3.1× Review Lift

Why NFC cards (and not QR codes alone)

QR codes work but the friction is higher — the guest must open their camera app, frame the code, tap the link, then write the review. NFC reduces this to a single tap, no camera, no app. In a fast-casual setting where the guest is mid-meal or mid-pickup, the friction reduction is the lever.

• QR code conversion to a posted review: roughly 0.7% of tabletop impressions in the control period.
• NFC tap conversion to a posted review: 2.1% of tap events in the deployment period.
• Per-card tap rate at tabletop: 5–7 taps / card / day across 412 stores.
• Per-card tap rate at takeout counter: 12–18 taps / card / day (higher because the cashier verbally invites the guest).

Why NTAG215 specifically

NTAG215 is the workhorse of NFC marketing cards — 504 bytes of user memory is enough for a Google Review URL plus a fallback redirect, the chip is universally readable by every iPhone (since iOS 13) and every NFC-enabled Android (since Android 4.4), and the chip-cost economics work at the customer's volume.

• NTAG213 (180 bytes) — considered, sufficient for the URL alone but no room for a fallback redirect.
• NTAG215 (504 bytes) — chosen, room for the URL + the fallback redirect + a tap-counter tag (private chip memory).
• NTAG216 (888 bytes) — overkill, ~$0.04 / unit more expensive with no functional benefit for this use case.
• NTAG424 DNA — considered for the future programme expansion (authenticated taps for loyalty), not used in this case study.

Deployment logistics across 412 stores

Cards needed to be per-store-encoded (each store has a unique Google review URL) and shipped to the store with the artwork matching the chain's brand standards. Proud Tek ran the per-store encoding in Shenzhen using a database of store IDs supplied by the customer's IT team, then shipped one branded card set per store via FedEx ground.

• Per-store encoding at Proud Tek factory using a Mühlbauer encoding line — 8,000 cards / hour throughput.
• Master URL pattern — https://g.page/r/<store-id>/review, with <store-id> drawn from the customer's IT team's CSV.
• Quality gate — every card read post-encoding to verify the URL matches the store-ID database. 100% test, 0 mis-encoded cards in the deployment lot.
• Total production-to-store lead time: 28 days (production 18 days, FedEx 6 days, store-team activation 4 days).

Operational results at 90 days post-deployment

• 3.07×Lift in monthly Google reviews per store
• +0.2 ⭐Chain-wide rating change (4.2 → 4.4)
• 2.1%NFC tap → posted-review conversion rate
• 412Stores deployed

The customer's CX team modelled the Google review programme on a 'review velocity' KPI — reviews per store per month, normalised by transaction volume. The 3.07× lift held steady through month 6 and showed no sign of habituation decay in the year-1 data. The customer expanded the programme to include hand-held cards for shift leaders and is piloting NTAG424 DNA cards as a future authenticated-loyalty card.

How the card program was built — chip, encoding, artwork, rollout

A review card only works if the guest never has to think about it, so the build came down to four deliberate choices: a universally-readable chip, a per-store URL burned in at the factory, brand-matched artwork, and a two-placement drop that put a card exactly where the guest's hands already were. The same specification is documented on the Google Review NFC card solution page.

• Chip: NXP NTAG215 — an NFC Forum Type 2 tag with 504 bytes of user memory, room for the Google Review URL plus a fallback redirect, readable by every iPhone since iOS 13 and every NFC-enabled Android since Android 4.4.
• Encoding: each card was written with its store's own Google Review URL (https://g.page/r/<store-id>/review) on Proud Tek's Mühlbauer line at 8,000 cards / hour, then 100% read-verified against the store-ID database — 0 mis-encoded cards in the deployment lot.
• Form factor: an 85 × 55 mm PVC card, tabletop matte finish with a spot UV brand mark and a QR code fallback printed on the reverse for the few phones without a tap path.
• Placement: one card per booth plus one hand-held card at the takeout counter — the counter card sees 12–18 taps a day because the cashier says the words out loud, against 5–7 taps a day on a passive tabletop card.
• Rollout: production to store shelf in 28 days (18 days production, 6 days FedEx ground, 4 days store-team activation), shipped as one branded card set per store across all 412 stores.

The tap-to-review path — one gesture from booth to posted review

This is the whole product in one gesture. The guest taps the card, the phone opens the store's Google Review URL, and the rating posts — no camera, no app, no typing a search into Maps. Removing those steps is exactly what moved tabletop conversion from 0.7% on a QR code to 2.1% on a tap.

• Tap: the guest holds their phone to the NTAG215 card at the booth or takeout counter — NFC Forum Type 2 tags are read natively, with no app to open first.
• Open: the phone reads the encoded per-store URL (https://g.page/r/<store-id>/review) and opens it directly in the browser.
• Post: the guest lands on the store's own Google review screen and leaves a rating — the single step the card exists to make frictionless.
• Fallback: phones without a tap path (older iPhone 7–X, non-NFC Android) use the QR code on the card's reverse, so no guest reaches a dead end.

What 90 days changed — reviews, rating, and sentiment

Every figure here is the customer's own measured result over the first 90 days against the pre-deployment control — reported, not modelled. The headline is review velocity: monthly reviews per store went from 14 to 43.

• Review volume: 14 reviews / store / month in the control period rose to 43 in the first 90 days — a 3.07× lift that held through month 6 with no habituation decay in the year-1 data.
• Rating: the chain-wide Google rating moved from 4.2 to 4.4 stars (+0.2 ⭐) as review volume climbed.
• Sentiment: 91% of deployment-period reviews were 4 or 5 stars, up from 84% in the control period — the prompt lifted review quality, not just count.
• Reach: the program ran across all 412 stores, and the customer has since expanded it to hand-held cards for shift leaders.

The deployment at a glance — chip, card, and measured results

One table for the reviewer who wants the specification and the outcomes side by side. Every figure is either the shipped specification or the customer's own reported result — nothing here is modelled.

What transfers — where a tap-to-review card fits next

The mechanics that made this work — a universally-readable NFC tag, a per-location review URL, and a card placed where the customer's hands already are — aren't specific to fast-casual restaurants. The same pattern moves onto any high-footfall, review-sensitive business; the Google Review cards for restaurants page documents the workflow, and the Google Review NFC card setup guide covers the per-store URL encoding end to end.

• Multi-location hospitality — cafés, bars, and quick-service groups where each site has its own Google Business Profile and its own review URL to encode.
• Personal-service chains — salons, clinics, gyms, and auto service, where a front-desk hand-off is the natural moment to invite a tap.
• Any counter with a verbal prompt — the takeout-counter pattern, where a staff member invites the tap, was the highest-converting placement and transfers to any service desk.
• Choosing the chip is the one real decision — the NTAG213 vs NTAG215 vs NTAG216 comparison covers when the extra memory of NTAG215 earns its keep over NTAG213.

FAQ