Conversions & Reservations

Most restaurant websites get the discovery part right — Google sends them traffic. The leak is later: visitors land, look around, don’t find what they’re looking for, and bounce. The reservation never happens. Conversions is the unglamorous middle of the funnel where almost all the lost money lives, and the fixes are usually small, unsexy, and under-an-hour.

The honest reservation funnel for a typical mid-volume independent looks like this: 1,000 monthly Google clicks → 600 land on a page that loads in under 3 seconds → 350 find the menu without pinch-zooming a PDF → 180 reach the reservation page → 90 actually book. That’s a 9% end-to-end conversion. Move it to 12% and you’ve added roughly $1,200–$3,000 in monthly margin without spending a dollar on ads. Move it from 9% to the upper-quartile 16% — which the “direct order” CTA hierarchy alone usually gets you — and you’re looking at $4,000+ in recovered margin per month.

This pillar collects the playbooks, glossary terms, and tools that close that gap. Read it linearly if you’re auditing your conversion funnel for the first time; jump to a section if you already know which step is leaking.

The six places conversions leak, in order of dollar impact

The 1% margin audit graded 50 independent restaurant websites for the six most common reservation-and-conversion leaks. Median operator leaks 0.94% of revenue; worst-quartile 2.3%. The six, ranked by per-incident impact:

1. Aggregator-only ordering, no direct path — 0.45% revenue impact when present. Every order goes through a 30% commission instead of a 3.6% processor fee. Also the easiest leak to spot in a five-second test of your menu page.
2. Reservation button doesn’t deeplink — 0.34% impact, present on 59% of sites audited. The OpenTable / Resy / Tock button sends the visitor to the platform’s home page instead of your venue’s page. One extra click is one reservation lost. Five-minute fix.
3. EN-only on a bilingual block — 0.31% impact, present on 38%. A real /es/ mirror with hreflang doubles addressable demand in roughly 35% of DMV neighborhoods. The $1,500 menu drop-in ships the menu page version in seven days; the full-site version is in any of the three site tiers.
4. No direct-order CTA above third-party links — 0.27% impact, present on 71%. The single fix that recovers the most margin: a primary-color “Order direct” button positioned above the DoorDash / Uber Eats links. Match the third-party menu price; the customer notices, you keep the 26.4 points.
5. Hours buried below the fold on mobile — 0.21% impact, present on 52%. Phones lose 800ms to a heavy hero photo; the visitor leaves before discovering you’re open. Move the hours block above the hero. 15 minutes.
6. Mobile LCP > 3.0s — 0.18% impact, present on 68%. The slowest leak to fix (1–3 hours of image compression and JS pruning) but the one that compounds with every other leak above it.

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. The last three are where a paid build earns its keep.

The DoorDash math is the conversion math

An aggregator order isn’t just a pricier channel; it’s a parallel funnel that competes with your direct funnel for the same visitor. The marginal lift on switching one order from DoorDash to direct, on a $42 ticket: $4.42 in margin. Across an operator doing 2,400 third-party orders per year, a 40% switch to direct is $4,243 added to the bottom line. The full walk is in An honest DoorDash math, 2026; the calculator that runs it against your numbers is at /tools/margin-math/.

Where the conversion side meets the margin side: every direct-order CTA above the third-party links isn’t just a UX move — it’s a margin lever. The conversion lift compounds with the margin lift; you’re moving more orders and keeping more of each one.

What converts isn’t what looks good

Three findings from the 1% audit that cut against operator instinct:

• Photography quality didn’t correlate with leakage. Phenomenal photos and operator-iPhone snapshots leaked at roughly the same rates. Photography drives consideration, not conversion. The $4,000 you’d spend on a photographer would — on a conversion-leakage basis — be better spent on direct-order plumbing first, then photos. The photo spec sheet is for after the funnel is fixed, not before.
• Older sites didn’t leak more than newer ones. Some of the lowest-leakage sites in the sample were built in 2018 and never touched again. The features that cause leaks — heavy JS, third-party reservation widgets that don’t deeplink, hero photos 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% leakage band. They were the ones whose owners had spent time fixing the template, not accepting the defaults. Wix vs. custom walks through the six places Wix breaks first.

The takeaway: don’t spend on a redesign before you’ve audited the funnel. The cost-per-percent-recovered on a leak fix is roughly $200; the cost-per-percent-recovered on a full rebuild is closer to $4,000.

Menu, hours, and reservation: the three pages that convert

Of the seven pages an independent restaurant website needs (listed here), three of them carry 80% of the conversion weight:

• Menu page. If it’s a PDF, you’re leaking. A real HTML menu loads instantly, indexes for “cacio e pepe near me” queries, and renders well on a 360px phone screen. The free Menu Converter turns pasted menu text into semantic HTML + JSON-LD Menu schema in two seconds. The studio-built version is the $1,500 menu drop-in.
• Hours block. The visitor wants to know if you’re open right now. Anything more than that is friction. Above the fold, plain text, “Open until 10 PM today” format wins over a static schedule table. Add holiday hours via the Holiday Hours Generator; the same tool emits the JSON-LD override and the GBP block.
• Reservation page. Or — better — the embedded reservation widget on the menu page. Either way: deeplink to your venue, not the platform’s home page. Use the platform’s native script (OpenTable Direct, Resy Widgets v2) rather than an iframe; the native script keeps the visitor on your domain and lets you measure the conversion in your own analytics.

What to do this week

Three actions, in order, total time about an hour:

1. Run the free Restaurant Audit on your own URL. The result tells you which of the six leaks are active. Save it to your Workshop and re-run it monthly.
2. Run Storefront Health as the composite read. The audit grades the technical leaks; Storefront Health grades the content + flow leaks. Together they cover the six.
3. Pick one leak. Fix it this week. Reservation deeplink (5 min), hours-above-the-fold (15 min), or direct-order CTA (30 min) are the three under-an-hour fixes. Pick whichever your audit ranked highest. Re-run the audit after; the score moves immediately.

Operators who can’t carve out the hour can outsource it: the $499 deep-read audit ships a Loom + 8-page action sheet in five business days, ranked by dollar impact, with credits toward a build inside 60 days. The $1,500 menu drop-in handles the menu-page leak directly; the Care Plan Light handles the hours-and-hygiene side monthly.

Where this topic touches the others

• Local SEO & Discovery — because the conversion funnel starts with the GBP click. The two pillars share more readers than any other pair on the site.
• Speed & Mobile — because the slow-load leak is upstream of every conversion fix. A 4-second LCP makes every other improvement marginal.
• Operations & Margin — because conversion-channel mix (direct vs. aggregator) is the largest single lever on independent-restaurant margin. The DoorDash math op-ed sits across both pillars.

The composite score you want to track: Storefront Health. Run it monthly, save the trend line, watch the conversions number compound. Everything in this pillar feeds that one metric.