Driver POV (Meta glasses) + restaurant reviews strategy

Who this is for

You need two content engines that feel native on IG Reels / TikTok / FB Reels and still move marketplace KPIs:

1. Driver POV series — “delivery guy with Meta glasses / a day in the life of an UmaMeats driver” (immersive, Toronto local).
2. Restaurant review posts — honest, delivery-aware takes that feature kitchens on UmaMeats (or launching soon).

Primary persona: diners (entertainment → trust → app CTA → first order).
Secondary: soft driver recruiting + restaurant partnership goodwill.

Use with: Diner social launch cookbook for zone sequencing, offers, and kill/scale discipline. Do not scale this content into empty menus.

North star: Profile visits → link clicks → first orders in live zones — not follower vanity.

---

A. Meta glasses / driver day-in-the-life series

Why this works for UmaMeats

Position the series as entertainment with a marketplace payoff, not a corporate “careers” video. End cards sell the app; the middle sells the feeling.

Series title options (pick one and stick):

• UmaMeats POV · Toronto deliveries
• Glasses on · dinner rush
• Drop #XX · [Neighbourhood]

---

Legal / safety / privacy rules (non-negotiable)

Film like a professional courier who respects people and the road.

People & privacy

• Do not film customers’ faces, doorways with visible unit numbers, or kids without explicit consent.
• Blur or crop: faces in crowds, doorbell cams pointing at you, mail with names, phone screens with PII.
• Restaurant staff / owners: ask before featuring them; a quick “OK to film for UmaMeats social?” on camera or WhatsApp is enough for organic UGC.
• No filming inside private residences.

Locations & identifiers

• Do not show full street addresses, apartment numbers, or license plates (blur if they appear).
• Prefer neighbourhood / street character (“King West dinner rush”) over exact drop pins.
• Avoid filming secure building entry codes, locker PINs, or bag seals with order IDs readable.

Road safety — Meta glasses while driving

• Assume glasses-on continuous recording while operating a vehicle is a liability. Ontario distraction / careless driving risk is real; insurance and brand risk are worse.
• Preferred filming modes:
  1. Bike / walk / scooter segments when legal and safe in your zone
  2. Parked — glasses or phone before leaving the curb
  3. Pickup moments — restaurant curb / lobby / counter (with consent)
  4. Phone B-roll held by a second person on sidewalk (not leaning into traffic)
• If the driver is in a car: glasses off while driving. Capture departure/arrival when parked only.
• Never film while handling the bag and crossing mid-block; safety > shot.

Music / rights

• Use platform library / licensed tracks only.
• No copyrighted restaurant TV audio blasting in the background if avoidable.

Disclosure

• If the featured driver is paid / staff, that’s fine for organic brand content; if a post is a paid ad, use platform branded-content / ads tools.
• Driver earnings claims must be honest ranges or clearly labeled examples (“this shift”), not guarantees.

---

Episode formats

Shoot masters you can cut to 15s, 30s, and 60–90s.

Cadence: 2–3 driver-POV posts/week (plus Stories scraps from the same shoot). Batch-film 1 long shift → cut 3–5 posts.

---

Shot list — Meta glasses POV + phone B-roll backup

Glasses POV (primary)

1. Hands grabbing insulated bag / thermal insert
2. Approaching restaurant door / open sign / menu board (no sensitive back-of-house without OK)
3. Counter handoff / bag sealed (blur order ticket names)
4. Bike rack lock / helmet clip / walk start (prove you’re not driving distracted)
5. Street-level Toronto texture: streetcar, mural, skyline slice, neighbourhood sign
6. Building entrance from a respectful distance (no unit #)
7. Bag set at door / “delivered” gesture — cut before any personal interaction if no consent
8. End: phone unlock → UmaMeats diner app home (or logo end-card)

Phone B-roll backup (always shoot — glasses fail, fog, battery, or safety cut)

1. Hero dish stills from the restaurant (with permission) or plated promo shots you already have
2. Side profile of driver with bag (face OK if driver is talent)
3. App screen: tracking map zoomed out (no customer pin detail)
4. Timelapse of bag packing at restaurant
5. Sunset / “dinner o’clock” establishing shot of the zone
6. Text overlays filmed as separate takes for captions

Audio

• Ambient street + short VO: “First drop in [Neighbourhood] — let’s go.”
• Or trending audio bed + burned-in captions (assume sound off).

---

Hook formulas (IG Reels / TikTok / FB Reels)

First 1–3 seconds decide the algorithm. On-screen text + motion.

1. POV: you’re the UmaMeats driver on [Neighbourhood] dinner rush
2. Glasses on. Hot bag. Toronto. Let’s deliver.
3. Stop scrolling if you’re hungry in [zone] tonight
4. Day in the life of a Toronto delivery driver (UmaMeats)
5. Pickup → dropoff in under [X] min (no address spoilers)
6. What $ [example] looks like on a [weekday] peak (example shift)
7. This is what “order placed” looks like on our side
8. Hidden gem pickup in [zone] — wait till you see the bag

Avoid: “Hey guys welcome back to the UmaMeats channel” · slow logo sting · driving footage with glasses UI in frame while moving.

---

Captions + keyword / hashtag banks

Caption skeleton (driver POV) [HOOK line]. Riding along on UmaMeats deliveries in [Neighbourhood]. Hungry? Order with [FIRST_ORDER_CODE] for [OFFER] → [APP_LINK]. Want to drive peaks? [DRIVER_SIGNUP_LINK].

Keywords to weave (1–2 natural phrases): Toronto food delivery, [neighbourhood] delivery, delivery driver Toronto, day in the life delivery, order takeout Toronto, dinner rush Toronto

Hashtags (5–10): #TorontoEats #TorontoFood #FoodDelivery #DeliveryDriver #DayInTheLife #Toronto + 1 zone (#KingWest #TheAnnex etc.) + #UmaMeats

Alt text (IG): “First-person view of an UmaMeats delivery driver picking up food in [Neighbourhood], Toronto.”

---

Persona mapping

Keep one hard CTA per post (usually diner). Put driver link in caption line 2 or Stories sticker, not competing end cards.

---

CTA templates (placeholders)

Diner-primary end card text: Order in [NEIGHBOURHOOD] · [FIRST_ORDER_CODE] · link in bio

Soft dual CTA caption: Hungry in [NEIGHBOURHOOD]? [FIRST_ORDER_CODE] for [OFFER] → [APP_LINK] Drive with us at peaks → [DRIVER_SIGNUP_LINK]

Stories sticker pack: Poll “Lunch or dinner?” · Link sticker · “Code [FIRST_ORDER_CODE]”

---

Production checklist (film this week)

Gear

• Meta glasses charged + storage clear (or skip glasses and shoot phone POV — still ship)
• Phone 9:16 backup + gimbal optional
• Power bank, lens wipe (steam/fog), spare bag for hero shots
• Brand end-card PNG (logo + code space)

Consent & safety

• Driver talent release (if not employee) on file
• Restaurant OK for any staff faces / logos close-up
• No glasses while driving plan written for the shift
• Blur pass scheduled before publish

Edit

• Captions burned in; hook text frame 0–2s
• Music from platform library
• Thumbnail: clearest food or bag handoff + 3–5 words
• UTM on every link: utm_source=instagram|tiktok|facebook&utm_medium=organic&utm_campaign=driver_pov

Publish

• Cross-post same day IG + TikTok; FB Reel within 24h
• Tag restaurant; location sticker = neighbourhood
• Pin best episode; add to Highlight “POV”

---

B. Restaurant review posts strategy

Positioning

UmaMeats is a local taste guide for delivery, not a Yelp clone.

• Only review restaurants live on UmaMeats or launching within ~7–14 days (say “coming to UmaMeats”).
• Voice: neighbour who orders delivery a lot — honest, specific, useful.
• Goal: diner discovery → order on our app; secondary: merchant goodwill (tag them, ask them to share).

Not our job: roast merchants in public, dunk on competitors’ apps, or invent “#1 in Toronto” claims.

---

Formats that convert

---

Review framework (delivery-relevant)

Score or speak to each — diners care about door experience, not only dine-in vibes.

1. Taste — what you’d reorder; name the dish
2. Portion — enough for 1? shareable?
3. Packaging for delivery — leaks, sogginess, venting, utensils
4. Value — price vs fullness vs quality (be fair)
5. Reorder likelihood — “would I tap again Friday?” (the conversion metric)

Optional sixth: ETA honesty / arrival temp — only if you experienced a real UmaMeats order.

On-screen score card (example): Taste 9 · Portion 8 · Packaging 9 · Value 8 · Reorder: yes

---

Caption templates + keywords

Honest take Honest take: [Dish] from [Restaurant] in [Neighbourhood]. Taste [X]/10 · packaging held up for delivery. Order on UmaMeats with [FIRST_ORDER_CODE] for [OFFER].

3 under $X 3 dishes under $[X] delivering in [Neighbourhood] on UmaMeats: 1) … 2) … 3) … Code [FIRST_ORDER_CODE].

Hidden gem Hidden gem in [Neighbourhood]: [Restaurant] — try the [signature]. Live on UmaMeats → [APP_LINK].

Unbox Delivery unbox: [Restaurant] [Dish]. Packaging [passed/almost]. Craving this in [zone]? [FIRST_ORDER_CODE].

Keywords: Toronto food, [neighbourhood] restaurants, food delivery review, best [cuisine] Toronto, order [cuisine] delivery

Hashtags: #TorontoEats #TorontoFood #[Neighbourhood] #FoodReview #FoodDelivery + 1 cuisine + #UmaMeats

---

Ethical rules

• Disclose if the meal was comped, gifted, or part of a partnership (Thanks [Restaurant] for having us / paid partnership tools when required).
• Don’t trash merchants publicly. If an order fails (cold, wrong, damaged): resolve offline; post only constructive notes or skip. Public roast burns supply-side trust.
• No fake scarcity (“only 2 left”) unless true in-app.
• Photos must match what diners can order now.
• Tag the restaurant; invite them to share — that’s free distribution.

---

How reviews feed the funnel

Cross-link reviews in Stories to the same week’s driver-POV pickup at that kitchen when possible (content universe).

---

Cross-posting map

UTM: utm_campaign=restaurant_review vs driver_pov so you can see which engine drives orders.

---

C. Combined 14-day content calendar

Assumes: live menus in 1–2 zones, [FIRST_ORDER_CODE] works, one driver talent available for 2–3 shoots.

Pre-flight (Day 0)

• Confirm Meta glasses or commit to phone POV (don’t block on hardware)
• Book 1 restaurant for consented pickup shoot + 2 review orders
• End-card + UTM links live
• Tracking sheet: date · platform · type (POV/review) · zone · link clicks · first orders

Week 1

Week 2

Steady state (day 15+): 2–3 POV / week + 2 reviews / week + daily Stories from the same assets.

---

D. Measurement

Funnel

Views → profile visits → link clicks → installs → first orders → D7 reorder

Track by campaign UTM (driver_pov vs restaurant_review).

Leading (check every posting day)

• 3s / average watch % on Reels
• Profile visits and link sticker taps
• Comments: “do you deliver to X?” / “what’s the code?”
• Restaurant shares / tags

Lagging

• First orders attributed to social
• CAC if boosting
• Merchant sentiment (DMs after reviews)

Kill rules

• Hook fails: < platform median 3s retention after 2 iterations → retire format
• Reviews that never mention UmaMeats / no CTA → rewrite
• Any privacy / safety complaint → pause series, fix process
• Boosted spend with healthy views but no link clicks for 3 days → fix CTA/landing, don’t raise budget

Scale rules

• One POV episode or review gets 2× median link CTR → boost geo-fenced to that zone only
• Restaurant shares your review → do a collab follow-up within 7 days
• Driver signup spikes from earnings posts → keep soft CTA but don’t let recruit creative replace diner CTAs on all posts

Scoreboard columns

Date | Type (POV/Review) | Platform | Zone | Views | Profile visits | Link clicks | First orders | Notes

---

Brand tone

Practical Toronto marketplace. Neighbourhood names. Real bags, real kitchens, real codes. No purple-glow startup fluff. No filming customers without consent. One CTA.

---

Related

• Diner social launch cookbook — zone launch, keywords, offers
• Personas: Diners · Drivers · Restaurants
• Campaign: Driver POV + restaurant reviews
• Channels: Instagram · TikTok · Facebook
• Getting started