A 60-table brasserie in Istanbul's Beşiktaş district was stuck at a 2.8% online reservation conversion rate. A 12-week A/B test program lifted it to 5.4% — a 1.9x improvement. Here is exactly what we tested, what we measured, and what surprised us.
CTA Copy and Form Field Count
The first test pitted "Reserve a Table" against "Make a Reservation." Two weeks, 50/50 traffic split, 4,200 sessions. "Reserve a Table" earned 23% more clicks — the verb felt more concrete and action-oriented. The second test compared a 5-field form (name, phone, email, date-time, party size, notes) against a 3-field one (name, phone, date-time). Completion was 71% on the 3-field form versus 48% on the 5-field.
Dropping email felt risky because confirmations needed it. We made it optional and made SMS confirmation the default. The phone was already captured, so the extra field added nothing.
OpenTable Embed vs Custom Modal
OpenTable's embed widget is easy to integrate, but the mobile experience lagged — we measured 3.2 seconds TTI on Turkish mobile traffic (which is 78% of our audience). A custom modal cut TTI to 0.8 seconds and lifted conversion 34%. Hotjar heatmaps showed users not scrolling inside the OpenTable iframe — they stalled at the top.
The custom modal dims the background, closes on ESC, and includes a focus trap. For accessibility we added ARIA-modal and role-dialog. We tested with VoiceOver and NVDA screen readers.
Microsoft Clarity for Friction Detection
Clarity's rage-click and dead-click reports revealed users tapping disabled past dates in the calendar — we had forgotten to grey them out. After the fix, modal completion rose 12%. Session recordings also showed mobile users failing to open the party-size dropdown; we increased the tap target from 32px to 44px.
Heatmaps confirmed the "Reserve" button belongs above the fold. On mobile we moved it directly below the hero image; on desktop we placed it inside the hero. After those two changes conversion reached 4.1%.
FAQ
What is a good free tool to start with? Microsoft Clarity is fully free with no session-recording cap. Hotjar offers 35 sessions/day free.
Minimum traffic per test? Practically, target 1,000+ sessions per week and 200+ conversions per variant for 95% confidence.
Did you drop OpenTable entirely? No — it remains a secondary CTA on desktop; the primary mobile flow now uses our custom modal.
Found this helpful? Share it.
Related articles
7 Smart Ways to Place QR Codes in Your Restaurant
Placement matters more than you think. These seven strategies maximize QR code s…
How to Reduce Waiter Workload by 40% Without Firing Anyone
Smart digital tools don't replace your team — they free them to focus on what ma…
12 Concrete Benefits of QR Menus (Backed by Real Data)
From eliminating print costs to boosting average order value by up to 31%, here …