Menu Analytics: What Guests Look At (and What It Tells You to Fix)

A paper menu is a one-way broadcast: you never learn what guests read, skipped, hesitated over or wanted translated. A digital menu quietly answers all of it. The skill is knowing which numbers deserve action — and which are just weather.

The four signals worth watching

• •Menu views by day and hour — your demand curve. A Tuesday-lunch spike you didn't know about is a staffing decision; a dead Thursday evening is a promotion slot.
• •Per-dish views — attention, before money changes hands. The gap between seen often and ordered often (compare against POS) is the most diagnostic number in the whole system: high views with low orders means the description, photo or price is refusing the sale at the last second.
• •Likes — a lightweight desirability signal, useful for relative ranking within a category. Guests like freely; they order under budget constraints. A much-liked, little-ordered dish is usually priced or portioned wrong, not unwanted.
• •Language usage — who your guests actually are. When 30% of menu sessions run in German, that's a marketing insight (target Saxon weekend traffic), a staffing hint, and an instruction to finish the German translations you left at 80%.

Reading QR scans honestly

Scan counts approximate table sessions, not revenue. Their best use is comparative: which table zones scan most (placement working), what share of scans happen outside opening hours (people checking the menu from home — your menu is marketing), and whether a new table-tent design moved the number.

Small-data discipline

A restaurant's menu data is small data: dozens to hundreds of events a day, not millions. That means weekly patterns beat daily noise — read the dashboard once a week, compare like-for-like (this Tuesday vs. recent Tuesdays), and change one thing at a time so the next week's read means something. ProMenu's analytics view (daily views over two weeks, top dishes, language split) is deliberately shaped for exactly this cadence.

From number to action

The loop that pays: notice a high-view low-like dish → rewrite its description or reshoot the photo → wait a week → re-read. Notice German sessions rising → complete the translation → feature a dish Germans keep viewing. Notice at-home scans peaking Friday afternoon → post the weekend special Friday at noon. Each loop is fifteen minutes; compounded, they are menu engineering with live instrumentation.

Takeaway

Weekly, comparative, one change at a time. The menu has been watching your guests attentively — analytics is just the discipline of listening back.