Menu Analytics — See Every Scan, Every Tap, Every Drop-Off

The most underrated thing about a QR menu — relative to a printed one — is not the cost saving or the multi-language support. It is the data. Every scan is a page load. Every category scroll is a measurable event. Every "Add to cart" tap is a high-intent signal. For the first time, restaurants have a real dataset about what diners look at, what they ignore, what they consider, and what they buy. The dashboard makes this data legible.

What QRSeva tracks

Scan-level metrics

• Total scans per day, hour, and day-of-week
• Scans by QR placement (table tents, counter, shopfront, social, business card)
• New vs returning diner share (based on device fingerprint, not personal data)
• Average time-on-menu per scan
• Bounce rate (scanned but closed within 5 seconds)

Item-level metrics

• Views per item (how many diners scrolled the item into the visible area)
• Dwell time per item (average seconds spent looking at the item)
• Tap-through rate (item taps that opened the full detail view)
• Order rate (item adds to cart, when ordering is enabled)
• Conversion rate (view → order)

Category-level metrics

• Category-level scroll depth (how far down the menu most diners go)
• Category jump-nav clicks (which categories diners actively navigate to)
• Time spent per category

What the data typically reveals

Within the first two weeks of running analytics, most restaurants discover three patterns they had no way to see before.

High-view, low-conversion items

These are items diners look at but don't order. Usually one of three things is wrong: the photo is bad (or missing), the description doesn't sell the dish, or the price feels off relative to the perceived value. Identifying these items is the single highest-leverage thing analytics does — fixing them, in priority order, lifts overall order rate fastest.

Quiet winners

The mirror image: items with low view counts but very high tap-through and order rates. These are dishes that convert when seen but get buried because of where they sit on the menu. Promoting them to the top of their category — or to a "bestsellers" group at the top of the menu — typically lifts their order count by 2-3x without touching the dish itself.

Daypart patterns

Most restaurants have a vague intuition about "lunch rush vs dinner rush" but no real numbers. Analytics shows exact scan-by-hour curves, often revealing surprising peaks (a third sub-rush at 4pm coffee-snack time, for example) that nobody had built a menu around. The fix is often a "high-tea menu" — same items, different ordering, surfaced only between 4 and 6pm.

Real example: a Pune café's 30-day transformation

A Pune-based café enabled QRSeva analytics and watched for a month. They discovered three things: their cheesecake (their pride dish) was being viewed by 84% of diners but ordered by only 8% — the photo was a stock image and the description was three words. They replaced both, kept the price, and order rate jumped to 19% in the next week. They also discovered a Korean noodle bowl, buried at position 23 in the "Asian" category, was converting at 31% — the highest of any item — but only 11% of diners ever scrolled to see it. They moved it to position 3. Orders quadrupled. Both changes took less than 15 minutes.

How the analytics work technically

Every page event (scroll, item view, tap) emits a lightweight event to QRSeva's servers. Events are anonymised and aggregated — we don't store the diner's name, phone, IP, or any other personal data. The aggregation runs in near-real-time, so the dashboard updates within seconds of a new scan.

Data retention is 12 months by default. You can export the raw aggregated data at any time as CSV or query it via API (paid feature). For restaurants that want longer retention or finer-grained analytics, an upgrade to the analytics-pro tier exists, but the free tier covers everything most restaurants need.

GDPR, privacy, and what we do not collect

QRSeva analytics is built to be privacy-respecting by design. We do not collect: diner names, phone numbers, email addresses, IP addresses, or any cookies that persist across restaurants. The only identifier is a device fingerprint that resets every 90 days — enough to distinguish "new vs returning" within a window, not enough to identify anyone.

This means the analytics work without a cookie consent banner, work in jurisdictions with strict privacy laws (EU, India's DPDP Act), and don't put restaurants in any compliance bind. Restaurants in EU markets running QR menus with traditional Google Analytics enabled have to handle GDPR-cookie-consent flows; QRSeva's built-in analytics avoid this entirely.

Comparing analytics to printed-menu 'data'

The asymmetry is stark. Printed menus give you nothing. Digital menus give you everything. For the first time, restaurant operators can run their menu like an e-commerce store runs its product catalogue — with measurement, iteration, and per-item optimisation.

How often to check the dashboard

Weekly. The dashboard updates in real-time but most decisions are weekly. On Monday morning, look at: top 3 underperforming items (high view, low conversion), top 3 hidden winners (low view, high conversion), the daypart pattern, and any new dishes you launched the previous week. Spend 15 minutes acting on what you see — re-shoot one photo, move one item up the menu, adjust one price. Compound this over a year and your menu becomes genuinely high-performing.

Daily checking is usually overkill. The signal-to-noise ratio at day-by-day is low — a quiet Tuesday is just a quiet Tuesday, not a signal to act. Weekly cadence aligns with the realistic edit cycle of a restaurant menu.