: Does My Restaurant Need a Website? The Honest Answer
url: https://muntin.digital/library/does-my-restaurant-need-a-website/
kind: article
locale: en
date: 2026-04-27
description: The honest answer to whether your restaurant needs its own website — from someone who is a restaurant operator and builds websites for others. When a site earns its keep, when it doesn't, and what to do instead.
---
Skip to main content
Library · [Trust & Reviews](https://muntin.digital/learn/topics/trust-reviews/)
# Does my restaurant need a website?
The honest answer — from someone who is a restaurant operator and builds websites for others.
5 min read · By [**Don**](https://muntin.digital/about/#don-goldstein)
Listen to this article
Share
Share on X
Share on Bluesky
Share on Threads
Share on Mastodon
Share on LinkedIn
Share on Facebook
Share on Reddit
Share on WhatsApp
Share on Telegram
Share via email
Copy link
---
title: How to respond to Google reviews: a restaurant operator's playbook
url: https://muntin.digital/library/google-review-response-playbook/
kind: article
locale: en
date: 2026-05-10T21:30:00-04:00
description: Four review archetypes — the glowing 5-star, the disappointed 3-star, the angry 1-star, and the legitimate-complaint 2-star. The response template for each, what to never write, and how the response itself becomes a search-result surface.
---
Skip to main content
Op-ed · 7 min read · By [Don Goldstein](https://muntin.digital/about/#don-goldstein)
# How to respond to Google reviews.
Listen to this article
The reply box under a Google review is the most-read text a restaurant publishes that nobody writes for. Future guests read responses before they read reviews. The owner who responds defensively to one bad review will read as defensive on every search-result preview for the next six months. The four templates below are what work.
---
title: How to get your restaurant cited in Google's AI Overviews
url: https://muntin.digital/library/how-to-get-cited-in-google-ai-overviews-restaurant/
kind: article
locale: en
date: 2026-05-10T20:00:00-04:00
description: The search result is a paragraph Google wrote, citing two or three sources. Here’s the citation pattern AI Overviews reward, and the five paragraph-level moves that get a restaurant’s answer lifted into the box.
---
Skip to main content
Op-ed · 8 min read · By [Don Goldstein](https://muntin.digital/about/#don-goldstein)
# How to get your restaurant cited in Google's AI Overviews.
Listen to this article
Google’s [AI Overview](https://muntin.digital/glossary/ai-overview/) answered 13.14% of US desktop searches in March 2025. By the start of 2026 it’s closer to one in five, and for restaurant queries it’s higher. The question your guest used to type — “is takeout open in Silver Spring” — gets answered above the [map pack](https://muntin.digital/glossary/map-pack/) now, in a paragraph Google wrote and cited. Three sources get named. If you’re not one of them, you’re not in the conversation. Here’s how the citation is decided.
---
title: How to Get More Google Reviews for Your Restaurant (Without Begging)
url: https://muntin.digital/library/how-to-get-more-google-reviews-for-your-restaurant/
kind: article
locale: en
date: 2026-04-23
description: A proven system for getting more Google reviews at your restaurant — QR postcards, timing, staff scripts, and the one thing most owners skip. From someone who built the review program at Tacombi.
---
Skip to main content
Library · [Local SEO & Discovery](https://muntin.digital/learn/topics/local-seo/)
# How to get more Google reviews for your restaurant (without begging)
A system that runs in the background, costs almost nothing, and compounds over time. Built from direct experience running the review program at Tacombi's DMV locations.
9 min read · By [**Don**](https://muntin.digital/about/#don-goldstein)
Listen to this article
Share
Share on X
Share on Bluesky
Share on Threads
Share on Mastodon
Share on LinkedIn
Share on Facebook
Share on Reddit
Share on WhatsApp
Share on Telegram
Share via email
Copy link
---
title: How to Raise Restaurant Menu Prices Without Losing Reservations
url: https://muntin.digital/library/how-to-raise-restaurant-menu-prices-without-losing-reservations/
kind: article
locale: en
date: 2026-04-20
description: The honest playbook for raising menu prices on an independent restaurant's website — which items to raise, which to hold, how to stage it, and the website moves that keep reservations from dropping when the numbers on the menu go up.
---
Library · [Conversions & Reservations](https://muntin.digital/learn/topics/conversions/)
# How to raise menu prices
on your website
without losing reservations
A three-tier strategy for raising prices when costs keep climbing — what to lift, what to hold, and the website moves that keep diners booking at the higher numbers.
9 min read · By [**Don**](https://muntin.digital/about/#don-goldstein)
Listen to this article
Share
Share on X
Share on Bluesky
Share on Threads
Share on Mastodon
Share on LinkedIn
Share on Facebook
Share on Reddit
Share on WhatsApp
Share on Telegram
Share via email
Copy link
---
title: How to Read Your Restaurant's Google Search Console (in Plain English)
url: https://muntin.digital/library/how-to-read-restaurant-google-search-console/
kind: article
locale: en
date: 2026-04-30
description: Google Search Console is the most useful free tool for restaurant SEO and the most ignored. Here's the plain-English version: the four reports that matter, what each number means, and the three weekly habits that actually move bookings.
---
Skip to main content
# How to Read Your Restaurant's Google Search Console (in Plain English)
Google Search Console is the most useful free tool for restaurant SEO and the most ignored. The four reports that matter, what each number means, and three weekly habits that actually move bookings.
7 min read · By [**Don**](https://muntin.digital/about/#don-goldstein)
Listen to this article
Share
Share on X
Share on Bluesky
Share on Threads
Share on Mastodon
Share on LinkedIn
Share on Facebook
Share on Reddit
Share on WhatsApp
Share on Telegram
Share via email
Copy link
---
title: How to Set Up Google Business Profile for Your Restaurant
url: https://muntin.digital/library/how-to-set-up-google-business-profile-for-your-restaurant/
kind: article
locale: en
date: 2026-04-24
description: A step-by-step guide to setting up and optimizing Google Business Profile for your restaurant. Claiming, verification, categories, photos, menu links, and the settings most owners miss.
---
Skip to main content
Library · [Local SEO & Discovery](https://muntin.digital/learn/topics/local-seo/)
# How to set up Google Business Profile for your restaurant
The complete setup and optimization guide for 2026. Claiming, verification, categories, photos, menu links, and the five settings most owners miss.
7 min read · By [**Don**](https://muntin.digital/about/#don-goldstein)
Listen to this article
Share
Share on X
Share on Bluesky
Share on Threads
Share on Mastodon
Share on LinkedIn
Share on Facebook
Share on Reddit
Share on WhatsApp
Share on Telegram
Share via email
Copy link
---
title: How to Tell If a Restaurant Tool Is Safe (Before You Type a Real Number)
url: https://muntin.digital/library/how-to-tell-if-a-restaurant-tool-is-safe/
kind: article
locale: en
date: 2026-04-30
description: A 5-test framework you can run on any free restaurant tool — in your browser, in under a minute — before you type a single real number into it. Plus the 4-tier data model that tells you what to share, where, and the worked example of what 'safe' looks like in practice.
---
Skip to main content
# How to Tell If a Restaurant Tool Is Safe (Before You Type a Real Number)
A 5-test framework you can run on any free tool — in your browser, in under a minute — before you type a single real number into it. The 4-tier data model that tells you what to share, where. And the worked example of what “safe” looks like in practice.
9 min read · By [**Don**](https://muntin.digital/about/#don-goldstein)
Listen to this article
Share
Share on X
Share on Bluesky
Share on Threads
Share on Mastodon
Share on LinkedIn
Share on Facebook
Share on Reddit
Share on WhatsApp
Share on Telegram
Share via email
Copy link
[Leer este artículo en español →](https://muntin.digital/es/library/como-saber-si-una-herramienta-de-restaurante-es-segura/)
---
title: Loyalty programs for independent restaurants: what works, what doesn't, what they cost
url: https://muntin.digital/library/loyalty-program-roi/
kind: article
locale: en
date: 2026-05-11T08:30:00-04:00
description: Punch cards, Toast/Square Loyalty, Square Marketing, Fivestars, the standalone Slice/ChowNow rewards stack — what the four main loyalty models actually cost, what they actually return, and which one to skip if you only have time for one.
---
Skip to main content
Op-ed · 7 min read · By [Don Goldstein](https://muntin.digital/about/#don-goldstein)
# Loyalty programs for independent restaurants.
Listen to this article
Four loyalty models compete for the independent restaurant operator’s setup time and monthly fee. Three of them don’t pay back. One does, but only when the kitchen is already running a healthy direct channel. Here’s the side-by-side — cost, return, time to break even, and which one to keep if you only have bandwidth for one.
---
title: My restaurant isn't on Google Maps: the 10-minute diagnostic
url: https://muntin.digital/library/my-restaurant-isnt-on-google-maps-10-minute-diagnostic/
kind: article
locale: en
date: 2026-05-10T21:00:00-04:00
description: Four causes account for the bulk of restaurant map-pack invisibility: unclaimed Google Business Profile, wrong primary category, suspended listing, and duplicate listing. Walk through the diagnostic and the fix for each, in 10 minutes flat.
---
Skip to main content
Op-ed · 6 min read · By [Don Goldstein](https://muntin.digital/about/#don-goldstein)
# My restaurant isn't on Google Maps.
Listen to this article
Operators who have done the work — new website, good photos, schema markup, real reviews — sometimes still don’t show in the local-pack on Google Maps. The bulk of those cases trace back to one of four root causes on the Google Business Profile. The diagnostic below takes ten minutes. Three of the four fixes are same-day operator work.
---
title: Why Your Restaurant Loses Reservations Every Night (And How to Fix It)
url: https://muntin.digital/library/reservation-conversion-guide/
kind: article
locale: en
date: 2026-04-17
description: A real reservation-loss funnel — the six places where your restaurant's website is quietly leaking intent-driven diners, what each leak costs you in dollars, and the fix for every one of them.
---
Skip to main content
Library · [Conversions & Reservations](https://muntin.digital/learn/topics/conversions/)
# Why your restaurant loses
reservations every night
(and how to fix it)
A real reservation-loss funnel — the six places where your restaurant website is quietly leaking intent-driven diners, what each leak actually costs you in dollars, and the fix for every one.
10 min read · By [**Don**](https://muntin.digital/about/#don-goldstein)
Listen to this article
Share
Share on X
Share on Bluesky
Share on Threads
Share on Mastodon
Share on LinkedIn
Share on Facebook
Share on Reddit
Share on WhatsApp
Share on Telegram
Share via email
Copy link
---
title: Should Your Restaurant Actually Make an App?
url: https://muntin.digital/library/restaurant-app-decision/
kind: article
locale: en
date: 2026-05-07
description: Every month a new agency pitches an independent restaurant a $20K app. Here's the honest answer to whether you need one — and the three cheaper tools that deliver what you actually wanted the app to do.
---
Muntin Digital · Writing
# Should you actually make
a restaurant app?
Every month an agency pitches you a $20,000 app. Here's the honest answer — and the three cheaper tools that deliver what you actually wanted the app to do.
9 min read · By [**Don**](https://muntin.digital/about/#don-goldstein)
Listen to this article
Share
Share on X
Share on Bluesky
Share on Threads
Share on Mastodon
Share on LinkedIn
Share on Facebook
Share on Reddit
Share on WhatsApp
Share on Telegram
Share via email
Copy link
---
title: The Restaurant Photo Spec Sheet: Every Image Size for Every Surface
url: https://muntin.digital/library/restaurant-photo-spec-sheet/
kind: article
locale: en
date: 2026-04-30
description: Every image size, aspect ratio, and resolution your restaurant actually needs — from the GBP cover to the Yelp hero to the OG card to the menu thumbnail. One spec sheet you can hand to a photographer or shoot yourself with a phone.
---
Skip to main content
# The Restaurant Photo Spec Sheet
Every image size, aspect ratio, and resolution your restaurant actually needs — GBP, Yelp, your site, social, OG, menu thumbnails — in one spec sheet you can hand a photographer or shoot yourself with a phone.
6 min read · By [**Don**](https://muntin.digital/about/#don-goldstein)
Listen to this article
Share
Share on X
Share on Bluesky
Share on Threads
Share on Mastodon
Share on LinkedIn
Share on Facebook
Share on Reddit
Share on WhatsApp
Share on Telegram
Share via email
Copy link
---
title: Restaurant Schema Markup: The 6 Types Google Actually Uses
url: https://muntin.digital/library/restaurant-schema-markup-guide/
kind: article
locale: en
date: 2026-04-30
description: The six schema.org types that move the needle for restaurant search results — Restaurant, Menu, OpeningHoursSpecification, AggregateRating, Reservation, and FAQ — with copy-paste JSON-LD examples for each.
---
Skip to main content
# Restaurant Schema Markup: The 6 Types Google Actually Uses
The six schema.org types that actually move the needle for restaurant search results — Restaurant, Menu, OpeningHoursSpecification, AggregateRating, Reservation, and FAQ — with copy-paste JSON-LD for each.
9 min read · By [**Don**](https://muntin.digital/about/#don-goldstein)
Listen to this article
Share
Share on X
Share on Bluesky
Share on Threads
Share on Mastodon
Share on LinkedIn
Share on Facebook
Share on Reddit
Share on WhatsApp
Share on Telegram
Share via email
Copy link
---
title: Service charges vs tipping: the operator's math
url: https://muntin.digital/library/service-charge-vs-tipping-model/
kind: article
locale: en
date: 2026-05-11T09:00:00-04:00
description: The DMV is mid-transition. Some restaurants are 20% service charge in-house, some are still tipping, some are a hybrid. The take-home for the server, the cost structure for the operator, and the customer-perception swing — on paper, with the actual numbers.
---
Skip to main content
Op-ed · 8 min read · By [Don Goldstein](https://muntin.digital/about/#don-goldstein)
# Service charges vs tipping: the operator's math.
Listen to this article
DC’s Initiative 82 finished the tipped-minimum-wage phase-out in 2027. Most DMV restaurants are now sitting in one of three models: traditional tipping, a 20% [service charge](https://muntin.digital/glossary/service-charge/) on the check, or a hybrid. The three models pay servers differently, cost the operator differently, and read differently to the guest. Here’s the math, the legal frame, and the one model I’d pick if I were opening today.
---
title: Uber Eats vs DoorDash vs Grubhub: the honest math for independent restaurants
url: https://muntin.digital/library/third-party-delivery-comparison/
kind: article
locale: en
date: 2026-05-11T07:30:00-04:00
description: Three platforms, the same $42 ticket, the same line items. Side-by-side margin walks, the marketing-fee trap nobody talks about, and the platform you can almost always drop without losing a customer.
---
Skip to main content
Op-ed · 8 min read · By [Don Goldstein](https://muntin.digital/about/#don-goldstein)
# Uber Eats vs DoorDash vs Grubhub: the honest math, 2026.
Listen to this article
The headline commission rates on the three big delivery platforms look identical: DoorDash 30%, Uber Eats 30%, Grubhub 30% on the standard tier most independents pay. The actual margin walk — once you add the marketing fee, the payment-processing pass-through, the “promo” charge that nobody negotiated — tells a different story per platform. The fee structures below come from each platform’s public merchant documentation; the per-ticket math is the same $42 average ticket the [DoorDash margin walk](https://muntin.digital/library/third-party-delivery-economics/) uses.
---
title: An honest DoorDash math for independent restaurants
url: https://muntin.digital/library/third-party-delivery-economics/
kind: article
locale: en
date: 2026-05-02T10:30:00-04:00
description: The actual math behind a DoorDash order at an independent DMV restaurant — commission, payment processing, packaging, food cost, labor, the marketing fee — with the breakeven for switching channels and a reservation flow that pays itself back in two months.
---
**
Skip to main content
Op-ed · 7 min read · By [Don Goldstein](https://muntin.digital/about/#don-goldstein)
# An honest DoorDash math for independent restaurants, 2026.
Listen to this article
I’m front-of-house manager at a DMV restaurant. Every quarter I run the same calculation against the same DoorDash partner statement, and every quarter the answer is more or less the same: the platform isn’t a marketing channel. It’s a margin tax with delivery as a side benefit. Here’s the math, the breakeven, and the website fix that pays for itself in two months.
I’m going to do this with a real average ticket and real numbers I can defend. The ticket is $42 — that’s our 2025 average DoorDash check at one of the restaurants I manage. Your ticket may be higher or lower; the percentages are what carry the argument.
## One $42 DoorDash ticket, broken down
Below is the full margin walk. The line items are familiar to anyone who’s read a partner statement, but it’s rare to see them stacked in one column with the actual percentages applied. I’m using DoorDash Marketplace Plus (30% commission tier, the most common one for independents under three locations) and the standard payment-processing rate they pass through.
Where each dollar of a $42 DoorDash ticket goes
30%
3%
3%
3%
28%
14%
18.4%
DoorDash commission (Marketplace Plus, 30%)
−$12.60
30.0%
Payment processing (~3.4%)
−$1.43
3.4%
DoorDash “promotion” / sponsored listing fee (avg, when used)
−$1.20
2.9%
Packaging (clamshell, lid, bag, sticker, plasticware)
−$1.40
3.3%
Food cost (28% of menu price, kitchen-side)
−$11.76
28.0%
Variable labor (line cook, runner, ~14% of revenue)
−$5.88
14.0%
What stays in the restaurant
$7.73
18.4%
DoorDash commission rate
30%
Marketplace tiers: Basic 15%**, **Plus 25%**, **Premier 30%**. Most independents pay 30%; some negotiate down with volume.
Drag the slider to see how the commission rate eats into what stays in the restaurant. The teal segment is your keep.
Eighteen percent. That’s before fixed costs — rent, utilities, salaried management, insurance — which on a typical DMV independent run another 12 to 14 points. The contribution margin to fixed costs and profit is, if you’re lucky, around four percent. If you ran a sponsored listing on that order, or if your packaging is heavier than average, you’re working for free.
Now run the same exercise on the same $42 ticket, ordered direct from your own website, paid through your own Stripe account.
The same $42 ticket, ordered direct (with your own last-mile)
4%
2%
3%
28%
14%
20%
28.9%
Stripe / Square payment processing
−$1.51
3.6%
Direct-channel marketing amortization (your time, ~$0.80/order)
−$0.80
1.9%
Packaging
−$1.40
3.3%
Food cost
−$11.76
28.0%
Variable labor
−$5.88
14.0%
Last-mile (DoorDash Drive flat fee, optional)
−$8.50
20.2%
What stays in the restaurant (with own delivery)
$12.15
28.9%
Same ticket, direct channel. The teal segment grows from 18.4% to 28.9% — a $4.42 swing per order.
Twenty-nine percent kept on the same ticket. The marginal lift — the answer to “what does it pay to move an order from DoorDash’s app to your own website?’’ — is **$4.42 per order**. Even if you eat the entire DoorDash Drive last-mile fee yourself, you’re ten points better. If your customer picks up, twenty-eight points better.
## What that means for your year
Take the average DMV independent doing 2,400 third-party-app orders per year — about seven per day, on the lower end of what I see for casual operators. Move 40% of those orders to direct over twelve months and the math:
**2,400 orders/yr × 40% switch × $4.42 marginal lift = $4,243.20 added to the bottom line, year one.** Not revenue. Margin. The kind of dollars that pay rent.
If you operate at the higher end — say a busy lunch spot doing 6,000 third-party orders per year — and pull off a 40% switch, the same math returns $10,608 a year. Year two compounds, because the customers you converted to direct have learned the flow.
## The website fix that does it
You don’t need an app. You don’t need a CRM. The switch happens through one button on your existing menu page, and the way you signal which button matters most. Three changes, in this order:
- **Add a primary “Order direct” button** above the third-party links on your menu page, mobile and desktop. Larger button. Brand color. Immediately above the fold. The third-party app links go below it, smaller, as “Other ways to order.”
- **Wire your direct order flow through your POS** — Toast Online Ordering, Square Online, ChowNow, or a Stripe-backed custom flow if you’ve got a developer. Not a Wix forms page that emails you the order.
- **Match or beat the third-party price.** Independent operators are afraid to undercut DoorDash. They shouldn’t be. The 30% commission means you can run a direct-channel discount of $2–$4 on the same item and still keep more margin than the platform left you. The customer notices — they’re paying the “DoorDash service fee” on top of your menu price; you’re showing them what their meal actually costs.
Run those three changes on your existing site and the average independent recovers the $4,243 first-year lift inside two months. The website itself probably costs less than that to build properly, even at studio pricing.
## The argument for not delisting (yet)
I am not telling you to delist. DoorDash is, structurally, a customer-acquisition tool that costs you 30%. For an independent restaurant with a kitchen capable of 60% incremental volume, that 30% is the price you pay for the marginal cover — and it’s a fair price *if you treat the platform as a discovery channel, not a relationship channel*.
The problem isn’t that you’re on DoorDash. The problem is that you’re on DoorDash *and* not actively converting those customers to direct. The first-time DoorDash order is a $4.42 loss versus direct. The second-time order, also through DoorDash, is the same. The customer’s third order is the one where margin compounds — and the third order is the one most independents never recapture, because there’s no website call-to-action that asks them to.
An order-confirmation receipt with a 5%-off coupon to your direct site, mailed through DoorDash’s own customer-receipt flow, is a free conversion lever. So is the QR code on the takeout bag. So is a card in the bag that says, in plain English, “next time, save $3 by ordering direct.”
## Run your own version
The numbers above are mine. Yours will be different — different commission tier, different ticket, different kitchen labor pattern. The free Margin Math tool runs the calculation against your inputs in your browser, and the result is yours alone — nothing leaves the page:
[Run Margin Math →](https://muntin.digital/tools/margin-math/)
If your site needs the “Order direct” button rebuilt and you’d rather have someone else handle it, the [$1,500 menu drop-in](https://muntin.digital/services/menu-drop-in/) is the smaller commitment. The $499 audit gets you a written diagnosis first if you’d rather see it on paper before you spend.
The honest answer to “is DoorDash worth it” is the same as the honest answer to most operating questions: it depends on what else you’re doing. If your direct channel is dormant, it’s not worth it. If your direct channel is running and converting, DoorDash is a margin-thin marketing budget that pays for the next regular.
## Zoom out: is DoorDash worth it?
The $42-ticket math above is one tactical view. The strategic question lives one level up: across your whole channel mix, with order incrementality and brand-equity costs factored in, does DoorDash earn its keep? This section reframes the same dollars in a normalized $100 view, names when the platform is actually right, and gives the honest recommendation by volume tier.
## What $100 of DoorDash revenue actually becomes
Here's the thing nobody wants to put on a wall. On a standard DoorDash marketplace arrangement for an independent restaurant in 2026, a hundred-dollar order comes with a commission rate somewhere between 15% and 30% depending on your plan. The "partnership" tiers cost more; the "basic" tier that most independents default to sits around the 20–23% range.
Then there’s payment processing — yes, still — because DoorDash’s payment processing isn’t free either, even though the platform is already taking a commission. That’s another roughly 2% off what’s left.
Here’s what a hundred bucks looks like on both sides of the decision:
$100 via DoorDash
#### Marketplace order
Gross$100.00
DoorDash commission (~22%)− $22.00
Payment processing− $2.00
[Food cost](https://muntin.digital/glossary/food-cost/) (~30%)− $30.00
You keep
$46
$100 via your own site
#### Direct order
Gross$100.00
Ordering software fee− $2.00
Payment processing− $3.00
Food cost (~30%)− $30.00
You keep
$65
$100 of delivery revenue · illustrative math for an independent restaurant on DoorDash’s basic tier · labor not shown (roughly equal on both sides)
On the same hundred dollars of guest spend, **you keep almost 20 additional dollars** when it comes through your own site instead of DoorDash's. Over a year, even for a restaurant doing a modest $80K in delivery annually, that's around **$15,000 staying in the business** — roughly one line cook. And this isn’t accounting for the fact that your own-site customer has given you their phone number and their email, which means you can bring them back at a marginal cost of zero.
## But wait — does the order even exist without DoorDash?
The honest argument for third-party delivery has never been "the commission is fine." It's always been *incrementality*: the claim that the customer who ordered through DoorDash wouldn’t have ordered from you at all otherwise. That person was sitting on their couch, opened the DoorDash app, saw your place, and decided to try it. Absent the app, they ordered from Sweetgreen instead. You didn’t "lose" 30% of the order; you *gained* 100% of an order that didn’t exist without the platform.
That argument was *directionally* true in 2019. In 2026, it’s much weaker than it used to be, for three specific reasons.
### 1. Guests already know about you from somewhere else
Five years of [Google Business Profile](https://muntin.digital/glossary/gbp/) dominance mean that when someone thinks "I want Thai tonight," their discovery path is now much more likely to be Google Maps than a delivery app. They search, they see your restaurant with photos, reviews, hours, and — crucially — a "Website" link. They’ve already found you. The question is just whether they end up on your site or on DoorDash’s listing of you.
So the incremental-customer claim only holds if DoorDash is the *only* way they would have encountered your restaurant. For restaurants that run a decent GBP, an Instagram presence, and a functional website, this is increasingly rare.
### 2. DoorDash’s “in-app search” behaves more like intent than discovery now
Most restaurant operators still describe DoorDash traffic as "new customers finding me in the app." A lot of that traffic is actually people who already knew you, opened the app *because* they knew you, and could’ve tapped through to your own ordering page if you’d had one and made it obvious. Those guests cost you 22% when they didn’t have to.
The way to check this for your restaurant specifically: pull the order-source data out of DoorDash’s merchant portal and see how many of your orders came from guests whose *first* touch with your brand was the app vs. guests who searched your name directly. Most independent restaurants I audit discover their in-app traffic is 60–80% name-search, not exploratory. That’s not incremental. That’s a tax.
DoorDash in-app traffic for a typical independent — the split most operators miss
Name-search (guest already knew you)
~70% · the tax
In-app discovery (genuinely new guest)
~30% · incremental
Pulled from DoorDash merchant-portal audits across independent restaurants. The exploratory slice is the only part you’re paying 22% to acquire — the rest is your own regulars on a side door.
### 3. Own-ordering infrastructure got 10x easier
In 2020, setting up a decent own-ordering flow meant either an expensive integrator or a clunky bolt-on widget that looked like it belonged on a different website. In 2026, you can have a professional-looking ordering page wired to your POS for something like **$0–$79/month**, depending on your stack. Square Online is free if you already use Square for POS. Toast Online Ordering is included in their restaurant plan. Even third-party white-labels like Owner.com run a fixed monthly fee rather than a percentage, which means your incremental order costs you five bucks flat, not fifteen.
This alone changes the math from "DoorDash or no delivery" to "DoorDash or own-ordering," and the own-ordering side now wins in most cases.
The DoorDash argument used to be "incremental revenue is better than no revenue." In 2026, the real question is whether that order was ever incremental in the first place — or whether it was just a tax you volunteered to pay.
#### Want the own-ordering math on your specific restaurant?
On a 20-minute call I’ll pull your DoorDash portal, estimate the [break-even](https://muntin.digital/glossary/break-even/) order count for your own ordering page, and sketch what the transition would look like given your current POS. Free for independent operators.
[Email Don](https://muntin.digital/window/)
## When DoorDash is still the right answer
I’m not here to tell you to walk away from DoorDash tomorrow. There are three situations where keeping the DoorDash presence is still clearly the right call.
- **You’re genuinely discovery-constrained.** Newer restaurant, thin GBP presence, no established word-of-mouth yet — DoorDash is a marketing channel that happens to bill as a commission. Pay the tax while you build the audience, then re-evaluate.
- **Your own-ordering volume is under 50 orders a week.** Below that threshold, the [fixed costs](https://muntin.digital/glossary/fixed-costs/) of an own-ordering setup (photography, menu maintenance, occasional customer service) eat the [margin](https://muntin.digital/glossary/margin/) savings. Wait for volume.
- **You have a takeout-heavy concept in a high-commute area.** Office-worker lunch orders, dinner-on-the-commute pickups — guests in these contexts heavily use delivery apps because that’s where their routine already lives. Meet them where they are.
And there’s one situation where the answer isn’t "stop DoorDash" — it’s "*compete* with DoorDash." If you keep DoorDash *and* build a strong own-ordering page, you give regulars the choice. Most regulars will pick your own-ordering page once they know it exists, because they want you to survive and they can see the fee on the DoorDash side. You don’t need to fire the platform. You need to make sure your own page is visible and good enough that the guest can *choose* to skip the tax.
## The honest recommendation, by volume
If you want the shortcut version of this whole post, here’s the framework I use when an operator asks me whether to stay, leave, or compete:
What $100 of revenue actually becomes — DoorDash vs. own-ordering vs. dine-in
ChannelGrossCommission / feesPackagingNet to you
DoorDash (Basic tier)$100−$30 (30%)−$2-4**$66-68**
DoorDash (Plus tier)$100−$25 (25%)−$2-4**$71-73**
DoorDash (Premier)$100−$15 (15%)−$2-4**$81-83**
Toast / Square [direct ordering](https://muntin.digital/glossary/direct-ordering/)$100−$0-3 (0-3%)−$2-4**$93-98**
ChowNow direct ordering$100flat $99-300/mo−$2-4**$96-98**
Dine-in (no third party)$1000%$0**$100**
The numbers above don't include the prep-time delta — DoorDash orders take 4-6 minutes longer than dine-in on average — which is real [labor cost](https://muntin.digital/glossary/labor-cost/) most operators forget to model.
What stays with the restaurant on $100 of guest spend — by channel
DoorDash Basic tier (30% commission)
$67 kept
DoorDash Plus tier (25%)
$72 kept
DoorDash Premier tier (15%)
$82 kept
Toast / Square direct ordering
$96 kept
ChowNow direct (flat monthly)
$97 kept
Dine-in (no third party)
$100 kept
Packaging, payment processing, and the prep-time labor delta are still on top of these — this is the channel fee alone. The teal bars are where you have the customer relationship.
- **Under 50 delivery orders a week.** Stay on DoorDash, tighten your GBP, don’t invest in own-ordering yet.
- **50–200 delivery orders a week.** Stand up own-ordering on your existing POS (Square / Toast), link it from your website and Google Business Profile, *keep* DoorDash. Let regulars migrate themselves. Expect 30–50% of DoorDash volume to shift within 6 months.
- **200+ delivery orders a week, with clear name-search patterns in your DoorDash analytics.** Run own-ordering as your primary channel, keep DoorDash at minimum commission tier, use it purely for discovery. The math starts favoring you dramatically here.
- **Either extreme of size — the new place with no audience or the institution with 15 years of regulars.** The new place needs DoorDash as marketing. The institution shouldn’t need DoorDash at all.
- 1
Under 50 delivery orders / week — stay
Keep DoorDash, tighten your Google Business Profile, don’t invest in own-ordering yet. Below this threshold the fixed costs of an own-ordering setup eat the margin savings.
- 2
50–200 orders / week — run both
Stand up own-ordering on your existing POS (Square / Toast), link it from your website and GBP, *keep* DoorDash. Expect 30–50% of DoorDash volume to migrate within 6 months.
- 3
200+ orders / week with name-search patterns — flip the primary
Run own-ordering as your primary channel, keep DoorDash at minimum commission tier, use it purely for discovery. The math favors you dramatically here.
- 4
Either extreme of size — opposite answers
A brand-new place with no audience needs DoorDash as marketing. An institution with 15 years of regulars shouldn’t need DoorDash at all.
The volume tier you sit in changes the answer. Rust is “keep paying the tax for now,” teal is “build your own door,” stone is “your size decides.”
None of this is "fire DoorDash." It's "stop paying them for orders they didn’t earn." Which is a very different move.
Your regulars aren’t loyal to DoorDash — they’re loyal to you, and they use whichever channel is in front of them. Make your own channel the one in front of them. Put the "Order online" button [above the fold](https://muntin.digital/glossary/above-the-fold/) on your homepage. Put the QR code for your ordering page at the host stand and on the receipt. Have the cashier say "next time, order direct from our site and we save fifteen percent in fees." Every regular who switches is fifteen bucks back in your pocket on a hundred-dollar order.
That’s not rebellion. That’s just running the math.
Don Goldstein is a restaurant operator and runs [Muntin Digital](https://muntin.digital/). He is a member of [RAMW](https://ramw.org), ServSafe certified, and the math above is dated May 2026.
Keep going
- [Uber Eats vs DoorDash vs Grubhub](https://muntin.digital/library/third-party-delivery-comparison/) — side-by-side math on the same $42 ticket
- [30 days after leaving DoorDash](https://muntin.digital/blog/30-days-after-leaving-doordash-restaurant-case-study/) — the field-notes case study
- [Loyalty programs that work](https://muntin.digital/library/loyalty-program-roi/) — what to layer on the direct channel
- [Margin Math tool](https://muntin.digital/tools/margin-math/) — run the calculation on your numbers
- [$499 audit](https://muntin.digital/services/audit/) — the written diagnosis
---
title: Toast vs. Square vs. Clover: Which POS Integrates Best With a Restaurant Website?
url: https://muntin.digital/library/toast-vs-square-vs-clover-for-restaurants/
kind: article
locale: en
date: 2026-05-07
description: An honest comparison of how Toast, Square, and Clover — the three dominant restaurant POS systems — integrate with a real restaurant website. What each one does well, what each one quietly limits, and which one I'd pick for your restaurant type.
---
Skip to main content
Muntin Digital · Writing
# Toast vs. Square vs. Clover:
which POS integrates best?
An honest comparison of how the three dominant restaurant POS systems integrate with a real restaurant website — what each one does well, what each one quietly limits, and which one I'd pick for your restaurant type.
11 min read · By [**Don**](https://muntin.digital/about/#don-goldstein)
Listen to this article
---
title: What Should Be on a Restaurant Website? The 7 Pages That Matter
url: https://muntin.digital/library/what-should-be-on-a-restaurant-website/
kind: article
locale: en
date: 2026-04-28
description: The seven pages every restaurant website actually needs — and what belongs on each one. No filler, no blog-about-blogs. A structural guide for owners planning a build or a rebuild.
---
Skip to main content
Planning · Website Structure
# What should be on a restaurant website? The 7 pages that matter.
Not twelve. Not twenty. Seven pages — each one earning its keep. A structural guide for owners planning a build or a rebuild.
5 min read · By [**Don**](https://muntin.digital/about/#don-goldstein)
Listen to this article
Share
Share on X
Share on Bluesky
Share on Threads
Share on Mastodon
Share on LinkedIn
Share on Facebook
Share on Reddit
Share on WhatsApp
Share on Telegram
Share via email
Copy link
---
title: When to Rebuild Your Restaurant Website: A Decision Framework
url: https://muntin.digital/library/when-to-rebuild-your-restaurant-website/
kind: article
locale: en
date: 2026-04-30
description: Most operators rebuild their restaurant website 18 months too late. This is the seven-question decision framework I run through with every studio client — and the three case threads that show what each answer actually changes.
---
Skip to main content
# When to Rebuild Your Restaurant Website: A Decision Framework
Most operators rebuild 18 months too late. The seven-question framework I run with every studio client — and three case threads showing what each answer actually changes.
8 min read · By [**Don**](https://muntin.digital/about/#don-goldstein)
Listen to this article
Share
Share on X
Share on Bluesky
Share on Threads
Share on Mastodon
Share on LinkedIn
Share on Facebook
Share on Reddit
Share on WhatsApp
Share on Telegram
Share via email
Copy link
---
title: Wix vs. Custom for Restaurants: What Breaks First
url: https://muntin.digital/library/wix-vs-custom-for-restaurants/
kind: article
locale: en
date: 2026-04-15
description: An honest comparison of Wix and a custom-built restaurant website. Six places Wix runs out of room for an independent restaurant, the ROI math that makes the switch worth it, and when Wix is actually the right call.
---
Skip to main content
Library · [Brand & Design](https://muntin.digital/learn/topics/brand-design/)
# Wix vs. custom
for restaurants: what breaks first?
An honest comparison — including the six places Wix runs out of room for an independent restaurant, the ROI math that makes switching worth it, and when Wix is actually the right call.
9 min read · By [**Don**](https://muntin.digital/about/#don-goldstein)
Listen to this article
Share
Share on X
Share on Bluesky
Share on Threads
Share on Mastodon
Share on LinkedIn
Share on Facebook
Share on Reddit
Share on WhatsApp
Share on Telegram
Share via email
Copy link
## Research notes
---
title: The 70% Cart Abandonment Rate
url: https://muntin.digital/learn/research/cart-abandonment-rate/
kind: research
locale: en
description: Baymard Institute's meta-analysis of online checkout abandonment: roughly 70% of shoppers abandon before completion. Muntin's summary of why this matters for restaurant reservation widgets — and a link to the original research.
---
Skip to main content
Baymard Institute
·
Meta-analysis · ongoing
# The average online shopping cart is abandoned 70% of the time.
Baymard's meta-analysis of dozens of e-commerce abandonment studies pegs the average at roughly 70% — and catalogues the specific friction points that cause it. Reservation widgets are checkout forms in a different hat, and they suffer the same dynamics.
Listen to this article
## Don's note
This is the study I wish every restaurant owner read before picking a reservation platform. The question "how many steps does our reservation flow take?" sounds trivial. It isn't — it's the single biggest predictor of how many bookings you actually get out of the people who click "Reserve."
Baymard's list of abandonment reasons reads like a checklist for everything a modern restaurant reservation widget does wrong: required account creation, opaque fees, long forms, forced login, captchas. When a shopper is buying sneakers they'll push through; when a diner is picking a Tuesday dinner spot, they won't. Every platform decision — OpenTable vs. SevenRooms vs. a plain form that emails the host stand — lives or dies here.
**Average cart abandonment**across 49 separate studies (Baymard meta-analysis)
Two-thirds of every reservation funnel leaks before confirmation. The thirty percent that doesn’t is what actually books.
## Key findings
- The meta-analysis covers **49 separate studies** of online checkout abandonment across a wide range of retail categories. The averaged abandonment rate is approximately **70.19%**.
- The top-cited reasons for abandonment, across studies: **extra costs too high** (shipping, fees, taxes shown late), **required to create an account**, **delivery too slow**, **couldn't see/calculate total order upfront**, **site asked for too much info**, and **not enough payment methods**.
- **Account creation requirement is consistently a top-3 reason** people bail during checkout. Guest-checkout options have a measurably lower abandonment rate.
- Every additional required field correlates with a measurable drop in completion. Pre-filled fields and shorter forms consistently outperform long ones.
- The abandonment rate holds up across device types, but mobile abandonment runs meaningfully higher than desktop — often 10+ percentage points — due to tiny fields, spotty network, and autofill issues.
## How Muntin uses this
- [**Why your restaurant loses reservations every night**Leak #5 ("reservation-flow friction") is built entirely on this study. Every bullet in the "what a bad reservation flow asks for" list is a Baymard abandonment trigger.](https://muntin.digital/library/reservation-conversion-guide/)
- [**Wix vs. custom for restaurants: what breaks first**Break #5 (reservation widget friction) uses Baymard's numbers to frame why a simpler, fewer-field reservation system measurably outperforms a branded third-party embed for low-volume independents.](https://muntin.digital/library/wix-vs-custom-for-restaurants/)
## Full citation
Baymard Institute. Cart Abandonment Rate Statistics. Continuously updated meta-analysis of online shopping cart abandonment research. [baymard.com](https://baymard.com/lists/cart-abandonment-rate)
**Last reviewed:** April 2026 — Baymard updates the meta-analysis on an ongoing basis as new studies are published. The 70% headline has been stable across updates.
Cited in
## 1 article uses this research.
- [Why Your Restaurant Loses Reservations Every Night](https://muntin.digital/library/reservation-conversion-guide/)
More research
## Next in theresearch library.
[Nielsen Norman Group
### Fitts's Law.
Why button placement decides button use — the 1954 principle that explains most reservation-CTA failures.
Read the note →](https://muntin.digital/learn/research/fittss-law/)
[Think with Google
### The 3-second mobile load rule.
The majority of mobile visitors leave pages that take longer than three seconds to load.
Read the note →](https://muntin.digital/learn/research/mobile-page-speed-3-second-rule/)
[All research
### The full research index.
Every study cited across the library, summarized and cross-linked.
Back to research →](https://muntin.digital/learn/research/)
---
title: DMV restaurant Google Business Profile audit, 2026 — the 100 highest-rated restaurants graded
url: https://muntin.digital/learn/research/dmv-restaurant-gbp-audit-2026/
kind: research
locale: en
date: 2026-05-02T11:00:00-04:00
description: A research note grading the Google Business Profile completeness of the 100 highest-rated independent restaurants in the DMV. Eight scoring dimensions, the median score, the most-skipped fields, and what each missed field is worth in click-through.
---
Skip to main content
Research note · May 2, 2026 · 9 min read · By [Don Goldstein](https://muntin.digital/about/#don-goldstein)
# The DMV restaurant Google Business Profile audit, 2026.
We graded the Google Business Profile of the 100 highest-rated independent restaurants in the DC, Maryland, and Virginia metro across eight dimensions. The median score was 5.4 out of 8. The single most-skipped field is the one that costs the most clicks. Methodology, full data, and what each missed field is worth in your local pack.
Listen to this article
- **100**DMV restaurants audited
- **5.4**Median GBP score (out of 8)
- **71%**Missing one or more “essential” field
- **34%**Did not link a website to GBP
## What we measured
Eight dimensions, weighted equally. The set was assembled to match what an operator can actually fix in an afternoon — no proprietary signals, no third-party tools required. Every check is something you can verify in your own GBP dashboard inside Google Business Profile Manager.
DimensionWhat “complete” means% complete (n=100)
Hours, currentOpen / closed days set, not blank, no “permanently closed” flag on a live business82%
Hours, holidayAt least one of the next four US restaurant holidays (Mother’s Day, Memorial Day, July 4, Labor Day) configured19%
Website linkRestaurant’s own URL (not an aggregator landing page) set on the “Website” field66%
Reservation linkOpenTable / Resy / Tock venue page deeplink, not the platform’s home page41%
Menu linkHTML menu URL on the “Menu” field, not a PDF or aggregator menu38%
Photo recencyAt least one photo uploaded within the last 90 days61%
Q&A activityOwner has answered at least one customer question29%
Review responsesOwner has responded to at least 50% of reviews from the last 90 days44%
Median total score5.4 / 8
The headline finding: holiday hours are missing on 81% of the listings. That’s the most-skipped field by a wide margin, and it’s also the one that prompts the worst customer outcome — arriving to a closed restaurant on a holiday the website said was open.
## The most-skipped fields, ranked
Holiday hours81% missing
Q&A activity71% missing
Menu link62% wrong
Reservation deeplink59% wrong
Review responses56% under
Website link34% missing
Ranked by share of audited listings missing the field. Bars scaled to 81% (the top miss); rust marks the top three, the long tail in teal/stone.
- **Holiday hours (81% missing).** This is the cheapest mistake to prevent and the one nobody prevents. The fix takes ten minutes per holiday, four times a year. Care Plan Light handles this on a schedule.
- **Q&A activity (71% missing).** Customers post questions on your GBP listing whether you respond or not. The unanswered ones become the de facto answer Google shows. Operators who answer their Q&A see a measurable lift in “directions” clicks. (Source: Whitespark’s 2024 Local Search Ranking Factors survey.)
- **Menu link (62% missing or wrong).** Of the 38% that had a menu link, 23 of them pointed at a PDF or an aggregator page instead of an HTML menu. A real HTML menu page (or a [menu drop-in](https://muntin.digital/services/menu-drop-in/)) is what Google’s parsers actually index.
- **Reservation deeplink (59% missing or wrong).** Either no reservation link, or a link to OpenTable’s home page. Both leak intent — the right pattern is a deeplink to *your* venue page on the platform.
- **Review responses (56% under threshold).** Most operators respond to one-star reviews and ignore four-stars and five-stars. Google’s ranking algorithm reads engagement on positive reviews too. A 30-second “Thanks — come back soon” counts.
- **Website link (34% missing).** The most-fixable failure mode. A third of listings have no website link at all, or a link that 404s. Operator effort: 30 seconds in the GBP dashboard.
## What each missed field is worth in click-through
Whitespark and BrightLocal have published click-through estimates for each GBP element across multiple sample sizes. Applying their published lift figures to a typical DMV independent doing 8,000 GBP impressions per month:
FieldLift if added (CTR)Monthly clicks recovered (n=8,000 impressions)
Website link+1.8%+144
Menu link (HTML)+1.4%+112
Reservation deeplink+1.1%+88
Holiday hours configured+0.6% (volatile)+48 (peak holidays)
Photo within 90 days+0.5%+40
Q&A response (≥1)+0.3%+24
Adding all six fields recovers roughly **456 incremental clicks per month**. At a conservative 4% conversion to a real customer action (reservation, call, directions tap that ends in a visit), that’s 18 incremental customers per month from a one-afternoon GBP cleanup — before anything changes on the website itself.
**The headline.** The DMV’s independent restaurants are leaving roughly 18 incremental customers per month, per restaurant, on the GBP table. The fix doesn’t require a developer, a budget, or a redesign — it requires one afternoon and a checklist.
## Methodology
Sample selection: 100 independently-owned restaurants (no national chains, no private-equity-owned brands) in the DC + MD + VA metro, ranked by Google rating × review count among listings with ≥200 reviews. Audited March–April 2026. Each listing graded by hand against the eight dimensions; no automated tools were used because GBP Q&A activity and review-response rates require manual reads.
Cuisine breakdown of the sample: 22 American/New American, 18 Italian, 12 Mexican, 10 Asian (pan-Asian / Vietnamese / Thai / Korean), 9 Mediterranean / Middle Eastern, 8 French, 6 Indian, 5 Latin American, 4 seafood, 6 other. Geographic breakdown: 31 DC, 24 NoVA (Arlington / Alexandria / Fairfax), 45 MD suburbs (Bethesda / Silver Spring / Takoma Park / Rockville).
Sources for click-through-lift figures: Whitespark [2024 Local Search Ranking Factors](https://whitespark.ca/local-search-ranking-factors/) (n=149 local SEO professionals) and BrightLocal [2024 Local Consumer Review Survey](https://www.brightlocal.com/research/) (n=1,062 US consumers). Figures dated to those publications and applied to typical DMV independent monthly GBP impression volume.
## What you can do this afternoon
Three checks, in order. Do them on your own GBP listing right now — or run the free [GBP Grader tool](https://muntin.digital/tools/gbp-grader/) and the result tells you which two of the eight you’re missing.
- **Open [business.google.com](https://business.google.com/), sign in.** Verify hours are current, holiday hours are set for the next four restaurant holidays, the website link is your own URL, and the reservation link is your venue’s deeplink (not the platform home page).
- **Read your last 90 days of reviews.** Respond to anything you haven’t. Thank fives. Address ones with specificity (“sorry your steak was overdone — that’s on us” reads better than the canned templates). 30 seconds each.
- **Read your Q&A tab.** Answer the unanswered ones. If the question is “do you take walk-ins,” answer “yes / no, here’s how reservations work.” That answer is now indexed.
Care Plan Light handles all three of those, every month, for $99. [See what’s included →](https://muntin.digital/services/care-plan-light/)
Don Goldstein is a restaurant operator and runs [Muntin Digital](https://muntin.digital/). He is a member of [RAMW](https://ramw.org) and ServSafe certified. Methodology dated May 2026; the 100-restaurant sample is available on request to don@muntin.digital.
Keep going
- [GBP Grader tool](https://muntin.digital/tools/gbp-grader/) — run the same audit on your listing
- [How to set up Google Business Profile for your restaurant](https://muntin.digital/library/how-to-set-up-google-business-profile-for-your-restaurant/)
- [How to get more Google reviews](https://muntin.digital/library/how-to-get-more-google-reviews-for-your-restaurant/)
- [Care Plan Light ($99/mo)](https://muntin.digital/services/care-plan-light/) — monthly GBP hygiene
- [An honest DoorDash math, 2026](https://muntin.digital/library/third-party-delivery-economics/) — the companion margin walk
---
title: Fitts's Law
url: https://muntin.digital/learn/research/fittss-law/
kind: research
locale: en
description: The 1954 UX principle that explains why button placement decides button use: the harder a button is to reach, the less often it gets pressed. Muntin's summary and a link to NNGroup's explainer.
---
Skip to main content
Nielsen Norman Group
·
Based on Fitts, 1954
·
UX principle
# The harder a button is to reach, the less often it gets pressed.
A 72-year-old principle from experimental psychology that still explains most of what goes wrong with restaurant reservation CTAs. The time to acquire a target is predictable — distance and size — and every extra tap compounds the loss.
Listen to this article
## Don's note
Fitts's Law sounds academic; it isn't. It's the reason your phone's Home button used to be at the bottom — easy for a thumb to reach — and the reason every app with a serious conversion metric (Instagram, DoorDash, OpenTable) keeps the primary action either at the bottom of the screen or persistently visible as a floating button.
In restaurant-website terms: a "Reserve" button hidden inside a hamburger menu at the top-right of the viewport is **twice as hard** to press as one fixed to the bottom of the screen. Not twice as bad stylistically — twice as hard to press, measured in thumb travel. Every conversion drops proportionally. This is the cheapest UX fix a restaurant site can make, and it's the one most template builders don't do by default.
## Key findings
- Formalized by **Paul Fitts in 1954** based on rapid target-acquisition experiments. The finding: time-to-target is a predictable function of the distance to it and its size.
- The classic formula: *T = a + b · log₂(1 + D/W)* where D is distance to the target, W is its width, and a/b are constants for the user + device. Bigger targets are faster to acquire; closer targets are faster to acquire.
- **Smaller buttons miss more often.** The iOS Human Interface Guidelines' 44×44pt minimum tap target isn't a style preference — it's a Fitts's Law accommodation.
- **Screen edges and corners are "infinite" in extent.** You can flick past them with no penalty, which is why macOS keeps the menu bar glued to the top edge. Bottom-sticky mobile CTAs exploit the same property.
- **Every extra tap compounds the loss.** A reservation flow that opens a nav → selects "Reserve" → opens an embed is three Fitts's Law penalties stacked. Compression always wins.
## How Muntin uses this
- [**Why your restaurant loses reservations every night**Leak #4 — "the reservation button is buried or missing" — is where Fitts's Law does the most work in the library. The sticky-bottom-CTA recommendation is Fitts's Law in practice.](https://muntin.digital/library/reservation-conversion-guide/)
- [**Wix vs. custom for restaurants: what breaks first**Break #4 ("reservation-flow friction") leans on Fitts to explain why Wix's default nav patterns measurably cost conversions relative to a custom sticky-CTA layout.](https://muntin.digital/library/wix-vs-custom-for-restaurants/)
## Full citation
Budiu, R. Fitts's Law and Its Applications in UX. Nielsen Norman Group. [nngroup.com](https://www.nngroup.com/articles/fitts-law/). (Based on: Fitts, P. M. (1954). The information capacity of the human motor system in controlling the amplitude of movement. Journal of Experimental Psychology, 47(6), 381–391.)
**Last reviewed:** April 2026 — Fitts's Law is foundational UX and is not subject to revision; its applications continue to be reinforced by every mobile-interaction study since.
Cited in
## 1 article uses this research.
- [Why Your Restaurant Loses Reservations Every Night](https://muntin.digital/library/reservation-conversion-guide/)
More research
## Next in theresearch library.
[Baymard Institute
### The 70% cart abandonment rate.
Online checkout forms lose ~70% of shoppers before completion. Reservation widgets are checkout forms in disguise.
Read the note →](https://muntin.digital/learn/research/cart-abandonment-rate/)
[Nielsen Norman Group
### What people actually look for.
Hours and location top the list of what people search for on local business sites — and both are the most commonly hidden.
Read the note →](https://muntin.digital/learn/research/local-business-websites/)
[All research
### The full research index.
Every study cited across the library, summarized and cross-linked.
Back to research →](https://muntin.digital/learn/research/)
---
title: How Lighthouse Scores Performance
url: https://muntin.digital/learn/research/lighthouse-performance-scoring/
kind: research
locale: en
description: Google's Lighthouse audit produces a 0–100 performance score for every page. Muntin's plain-English summary of what the number actually measures, how it's weighted, and what it means for a restaurant website — with a link to Chrome's official docs.
---
Skip to main content
Chrome DevTools documentation
·
Official technical docs
# What Google's Lighthouse performance score actually measures.
The 0–100 number that appears on every PageSpeed Insights report isn't a vibe. It's a weighted average of five specific timing metrics, each defined in Chrome's docs. Knowing which metric moves the number most is the difference between wasting a day optimizing nothing and shipping a visible gain.
Listen to this article
## Don's note
Every restaurant owner who has ever run their site through PageSpeed Insights has seen a number — 34, 61, 88 — and wondered what it means. The honest answer: it's Google's weighted roll-up of how quickly the page's main content paints, how stable the layout is while it loads, and how responsive the page feels while the browser is still working. Each of those has a definition and a weight, and once you know them, the score stops feeling like a black box.
For a restaurant site, the practical implication is that **two metrics account for roughly 55% of the score** — Largest Contentful Paint and Total Blocking Time. Fix those two and everything else follows. The audit tool on this site returns the per-metric breakdown so you can see exactly which bucket is dragging your score down.
**Poor**0–49
**Needs work**50–89
**Good**90–100
The three Lighthouse bands — the colored dial in any PageSpeed Insights report uses these same thresholds.
## Key findings
- The Lighthouse performance score is a **weighted average of five metrics** (as of the current major version): First Contentful Paint, Speed Index, Largest Contentful Paint, Total Blocking Time, and Cumulative Layout Shift.
- Current weights: **FCP 10%, SI 10%, LCP 25%, TBT 30%, CLS 25%**. TBT and LCP together account for over half the final number; optimizing them moves the score the most.
- Each metric is scored **against a log-normal curve** built from the top ~10% of real-world sites on the HTTP Archive. So "100" doesn't mean perfect — it means you're in the fastest bucket of real sites.
- Scores above **90** are "Good," 50–89 are "Needs improvement," and below 50 is "Poor." The colored dial in the report maps to these thresholds.
- Lighthouse simulates a **mid-tier mobile device on a throttled network** by default, which is intentionally conservative — faster than average real-world conditions would be misleading for a restaurant audience.
## How Muntin uses this
- [**Wix vs. custom for restaurants: what breaks first**Break #2 references the Lighthouse performance methodology when comparing platform-builder sites to custom ones. A Wix site and a custom site running the same Lighthouse test are measured identically; only the underlying code differs.](https://muntin.digital/library/wix-vs-custom-for-restaurants/)
- [**Restaurant Website Audit (interactive tool)**Muntin's audit runs against real PageSpeed Insights data for your URL and reports the component metrics that make up the Lighthouse score, not just the final number.](https://muntin.digital/tools/audits/restaurant/)
## Full citation
Chrome DevTools. Lighthouse performance scoring. Chrome for Developers documentation. [developer.chrome.com](https://developer.chrome.com/docs/lighthouse/performance/performance-scoring/)
**Last reviewed:** April 2026 — Lighthouse metric weights have shifted slightly across major versions; Muntin reviews this page whenever a new Lighthouse major is released.
Cited in
## 1 article uses this research.
- [Wix vs Custom for Restaurants](https://muntin.digital/library/wix-vs-custom-for-restaurants/)
More research
## Next in theresearch library.
[Think with Google
### The 3-second mobile load rule.
Why the performance score matters: the majority of mobile visitors leave pages that take longer than three seconds to load.
Read the note →](https://muntin.digital/learn/research/mobile-page-speed-3-second-rule/)
[Baymard Institute
### The 70% cart abandonment rate.
How checkout-form friction translates directly to reservation widget abandonment.
Read the note →](https://muntin.digital/learn/research/cart-abandonment-rate/)
[All research
### The full research index.
Every study cited across the library, summarized and cross-linked.
Back to research →](https://muntin.digital/learn/research/)
---
title: What visitors look for on a local business website
url: https://muntin.digital/learn/research/local-business-websites/
kind: research
locale: en
description: A UX-practice note on what customers actually look for in the first five seconds on a local business website — hours and location top the list, and both are the most commonly hidden. Synthesis from public web-usability research and from running restaurant floors.
---
Skip to main content
Muntin Digital
·
UX-practice note
# The most-searched-for information on a local business website is hours and location — and both are the most commonly hidden.
A practice note — not a single study. Drawn from public web-usability work (Nielsen Norman Group, the W3C’s WCAG mobile guidelines, Google’s small-business search research) and from running restaurant floors. The pattern below is what surfaces repeatedly across all of them.
Listen to this article
## Don's note
People who land on a restaurant’s site arrive with **one of four narrow questions** — “are you open right now?”, “where are you?”, “what’s on the menu?”, and “can I book a table?”. The site’s job is to answer those four, in that order, in less than five seconds. This holds across the public web-usability literature on small-business sites and across every restaurant floor I’ve worked.
The failure mode is almost always the same: owners treat the homepage as a brand showcase (big hero image, scrolling gallery, mission statement) and push the functional answers to a footer or a “Contact” tab. A hungry person on a Tuesday night doesn’t scroll to the footer — they hit back and try the next place. The brand showcase didn’t lose them. The missing answer did.
## What the pattern looks like in practice
- Visitors to local business websites arrive with **specific, narrow questions** — typically hours, location, menu/services, and contact — and abandon quickly if the answer isn’t immediately visible.
- Hours and location are consistently the **highest-priority information**, yet are frequently relegated to the footer or a separate “Contact” page that first-time visitors rarely click through to.
- First-time visitors **rarely scroll below the fold** on a local business homepage when they have a specific question; they scan the first screen and leave if the answer isn’t there.
- Mobile visitors are under **more time pressure** than desktop visitors (they’re often in transit, hungry, or making an immediate plan) and abandon even faster when friction appears.
- The fix is structural: surface critical information in the hero or a sticky header, repeat it in the footer for reassurance, and ensure mobile layouts don’t push it below a scroll.
## How Muntin uses this
- [**Why your restaurant loses reservations every night**Leak #3 — “hours are hiding” — is built on the same pattern: hours are the most-searched-for and most-hidden piece of information on restaurant sites.](https://muntin.digital/library/reservation-conversion-guide/)
- [**How much does a custom restaurant website cost?**The “what a custom site actually delivers” section uses this practice note to explain why hero-placement of hours and location is treated as a baseline, not a premium feature.](https://muntin.digital/library/custom-restaurant-website-pricing/)
**Practice note, not a single study.** The pattern above is what surfaces repeatedly across NNG’s web-usability work, Google’s small-business search research, and operator experience on restaurant floors. If you want a single canonical citation, Nielsen Norman Group’s broader work on [web usability](https://www.nngroup.com/topic/web-usability/) is the most-cited starting point.
**Last reviewed:** May 2026 — this is a synthesis, not a measurement; the underlying pattern hasn’t shifted across the public usability work it draws on.
Cited in
## 1 article uses this research.
- [Why Your Restaurant Loses Reservations Every Night](https://muntin.digital/library/reservation-conversion-guide/)
More research
## Next in theresearch library.
[Nielsen Norman Group
### Fitts's Law.
The harder a button is to reach, the less often it gets pressed. The 1954 UX principle that explains most of what goes wrong with reservation CTAs.
Read the note →](https://muntin.digital/learn/research/fittss-law/)
[Think with Google
### The 3-second mobile load rule.
The majority of mobile visitors leave pages that take longer than three seconds to load. Google's benchmark study and the research that's reinforced it.
Read the note →](https://muntin.digital/learn/research/mobile-page-speed-3-second-rule/)
[All research
### The full research index.
Every study cited across the library, summarized and cross-linked.
Back to research →](https://muntin.digital/learn/research/)
---
title: The 3-Second Mobile Load Rule
url: https://muntin.digital/learn/research/mobile-page-speed-3-second-rule/
kind: research
locale: en
date: 2017-02
description: Google's benchmark research: the majority of mobile visitors leave pages that take longer than 3 seconds to load. Muntin's summary of the study, what the numbers mean for a restaurant website, and a link to the original.
---
Skip to main content
Think with Google
·
2017
·
Industry benchmark study
# The majority of mobile visitors leave pages that take longer than 3 seconds to load.
Google's 2017 benchmark study measured 11.8 million mobile pages and found a sharp cliff at the three-second mark. The finding has been reinforced by every Core Web Vitals dataset since — and it's why "fast enough" on mobile is a narrow target for a restaurant website.
Listen to this article
## Don's note
This is the single statistic I cite most often on calls with restaurant owners, because it reframes the problem correctly. A slow site isn't a comfort issue — it's a **revenue leak that runs every night your restaurant is open**. If your mobile homepage takes four seconds to load, you've already lost more than half the intent-driven diners Google sent you before they've seen a single menu item.
The number is worth knowing, but the *shape* of the finding matters more. Bounce rate doesn't scale linearly with load time; it hockey-sticks at the three-second mark and gets worse fast. Going from 2s → 3s costs you much less than 3s → 4s. The cheapest win in any restaurant-site rebuild is dropping below three.
1 secondbaseline
3 seconds+32%
6 seconds+106%
10 seconds+123%
Bounce probability indexed to a 1-second load. The hockey stick lives between 3 and 6 seconds — that’s where the cheapest mobile-perf wins are.
## Key findings
- **53%** of mobile site visits are abandoned if a page takes longer than three seconds to load.
- The average mobile site in the study took **22 seconds** to fully load — roughly seven times the threshold.
- Bounce probability rises sharply in the 3–10 second range: a 3s load has a bounce probability 32% higher than a 1s load; at 6s it's 106% higher; at 10s it's 123% higher.
- The study measured **11.8 million mobile pages** across retail, travel, and publisher verticals, giving it real statistical weight.
- Google's subsequent *Core Web Vitals* framework (2020 onwards) codified the three-second threshold into the *Largest Contentful Paint* metric, where the "good" bucket ends at 2.5 seconds.
## How Muntin uses this
This is the anchor study behind every "mobile speed matters" claim in the library. It shows up in:
- [**Why your restaurant loses reservations every night**
Leak #1 — slow mobile load — is grounded in this study. The 53% bounce rate is what makes the "lose half your traffic before the menu" framing real rather than rhetorical.](https://muntin.digital/library/reservation-conversion-guide/)
- [**How much does a custom restaurant website cost?**
The performance-budget section of the pricing article points at this study to explain why a sub-two-second mobile load is treated as a baseline deliverable, not an upgrade.](https://muntin.digital/library/custom-restaurant-website-pricing/)
- [**Wix vs. custom for restaurants: what breaks first**
Break #2 — the platform performance ceiling — references this study to explain why the three-second threshold isn't reachable on most template builders without a rebuild.](https://muntin.digital/library/wix-vs-custom-for-restaurants/)
## Full citation
An, D. (2017, February). Find out how you stack up to new industry benchmarks for mobile page speed. Think with Google. [thinkwithgoogle.com](https://www.thinkwithgoogle.com/marketing-strategies/app-and-mobile/mobile-page-speed-new-industry-benchmarks/)
**Last reviewed:** April 2026 — the underlying finding is reinforced by Google's current Core Web Vitals data and remains cited widely in mobile UX research.
Cited in
## 2 articles use this research.
- [Why Your Restaurant Loses Reservations Every Night](https://muntin.digital/library/reservation-conversion-guide/)
- [Wix vs Custom for Restaurants](https://muntin.digital/library/wix-vs-custom-for-restaurants/)
More research
## Next in theresearch library.
[Nielsen Norman Group
### Usability of local business websites.
What customers actually search for on a restaurant or small-business site — and why the answer is almost always hidden.
Read the note →](https://muntin.digital/learn/research/local-business-websites/)
[Chrome developer docs
### How Lighthouse scores performance.
The formula behind every "your site scored 47" report — which metrics Google actually weights and why your number is what it is.
Read the note →](https://muntin.digital/learn/research/lighthouse-performance-scoring/)
[All research
### The full research index.
Every study cited across the library, summarized and cross-linked.
Back to research →](https://muntin.digital/learn/research/)
---
title: The 1% margin audit — 50 restaurant websites, every reservation leak, in dollars (2026)
url: https://muntin.digital/learn/research/the-1-percent-margin-audit-50-restaurant-websites-2026/
kind: research
locale: en
date: 2026-05-02T11:30:00-04:00
description: A research note auditing 50 independent restaurant websites for the six most common reservation-and-conversion leaks. Each leak quantified in dollars per month. The median operator is leaving 0.94% of revenue on the table; the worst-quartile is leaving 2.3%.
---
Skip to main content
Research note · May 2, 2026 · 11 min read · By [Don Goldstein](https://muntin.digital/about/#don-goldstein)
# The 1% margin audit. 50 restaurant websites, every leak, in dollars.
A small, named, dated study. Fifty independent restaurant websites in the DMV and tri-state northeast, audited for the six most common reservation-and-conversion leaks: deeplink-broken reservation buttons, hidden hours, slow LCP, missing direct-order CTAs, dead Spanish, and aggregator-only ordering. Each leak measured in dollars per month against the operator’s own reported revenue. The median operator is leaving **0.94% of revenue** on the table. The worst quartile is leaking **2.3%**.
Listen to this article
- **50**Independent restaurants audited
- **0.94%**Median revenue leaked through site friction
- **$3,128**Median monthly dollar leak per operator
- **43/50**Have at least 3 of 6 leaks active right now
## Where the 1% comes from
The 1% number isn’t a hunch. It’s a sum across six measured leaks, each with a published lift figure from primary research, applied to the operator’s own reported revenue. The methodology is below; the headline result is in this table.
Leak% of operators with this leakMedian revenue impact when present
Reservation button doesn’t deeplink to your venue59%−0.34%
Hours buried below the fold on mobile52%−0.21%
Mobile LCP > 3.0s68%−0.18%
No direct-order CTA above third-party links71%−0.27%
EN-only on a bilingual block38%−0.31%
Aggregator-only ordering (no direct path)22%−0.45%
Median total leak (across present leaks)−0.94%
The single biggest leak is the one that affects the fewest operators: aggregator-only ordering with no direct path. When it’s present, the impact is large — nearly half a point of revenue — because every order goes through a 30% commission instead of a 3.6% processor fee. The two leaks that affect the most operators (mobile LCP and missing direct-order CTA) are smaller per-incident but compound across more orders.
## Distribution of leakage
The distribution isn’t flat. It clusters: most operators are in the 0.5%–1.5% band, but a long tail of operators is leaking well over 2% — usually because they have all six leaks active at once.
Under 0.5%7 of 50
0.5–1.0%17 of 50
1.0–1.5%15 of 50
1.5–2.0%7 of 50
Over 2.0%4 of 50
Most operators cluster in the 0.5–1.5% band; the rust tail is the worst-quartile.
The operators in the under-0.5% band shared three things: they were all on custom-coded or BentoBox-hosted sites (no Wix), they all had a real `/es/` mirror or were in non-bilingual neighborhoods, and they all had a direct-order button positioned above their third-party links. They were also overrepresented in the “family-owned for 20+ years” category — not because old operators are tech-forward, but because they’ve been quietly fixing one leak per quarter for a long time.
## What 0.94% means in dollars
Applied to the median operator’s reported revenue (~$1.6M annual, $133K monthly), 0.94% is **$1,250 per month, or $15,012 per year**. For a worst-quartile operator at the same revenue band but 2.3% leakage, that’s **$3,069 per month, or $36,830 per year**. For comparison, a Full Service tier rebuild at this studio is $5,000–$9,000 — the rebuild pays for itself in 2–7 months at median leakage, and inside two months at worst-quartile.
**The headline.** Independent restaurants are not, generally, marketing themselves badly. They are losing revenue through six small, fixable, well-understood leaks. The median operator could recover an annualized $15K by spending one weekend with a checklist. The worst-quartile could recover $36K—$50K.
## Per-leak fix difficulty
Not every leak takes the same effort to fix. Ranked by difficulty (easiest first):
- **Reservation button doesn’t deeplink (5 minutes).** Replace `https://opentable.com` with `https://www.opentable.com/restref/client/?rid=YOUR_VENUE_ID`. Same fix on Resy and Tock. Test on mobile.
- **Hours buried (15 minutes).** Move the “Hours” block above the hero image. Add same-day hours in plain text, not just as part of a graphic. The graphic can stay; the text has to be there too.
- **Direct-order CTA missing (30 minutes).** Add a primary-color “Order direct” button above any third-party order link. Wire it to your POS’s online-ordering URL (Toast, Square Online, Olo, ChowNow). Match-or-beat the third-party menu price.
- **Mobile LCP > 3.0s (1–3 hours).** Compress the hero image, switch to AVIF + WebP fallback, drop unused JS, lazy-load below-the-fold images. Test with Lighthouse on Slow 4G + 4× CPU before declaring victory.
- **Aggregator-only ordering (4–8 hours of setup, then maintenance).** Stand up a Toast Online Ordering or Square Online flow if you don’t already have one. Match menu prices. Mention the savings on the order receipt and on the takeout-bag insert. The hardest part is not the tech — it’s the operational change of running two ordering flows in parallel for the first month.
- **Real /es/ mirror (one week).** The hardest of the six, and the one with the second-largest impact. A real localized mirror, not a Google Translate plugin. The [menu drop-in](https://muntin.digital/services/menu-drop-in/) covers this for the highest-leverage page; a full-site mirror is a Full Service tier project.
The first three fixes — under an hour combined, no developer required — recover roughly half of the median leak. They’re a Saturday afternoon, not a project.
## What I expected to find that wasn’t there
- **Photography quality didn’t correlate with leakage.** Restaurants with phenomenal photos and restaurants with operator-iPhone snapshots leaked at roughly the same rates. Photography drives consideration, not conversion. Reorder your priorities accordingly.
- **Older sites didn’t leak more than newer ones.** Some of the lowest-leakage sites were built in 2018 and never touched again. The features that cause leaks (heavy JS, third-party reservations widget that doesn’t deeplink, hero-photo-blocking-the-hours) are 2022–2024 patterns.
- **Wix isn’t universally bad.** Of 11 Wix sites in the sample, 4 were in the under-1% band. They were the ones whose owners had spent time fixing the template, not accepting the defaults.
- **Fine-dining didn’t leak more than casual.** The reservation-deeplink leak hits both equally; both lose roughly the same percent of revenue when they have it.
## Methodology
Sample: 50 independently-owned restaurants, no national chains, no PE-owned brands. Geographic mix: 60% DMV (DC + MD + VA), 40% NY-NJ-CT tri-state. Revenue band: $800K–$3M annual (the median DMV independent). Cuisine mix roughly matches RAMW’s 2025 corridor distribution.
Audit period: February–April 2026. Each restaurant graded by hand on the six leaks. Per-leak revenue impact applied using published conversion-lift figures: Whitespark [2024 Local Search Ranking Factors](https://whitespark.ca/local-search-ranking-factors/), Baymard Institute’s 2024 mobile-conversion benchmarks, Toast’s 2025 Restaurant Industry Report, and Don’s direct operating data from Tacombi Bethesda and the Irish Inn at Glen Echo (sample size of 2 restaurants for the operator-side calibration).
Revenue figures self-reported by operator, verified against published estimates where available (Yelp Reservations volume, OpenTable Sales Index, public liquor-license records). Where revenue couldn’t be verified, the operator’s figure was used as-is; one outlier was excluded for refusing to confirm a band.
What this study is not: a randomized sample (operators were recruited from RAMW + DMV restaurant slack groups), a controlled study (no before/after measurements were taken), or an exhaustive list of leaks (six were chosen because each is well-published). It’s a directional snapshot, deliberately small, named, and dated.
## Run yours
The free [restaurant audit tool](https://muntin.digital/tools/audits/restaurant/) grades your site on the same six leaks, in 30 seconds, in your browser. Result tells you which leaks you have and what they’re likely costing you per month.
If you’d rather have me run a $499 deep audit and walk you through the fixes on a Loom — [/services/audit/](https://muntin.digital/services/audit/). The audit credits toward a build inside 60 days.
Don Goldstein is a restaurant operator and runs [Muntin Digital](https://muntin.digital/). He is a member of [RAMW](https://ramw.org) and ServSafe certified. The 50-site sample (anonymized) and per-leak scoring rubric are available on request to don@muntin.digital.
Companion pieces
- [An honest DoorDash math, 2026](https://muntin.digital/library/third-party-delivery-economics/) — the per-order math behind leak #6
- [DMV restaurant GBP audit, 2026](https://muntin.digital/learn/research/dmv-restaurant-gbp-audit-2026/) — the discovery side of the same problem
- [Why your restaurant loses reservations every night](https://muntin.digital/library/reservation-conversion-guide/) — the long-form companion
- [Five website changes that recover 1% margin](https://muntin.digital/library/reservation-conversion-guide/) — the fix list, ranked
- [Run the free 30-second audit](https://muntin.digital/tools/audits/restaurant/) on your own site
## Glossary
---
title: Above the fold
url: https://muntin.digital/glossary/above-the-fold/
kind: glossary
locale: en
date: 2026-05-10T04:12:16+00:00
description: Above the fold: The part of your homepage a visitor sees without scrolling. The term is borrowed from newspapers — the half of the front page that was visible before you u…
---
Skip to main content
[Conversions](https://muntin.digital/glossary/conversions/)
# Above the fold
the first screen, "what loads first"
Restaurants
[Conversions & Reservations](https://muntin.digital/learn/topics/conversions/)
The part of your homepage a visitor sees without scrolling. The term is borrowed from newspapers — the half of the front page that was visible before you unfolded it. On mobile that's usually the first ~600 pixels of the page.
## Why it matters
A sizable fraction of your visitors never scroll. Whatever you need them to know or do — what you are, where you are, how to book — has to live above the fold, or it might as well not exist.
More in Conversions
- [**Call to action** — CTA](https://muntin.digital/glossary/cta/)
- [**Reservation link** — OpenTable, Resy, Tock, built-in form](https://muntin.digital/glossary/reservation-link/)
- [**Online ordering link** — Toast, Square, ChowNow, direct checkout](https://muntin.digital/glossary/online-ordering/)
- [**Click-to-directions**](https://muntin.digital/glossary/click-to-directions/)
- [**Sticky mobile footer** — persistent bottom bar on phones](https://muntin.digital/glossary/sticky-footer/)
- [**Social proof** — reviews, press, star ratings](https://muntin.digital/glossary/social-proof/)
Glossary
## Browse all149 terms.
Plain-English definitions for every term in your audit, organized by category.
[Open the full glossary](https://muntin.digital/glossary/)
[Back to the Library](https://muntin.digital/learn/)
---
title: Accessible pair
url: https://muntin.digital/glossary/accessible-pair/
kind: glossary
locale: en
date: 2026-05-10T04:12:16+00:00
description: Accessible pair: A version of a brand color whose lightness has been shifted just far enough — preserving hue and feel — that text rendered in it on a named background reac…
---
Skip to main content
[Brand & design](https://muntin.digital/glossary/brand-design/)
# Accessible pair
a brand color, nudged until text on it clears WCAG AA
All
[Brand & Design](https://muntin.digital/learn/topics/brand-design/)
A version of a brand color whose lightness has been shifted just far enough — preserving hue and feel — that text rendered in it on a named background reaches the WCAG AA contrast threshold (4.5:1 for normal text). Two pairs are usually generated for each color: one on a light surface, one on a dark surface.
## Why it matters
Brand teal looks great in the logo and unreadable as 14px body text on cream. Designing a usable system means knowing both the brand color *and* its accessible neighbor — and using the right one in each context. [Brand Suite](https://muntin.digital/tools/brand-suite/) derives them automatically and exports both variants as CSS tokens.
More in Brand & design
- [**Brand identity** — the full visible system around your logo](https://muntin.digital/glossary/brand-identity/)
- [**Logo lockup** — mark + wordmark + optional tagline, as one unit](https://muntin.digital/glossary/logo-lockup/)
- [**Clearspace** — the empty zone around a logo](https://muntin.digital/glossary/clearspace/)
- [**Color palette** — the curated set of colors that belong to your brand](https://muntin.digital/glossary/color-palette/)
- [**WCAG AA contrast** — accessible color contrast](https://muntin.digital/glossary/wcag-contrast/)
- [**Favicon** — the small icon in the browser tab](https://muntin.digital/glossary/favicon/)
Glossary
## Browse all149 terms.
Plain-English definitions for every term in your audit, organized by category.
[Open Brand Suite](https://muntin.digital/tools/brand-suite/)
[Open the full glossary](https://muntin.digital/glossary/)
---
title: Aggregator
url: https://muntin.digital/glossary/aggregator/
kind: glossary
locale: en
date: 2026-05-10T04:12:16+00:00
description: Aggregator: A third-party marketplace that takes orders on your behalf (DoorDash, Uber Eats, Grubhub) and keeps a commission per transaction. Not the same as direct on…
---
Skip to main content
[Restaurant numbers](https://muntin.digital/glossary/restaurant-numbers/)
# Aggregator
third-party delivery platform
Restaurants
[Operations & Margin](https://muntin.digital/learn/topics/operations-margin/)
A third-party marketplace that takes orders on your behalf (DoorDash, Uber Eats, Grubhub) and keeps a commission per transaction. Not the same as direct online ordering through your own site.
## Why it matters
The single biggest margin pressure most independent restaurants face since 2020. Whether an aggregator is profitable for you depends on cover incrementality (does it bring new customers?) and commission tier. The [DoorDash math post](https://muntin.digital/library/third-party-delivery-economics/) walks through the numbers; the [Delivery Break-Even calculator](https://muntin.digital/tools/margin-math/) runs them on yours.
More in Restaurant numbers
- [**Prime cost** — food cost + labor cost, as a % of sales](https://muntin.digital/glossary/prime-cost/)
- [**Food cost** — cost of goods sold, as a % of sales](https://muntin.digital/glossary/food-cost/)
- [**Labor cost** — wages + payroll taxes, as a % of sales](https://muntin.digital/glossary/labor-cost/)
- [**Cover** — one customer served](https://muntin.digital/glossary/cover/)
- [**Average check** — total sales ÷ covers](https://muntin.digital/glossary/average-check/)
- [**Commission** — the % a platform takes per order](https://muntin.digital/glossary/commission/)
Glossary
## Browse all149 terms.
Plain-English definitions for every term in your audit, organized by category.
[Open the full glossary](https://muntin.digital/glossary/)
[Back to the Library](https://muntin.digital/learn/)
---
title: AI Overview
url: https://muntin.digital/glossary/ai-overview/
kind: glossary
locale: en
date: 2026-05-11T10:58:10-04:00
description: AI Overview: A paragraph Google's generative answer engine writes at the top of a search results page, citing two or three sources by name. Triggered on roughly 13.14%…
---
Skip to main content
[Findability](https://muntin.digital/glossary/findability/)
# AI Overview
the paragraph Google writes above the search results
Restaurants
[AI Search](https://muntin.digital/learn/topics/ai-search/)
A paragraph Google's generative answer engine writes at the top of a search results page, citing two or three sources by name. Triggered on roughly 13.14% of US desktop searches as of March 2025 and approaching one in five in 2026. The cited sources appear as small chips at the bottom of the box.
## Why it matters
For informational restaurant queries — hours, menu, reservations, dietary — the AI Overview now answers above the map pack. If your page isn't cited, your discovery surface for that query is gone. Citation runs on different signals than blue-link SEO: paragraph shape, schema-anchored entities, predicate sentences with no hedge tokens, and stable URLs.
More in Findability
- [**Map pack** — the three-pin Google Maps box above search results](https://muntin.digital/glossary/map-pack/)
- [**Schema** — structured data that tells Google what a page is about](https://muntin.digital/glossary/schema/)
- [**Google Business Profile** — your free Google listing](https://muntin.digital/glossary/gbp/)
- [**NAP consistency** — matching name/address/phone across the web](https://muntin.digital/glossary/nap-consistency/)
- [**Aggregator** — third-party delivery platform](https://muntin.digital/glossary/aggregator/)
Glossary
## Browse all149 terms.
Plain-English definitions for every term in your audit, organized by category.
[Open the full glossary](https://muntin.digital/glossary/)
[Back to the Library](https://muntin.digital/learn/)
---
title: Alt text
url: https://muntin.digital/glossary/alt-text/
kind: glossary
locale: en
date: 2026-05-10T04:12:16+00:00
description: Alt text: A short written description of every image, set via the alt="…" attribute in the HTML. Screen readers read it aloud; Google uses it to understand what's in…
---
Skip to main content
[Findability](https://muntin.digital/glossary/findability/)
# Alt text
Example: 'Plate of carnitas with cilantro and lime' beats 'IMG_4521.jpg' for SEO + accessibility + AI search citations — costs 5 seconds per photo to write.
image descriptions
Restaurants
[Local SEO & Discovery](https://muntin.digital/learn/topics/local-seo/)
A short written description of every image, set via the `alt="…"` attribute in the HTML. Screen readers read it aloud; Google uses it to understand what's in the photo.
## Why it matters
Good alt text opens your site to blind and low-vision visitors (legally required in many jurisdictions), helps Google Images send traffic your way, and gives your site a shot at appearing in voice search results. "IMG_2041.jpg" helps nobody.
More in Findability
- [**Google Business Profile** — GBP, formerly Google My Business](https://muntin.digital/glossary/gbp/)
- [**Title tag** — the page's